1 Quick xlsx reader functions
My functions to perform the various overrepresentation/enrichment analyses generally assume the result of extract_significant_genes() as the input. Sadly, I do not want to save those data structures to disk because they can be quite monstrously large when serialized (because I keep a reference to the input in them which gets dereferenced on save()). As a result, it is far simpler/safer to just read the xlsx outputs produced by combine_de_tables and extract_significant_genes. Ergo the following two little functions.
table_reader <- function(xlsx, sheet = "outcome") {
input_df <- openxlsx::readWorkbook(xlsx, sheet = sheet, startRow = 2)
rownames(input_df) <- input_df[["row.names"]]
input_df[["row.names"]] <- NULL
print(dim(input_df))
if (nrow(input_df) < 5000) {
warning("Something appears wrong with the de table: ", input_xlsx)
}
return(input_df)
}
sig_reader <- function(xlsx, sheet = "outcome", direction = "up") {
this_sheet <- paste0(direction, "_deseq_", sheet)
input_df <- openxlsx::readWorkbook(xlsx, sheet = this_sheet, startRow = 2)
rownames(input_df) <- input_df[["row.names"]]
input_df[["row.names"]] <- NULL
dim(input_df)
if (nrow(input_df) < 20) {
message("There are less than 20 rows in this significance table, it is unlikely to be interesting.")
}
return(input_df)
}I have been having some difficulties getting the treeplot fonts to match our readability goals. The set of things I have tried so far are:
- Setting a geom_text(size = x): this failed because it didn’t know to which element(s) to apply the new size.
- Changing the options for theme_bw(): this does change some font elements, but not the ones that I wanted. I tried more than a few, most importantly theme(text = element_text(size = x))
- Changing the default for theme_bw via theme_update(text = element_text(size = x)): this also changes some text, but not all.
- Invoking update_geom_defaults(“text”, aes(size = 10)) at the stop of my document. This does not seem to do much of anything.
- Playing with options in enrichplot::treeplot(), notably the cex and fontsize options. These also help with some, but not all fonts. I also spent a fair amount of time reading the code for treeplot() and decided to try #6 before I got too frustrated.
- Just changing the canvas size of my output pdf. This worked great, though it may require setting the wordwrap options in treeplot.
With that in mind, the following two parameters are going to set the width and height of the output pdf documents which will be sent to Maria Adelaida in order to arrange into the final figures via inkscape.
treeplot_height <- 6
treeplot_width <- 12
wrap_width <- 24
desired_go_ont <- "BP"2 Introduction
I once again moved the ontology analyses to their own document. This is primarily because I want a fresh notebook to play around with these and get some of the resulting clutter out of the DE notebooks.
The con of doing this is that some of my enrichment methods are smart enough to directly take the output from combine_de_tables()/extract_significant_genes(). I guess I can dispatch a method to take the xlsx file as input…
Thus, for each enrichment analysis that was in the differential expression documents, there will now be a few stanzas here, one which loads the appropriate xlsx file, one which performs gProfiler, and one which uses clusterProfiler. On occasion there may be another with random stuff like me poking at transcription factors or other side interests…
In addition, I am going to do the Tumaco-only analyses first, because they are what are actually in the paper.
2.1 Gene Set Enrichment
The gene set enrichment will follow each DE analysis during this document. I am adding a series of explicitly GSEA analyses in this most recent iteration, in these I will pass the full DE table and check the distribution of logFC values against the genes in each category as opposed to the simpler over-enrichment of the high/low DE genes.
Most (all?) of the gene set enrichment analyses performed in this document are a combination of gProfiler2 (Kolberg et al. (2020)) and clusterProfiler (Yu et al. (2012)). There are functions available in hpgltools to also perform a few other methods, but for the moment I am sticking to only these two. Oh yeah, I need to wrap clusterProfiler’s tree (taken from topGO) plotter because it spams plots everywhere…
input_xlsx <- glue("analyses/3_cali_and_tumaco/DE_Cure_Fail/Clinical_Samples/tc_clinical_cf_table_sva-v{ver}.xlsx")
all_de_cf_table <- table_reader(input_xlsx, "outcome")## [1] 12162 68all_de_cf_up <- sig_reader(glue("analyses/3_cali_and_tumaco/DE_Cure_Fail/Clinical_Samples/tc_clinical_cf_sig_sva-v{ver}.xlsx"), "outcome", "up")
all_de_cf_down <- sig_reader(glue("analyses/3_cali_and_tumaco/DE_Cure_Fail/Clinical_Samples/tc_clinical_cf_sig_sva-v{ver}.xlsx"), "outcome", "down")
all_de_cf <- rbind.data.frame(all_de_cf_up, all_de_cf_down)
tumaco_xlsx <- glue("analyses/4_tumaco/DE_Cure_Fail/t_all_visitcf_sig_sva-v{ver}.xlsx")
t_de_cf <- openxlsx::readWorkbook(tumaco_xlsx, sheet = 3, startRow = 2)
rownames(t_de_cf) <- t_de_cf[["row.names"]]
t_de_cf[["row.names"]] <- NULL
t_de_cf_up <- t_de_cf[, c("deseq_logfc", "deseq_adjp", "deseq_basemean", "deseq_lfcse")]
t_de_cf <- openxlsx::readWorkbook(tumaco_xlsx, sheet = 4, startRow = 2)
rownames(t_de_cf) <- t_de_cf[["row.names"]]
t_de_cf[["row.names"]] <- NULL
t_de_cf_down <- t_de_cf[, c("deseq_logfc", "deseq_adjp", "deseq_basemean", "deseq_lfcse")]
t_de_cf <- rbind.data.frame(t_de_cf_up, t_de_cf_down)2.2 gProfiler search of all Tumaco samples
The following gProfiler searches use the all_gprofiler() function instead of simple_gprofiler(). As a result, the results are separated by {contrast}_{direction}. Thus ‘outcome_down’.
The same plots are available as the previous gProfiler searches, but in many of the following runs, I used the dotplot() function to get a slightly different view of the results.
2.4 All Tumaco clinical samples cure/fail
The xlsx files are:
- “{cf_prefix}/All_Samples/t_clinical_cf_table_sva-v{ver}.xlsx”
- “{cf_prefix}/All_Samples/t_clinical_cf_sig_sva-v{ver}.xlsx”
input_xlsx <- glue("{cf_prefix}/All_Samples/t_clinical_cf_table_sva-v{ver}.xlsx")
t_clinical_cf_table_sva <- table_reader(input_xlsx, "outcome")## [1] 14156 68input_xlsx <- glue("{cf_prefix}/All_Samples/t_clinical_cf_sig_sva-v{ver}.xlsx")
t_clinical_cf_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_clinical_cf_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")And without biopsies
input_xlsx <- glue("{cf_prefix}/All_Samples/t_clinical_nobiop_cf_table_sva-v{ver}.xlsx")
t_clinicalnb_cf_table_sva <- table_reader(input_xlsx, "outcome")## [1] 11910 68input_xlsx <- glue("{cf_prefix}/All_Samples/t_clinical_nobiop_cf_sig_sva-v{ver}.xlsx")
t_clinicalnb_cf_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_clinicalnb_cf_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_clinicalnb_cf_sig_sva_both <- rbind.data.frame(t_clinicalnb_cf_sig_sva_up,
t_clinicalnb_cf_sig_sva_down)2.4.1 gProfiler
After most of the DE analyses, the set of significantly DE genes gets passed to gProfiler2 and clusterProfiler. One slightly neat thing in my gprofiler (and goseq/topGO/gostats) function(s): it coerces the result into the same datastructure produced by clusterProfiler, thus one may play with the various plotting functions in the enrichplot (Yu (n.d.)) package. This is kind of fun because gProfiler2 provides easy access to a few datasets:
- GO: (Ashburner et al. (2000))
- KEGG: (Kanehisa and Goto (2000))
- Reactome: (Croft et al. (2011))
- WikiPathways: (Kutmon et al. (2016))
- Transfac: (Wingender et al. (1996))
- miRTarBase: (Hsu et al. (2011))
- The Human Protein Atlas: (Pontén et al. (2011))
- Corum: (Giurgiu et al. (2019))
- Human phenotype ontology: (Köhler et al. (2017))
2.4.1.1 Genes higher in cure
My general sense is that the comparisons of primary interest are reactome and one or more GO. I suspect that if there are lots of transcription factors, that might prove interesting.
t_cf_clinical_gp_up <- simple_gprofiler(
t_clinical_cf_sig_sva_up,
excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinical_cure_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_clinical_gp_up## Error in eval(expr, envir, enclos): object 't_cf_clinical_gp_up' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_clinical_gp_up[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinical_gp_up' not foundt_cf_clinical_gp_go_up_tree <- sm(enrichplot::treeplot(go_termsim,
label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = "images/overrepresentation/t_cf_clinical_gp_up_tree.pdf",
width = treeplot_width, height = treeplot_height)## Warning in pp(file = "images/overrepresentation/t_cf_clinical_gp_up_tree.pdf",
## : The directory: images/overrepresentation does not exist, will attempt to
## create it.t_cf_clinical_gp_go_up_tree## Error in eval(expr, envir, enclos): object 't_cf_clinical_gp_go_up_tree' not founddev.off()## png
## 2t_cf_clinical_gp_go_up_tree## Error in eval(expr, envir, enclos): object 't_cf_clinical_gp_go_up_tree' not foundenrichplot::dotplot(t_cf_clinical_gp_up[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinical_gp_up' not foundreactome_termsim <- enrichplot::pairwise_termsim(t_cf_clinical_gp_up[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinical_gp_up' not foundt_cf_clinical_gp_reac_up_tree <- sm(enrichplot::treeplot(reactome_termsim,
label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'reactome_termsim' not foundpp(file = "images/overrepresentation/t_cf_clinical_gp_up_tree.pdf",
width = treeplot_width, height = treeplot_height)
t_cf_clinical_gp_reac_up_tree## Error in eval(expr, envir, enclos): object 't_cf_clinical_gp_reac_up_tree' not founddev.off()## png
## 2t_cf_clinical_gp_reac_up_tree## Error in eval(expr, envir, enclos): object 't_cf_clinical_gp_reac_up_tree' not foundpp(file = "images/overrepresentation/t_cf_clinical_gp_up_dot.pdf",
width = treeplot_width, height = treeplot_height)
enrichplot::dotplot(t_cf_clinical_gp_up[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinical_gp_up' not founddev.off()## png
## 2enrichplot::dotplot(t_cf_clinical_gp_up[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinical_gp_up' not foundtf_termsim <- enrichplot::pairwise_termsim(t_cf_clinical_gp_up[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinical_gp_up' not found## The treeplot fails for this for some reason?
##t_cf_clinical_gp_tf_up_tree <- sm(enrichplot::treeplot(tf_termsim))
enrichplot::dotplot(t_cf_clinical_gp_up[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinical_gp_up' not foundenrichplot::dotplot(t_cf_clinical_gp_up[["WP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinical_gp_up' not foundNo biopsies!
t_cf_clinicalnb_gp_up <- simple_gprofiler(
t_clinicalnb_cf_sig_sva_up,
excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinicalnb_cure_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_clinicalnb_gp_up## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_up' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_up[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinicalnb_gp_up' not foundt_cf_clinicalnb_gp_go_up_tree <- sm(enrichplot::treeplot(go_termsim,
label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = "figures/t_cf_clinicalnb_gp_up_tree.svg",
width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_go_up_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_go_up_tree' not founddev.off()## png
## 2t_cf_clinicalnb_gp_go_up_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_go_up_tree' not foundenrichplot::dotplot(t_cf_clinicalnb_gp_up[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinicalnb_gp_up' not foundreactome_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_up[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinicalnb_gp_up' not foundt_cf_clinicalnb_gp_reac_up_tree <- sm(enrichplot::treeplot(reactome_termsim,
label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'reactome_termsim' not foundpp(file = "images/overrepresentation/t_cf_clinicalnb_gp_up_tree.pdf",
width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_reac_up_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_reac_up_tree' not founddev.off()## png
## 2t_cf_clinicalnb_gp_reac_up_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_reac_up_tree' not foundpp(file = "images/overrepresentation/t_cf_clinicalnb_gp_up_dot.pdf",
width = treeplot_width, height = treeplot_height)
enrichplot::dotplot(t_cf_clinicalnb_gp_up[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinicalnb_gp_up' not founddev.off()## png
## 2enrichplot::dotplot(t_cf_clinicalnb_gp_up[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinicalnb_gp_up' not foundtf_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_up[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinicalnb_gp_up' not found## The treeplot fails for this for some reason?
##t_cf_clinicalnb_gp_tf_up_tree <- sm(enrichplot::treeplot(tf_termsim))
enrichplot::dotplot(t_cf_clinicalnb_gp_up[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinicalnb_gp_up' not foundenrichplot::dotplot(t_cf_clinicalnb_gp_up[["WP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinicalnb_gp_up' not found2.4.1.2 Genes higher in fail
The only significant results for this group appear to be GO.
t_cf_clinical_gp_down <- simple_gprofiler(
t_clinical_cf_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinical_fail_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_clinical_gp_down## Error in eval(expr, envir, enclos): object 't_cf_clinical_gp_down' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_clinical_gp_down[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinical_gp_down' not foundt_cf_clinical_gp_go_down_tree <- sm(enrichplot::treeplot(go_termsim))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = "images/overrepresentation/t_cf_clinical_gp_down_tree.pdf",
width = treeplot_width, height = treeplot_height)
t_cf_clinical_gp_go_down_tree## Error in eval(expr, envir, enclos): object 't_cf_clinical_gp_go_down_tree' not founddev.off()## png
## 2No biopsies!
t_cf_clinicalnb_gp_down <- simple_gprofiler(
t_clinicalnb_cf_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinicalnb_fail_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_clinicalnb_gp_down## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_down' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_down[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinicalnb_gp_down' not foundt_cf_clinicalnb_gp_go_down_tree <- sm(enrichplot::treeplot(go_termsim))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = "images/overrepresentation/t_cf_clinicalnb_gp_down_tree.pdf",
width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_go_down_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_go_down_tree' not founddev.off()## png
## 2t_cf_clinicalnb_gp_go_down_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_go_down_tree' not found## And both, this is not usually where I put this, but the
## clinical samples are a bit of a special case.
t_cf_clinicalnb_gp_both <- simple_gprofiler(
t_clinicalnb_cf_sig_sva_both,
excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinicalnb_fail_up_gp-v{ver}.xlsx"))## Error in function (type, msg, asError = TRUE) : Recv failure: Connection reset by peert_cf_clinicalnb_gp_both## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_both' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_both[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinicalnb_gp_both' not foundt_cf_clinicalnb_gp_go_both_tree <- sm(enrichplot::treeplot(go_termsim))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = "images/overrepresentation/t_cf_clinicalnb_gp_both_tree.pdf",
width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_go_both_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_go_both_tree' not founddev.off()## png
## 2t_cf_clinicalnb_gp_go_both_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_go_both_tree' not foundreac_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_both[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_clinicalnb_gp_both' not foundt_cf_clinicalnb_gp_reac_both_tree <- sm(enrichplot::treeplot(reac_termsim))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'reac_termsim' not foundpp(file = "images/overrepresentation/t_cf_clinicalnb_gp_both_reac_tree.pdf",
width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_reac_both_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_reac_both_tree' not founddev.off()## png
## 2t_cf_clinicalnb_gp_reac_both_tree## Error in eval(expr, envir, enclos): object 't_cf_clinicalnb_gp_reac_both_tree' not found2.4.2 clusterProfiler
The following essentially repeats the gProfiler2 invocation using clusterProfiler. I have a couple of functions which compare the GO results from various methods, perhaps I should dig it out and see how similar the results are using these two tools. My assumption is that the primary differences should arise from the fact that gProfiler theoretically is updating their GO data over time and cProfiler uses the information in org.Hs.eg.db, which afaik has not changed in quite a while.
2.4.2.1 Genes higher in cure
t_cf_clinical_cp_up <- simple_clusterprofiler(
t_clinical_cf_sig_sva_up, de_table = t_clinical_cf_table_sva,
excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinical_cure_up_cp-v{ver}.xlsx"))## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Reading KEGG annotation online: "https://rest.kegg.jp/link/hsa/pathway"...## Reading KEGG annotation online: "https://rest.kegg.jp/list/pathway/hsa"...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Writing a sheet containing the legend.
##
## Writing the BP data.
##
## Loading required namespace: Vennerable
##
## Writing the MF data.
##
## Writing the KEGG data.
##
## Finished writing excel file.enrichplot::dotplot(t_cf_clinical_cp_up[["enrich_objects"]][["BP_all"]])
2.4.2.2 Genes higher in fail
t_cf_clinical_cp_down <- simple_clusterprofiler(
t_clinical_cf_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinical_fail_up_cp-v{ver}.xlsx"))## Unable to find the fold-change column in the de table, not doing gsea.## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Writing a sheet containing the legend.
##
## Writing the BP data.
##
## Writing the MF data.
##
## Writing the CC data.
##
## Writing the KEGG data.
##
## Finished writing excel file.enrichplot::dotplot(t_cf_clinical_cp_down[["enrich_objects"]][["BP_all"]])
2.4.2.3 GSEA
GSEA is not associated with either up nor down, it takes the rank order of genes with respect (in this case) to fold-change.
t_cf_clinical_topn_gsea <- plot_topn_gsea(t_cf_clinical_cp_up)
t_cf_clinical_topn_gsea[["GO"]][[1]]
t_cf_clinical_topn_gsea[["GO"]][[2]]
t_cf_clinical_topn_gsea[["GO"]][[3]]
t_cf_clinical_topn_gsea[["GO"]][[4]]
t_cf_clinical_topn_gsea[["GO"]][[5]]
2.5 Visit 1 vs. other visits
The relevant xlsx files are:
- “{xlsx_prefix}/DE_Visits/tv1_vs_later_tables-v{ver}.xlsx”
- “{xlsx_prefix}/DE_Visits/tv1_vs_later_sig-v{ver}.xlsx”
input_xlsx <- glue("{xlsx_prefix}/DE_Visits/tv1_vs_later_tables-v{ver}.xlsx")
tv1_vs_later_table <- table_reader(input_xlsx, "later_vs_first")## [1] 11910 68input_xlsx <- glue("{xlsx_prefix}/DE_Visits/tv1_vs_later_sig-v{ver}.xlsx")
tv1_vs_later_up_sig <- sig_reader(input_xlsx, "later_vs_first")
tv1_vs_later_down_sig <- sig_reader(input_xlsx, "later_vs_first", "down")## There are less than 20 rows in this significance table, it is unlikely to be interesting.2.5.1 Increased in visit 1
I am not likely to do the decreased in visit 1, there are only 7 genes.
2.5.1.1 enrichment: gProfiler
tv1later_up_gp <- simple_gprofiler(
tv1_vs_later_up_sig,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/tv1_vs_later_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eetv1later_up_gp## Error in eval(expr, envir, enclos): object 'tv1later_up_gp' not foundenrichplot::dotplot(tv1later_up_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 'tv1later_up_gp' not foundenrichplot::dotplot(tv1later_up_gp[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 'tv1later_up_gp' not found2.5.1.2 enrichment: clusterProfiler
Note to self: make some classes for plot_topn_gsea so it can handle the various inputs it is likely to receive.
tv1later_up_cp <- simple_clusterprofiler(
tv1_vs_later_up_sig, de_table = tv1_vs_later_table,
orgdb = "org.Hs.eg.db", kegg_prefix = "hs", do_kegg = TRUE)## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.enrichplot::dotplot(tv1later_up_cp[["enrich_objects"]][["MF_all"]])
enrichplot::dotplot(tv1later_up_cp[["enrich_objects"]][["BP_all"]])
written <- write_cp_data(
tv1later_up_cp,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/tv1_vs_later_up_cp-v{ver}.xlsx"))## Writing a sheet containing the legend.
## Writing the MF data.
## Writing the CC data.
## Finished writing excel file.2.5.1.3 GSEA: clusterProfiler
tv1later_topn_gsea <- plot_topn_gsea(tv1later_up_cp)
tv1later_topn_gsea[["GO"]][[1]]
tv1later_topn_gsea[["GO"]][[2]]
2.6 Patient Sex
The relevant input xlsx files are:
- {xlsx_prefix}/DE_Sex/t_sex_table-v{ver}.xlsx”))
- {xlsx_prefix}/DE_Sex/t_sex_sig-v{ver}.xlsx”))
Oh, I messed up and put the DE results in the GSEA directory!
Let us see if we observe general male/female differences in the data. This has an important caveat: there are few female failures in the dataset and so the results may reflect that.
input_xlsx <- glue("{xlsx_prefix}/DE_Sex/t_sex_cure_table-v{ver}.xlsx")
t_sex_table <- table_reader(input_xlsx, "male_vs_female")## [1] 13971 68input_xlsx <- glue("{xlsx_prefix}/DE_Sex/t_sex_cure_sig-v{ver}.xlsx")
t_sex_up_sig <- sig_reader(input_xlsx, "male_vs_female")
t_sex_down_sig <- sig_reader(input_xlsx, "male_vs_female", "down")2.6.1 gProfiler
2.6.1.1 Increased in men
t_sex_up_gp <- simple_gprofiler(
t_sex_up_sig,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/sex_male_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_sex_up_gp## Error in eval(expr, envir, enclos): object 't_sex_up_gp' not foundenrichplot::dotplot(t_sex_up_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_sex_up_gp' not foundenrichplot::dotplot(t_sex_up_gp[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_sex_up_gp' not foundenrichplot::dotplot(t_sex_up_gp[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_sex_up_gp' not found2.6.1.2 Increased in women
t_sex_down_gp <- simple_gprofiler(
t_sex_down_sig,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/sex_female_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_sex_down_gp## Error in eval(expr, envir, enclos): object 't_sex_down_gp' not foundGiven the results from gProfiler, I do not have high expectations for clusterProfiler, so I am not likely to take the time to write them out. It seems like male/female is confounded with cure/fail.
2.6.2 clusterProfiler
2.6.2.1 Increased in men
t_sex_up_cp <- simple_clusterprofiler(
t_sex_up_sig, de_table = t_sex_table,
orgdb = "org.Hs.eg.db", kegg_prefix = "hs", do_kegg = TRUE)## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.enrichplot::dotplot(t_sex_up_cp[["enrich_objects"]][["BP_all"]])
2.6.2.2 Increased in women
t_sex_down_cp <- simple_clusterprofiler(
t_sex_down_sig, de_table = t_sex_table,
orgdb = "org.Hs.eg.db", kegg_prefix = "hs", do_kegg = TRUE)## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...enrichplot::dotplot(t_sex_down_cp[["enrich_objects"]][["BP_all"]])
2.6.2.3 GSEA
t_sex_topn_gsea <- plot_topn_gsea(t_sex_up_cp)
t_sex_topn_gsea[["GO"]][[1]]
2.7 Ethnicity
Once again, the relevant files to load:
- “{xlsx_prefix}/DE_Ethnicity/t_ethnicity_table-v{ver}.xlsx”
- “{xlsx_prefix}/DE_Ethnicity/t_ethnicity_sig-v{ver}.xlsx”
input_xlsx <- glue("{xlsx_prefix}/DE_Ethnicity/t_ethnicity_table-v{ver}.xlsx")
t_ethnicity_mestizo_indigenous <- table_reader(input_xlsx, "mestizo_indigenous")## [1] 14156 68t_ethnicity_mestizo_afrocol <- table_reader(input_xlsx, "mestizo_afrocol")## [1] 14156 68t_ethnicity_indigenous_afrocol <- table_reader(input_xlsx, "indigenous_afrocol")## [1] 14156 68input_xlsx <- glue("{xlsx_prefix}/DE_Ethnicity/t_ethnicity_sig-v{ver}.xlsx")
t_ethnicity_mestizo_indigenous_up <- sig_reader(input_xlsx, "mestizo_indigenous")
t_ethnicity_mestizo_indigenous_down <- sig_reader(input_xlsx, "mestizo_indigenous", "down")
t_ethnicity_mestizo_afrocol_up <- sig_reader(input_xlsx, "mestizo_afrocol")
t_ethnicity_mestizo_afrocol_down <- sig_reader(input_xlsx, "mestizo_afrocol", "down")
t_ethnicity_indigenous_afrocol_up <- sig_reader(input_xlsx, "indigenous_afrocol")
t_ethnicity_indigenous_afrocol_down <- sig_reader(input_xlsx, "indigenous_afrocol", "down")2.7.1 gProfiler
2.7.1.1 Increased in mestizo vs indigenous
mestizo_indigenous_up_gp <- simple_gprofiler(
t_ethnicity_mestizo_indigenous_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_mi_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eemestizo_indigenous_up_gp## Error in eval(expr, envir, enclos): object 'mestizo_indigenous_up_gp' not found2.7.1.2 Increased in indigenous vs mestizo
indigenous_mestizo_up_gp <- simple_gprofiler(
t_ethnicity_mestizo_indigenous_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_im_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eeindigenous_mestizo_up_gp## Error in eval(expr, envir, enclos): object 'indigenous_mestizo_up_gp' not found2.7.1.3 Increased in mestizo vs afrocolombian
mestizo_afrocol_up_gp <- simple_gprofiler(
t_ethnicity_mestizo_afrocol_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_ma_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eemestizo_afrocol_up_gp## Error in eval(expr, envir, enclos): object 'mestizo_afrocol_up_gp' not found2.7.1.4 Increased in afrocolombian vs mestizo
afrocol_mestizo_up_gp <- simple_gprofiler(
t_ethnicity_mestizo_afrocol_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_am_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eeafrocol_mestizo_up_gp## Error in eval(expr, envir, enclos): object 'afrocol_mestizo_up_gp' not found2.7.1.5 Increased in indigenous vs afrocolombian
indigenous_afrocol_up_gp <- simple_gprofiler(
t_ethnicity_indigenous_afrocol_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_ia_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eeindigenous_afrocol_up_gp## Error in eval(expr, envir, enclos): object 'indigenous_afrocol_up_gp' not found2.7.1.6 Increased in afrocolombian vs indigenous
afrocol_indigenous_up_gp <- simple_gprofiler(
t_ethnicity_indigenous_afrocol_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_ai_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eeafrocol_indigenous_up_gp## Error in eval(expr, envir, enclos): object 'afrocol_indigenous_up_gp' not found2.7.2 clusterProfiler
2.7.2.1 Increased in mestizo vs indigenous
mestizo_indigenous_up_cp <- simple_clusterprofiler(
t_ethnicity_mestizo_indigenous_up,
de_table = t_ethnicity_mestizo_indigenous)## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.mestizo_indigenous_up_cp## A set of ontologies produced by clusterprofiler.2.7.2.2 Increased in indigenous vs mestizo
indigenous_mestizo_up_cp <- simple_clusterprofiler(
t_ethnicity_mestizo_indigenous_down)## Unable to find the fold-change column in the de table, not doing gsea.## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.indigenous_mestizo_up_cp## A set of ontologies produced by clusterprofiler.2.7.2.3 Increased in mestizo vs afrocolombian
mestizo_afrocol_up_cp <- simple_clusterprofiler(
t_ethnicity_mestizo_afrocol_up,
de_table = t_ethnicity_mestizo_afrocol)## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.mestizo_afrocol_up_cp## A set of ontologies produced by clusterprofiler.2.7.2.4 Increased in afrocolombian vs mestizo
afrocol_mestizo_up_cp <- simple_clusterprofiler(t_ethnicity_mestizo_afrocol_down)## Unable to find the fold-change column in the de table, not doing gsea.afrocol_mestizo_up_cp## A set of ontologies produced by clusterprofiler.2.7.2.5 Increased in indigenous vs afrocolombian
indigenous_afrocol_up_cp <- simple_clusterprofiler(
t_ethnicity_indigenous_afrocol_up,
de_table = t_ethnicity_indigenous_afrocol)## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.indigenous_afrocol_up_cp## A set of ontologies produced by clusterprofiler.2.7.2.6 Increased in afrocolombian vs indigenous
afrocol_indigenous_up_cp <- simple_clusterprofiler(t_ethnicity_indigenous_afrocol_down)## Unable to find the fold-change column in the de table, not doing gsea.## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.afrocol_indigenous_up_cp## A set of ontologies produced by clusterprofiler.2.8 Visit 1 cure/fail
It looks like there are very few groups in the visit 1 significant genes.
The relevant xlsx files are:
- “{cf_prefix}/Visits/t_clinical_v1_cf_table_sva-v{ver}.xlsx”
- “{cf_prefix}/Visits/t_clinical_v1_cf_sig_sva-v{ver}.xlsx”
input_xlsx <- glue("{cf_prefix}/Visits/t_clinical_v1_cf_table_sva-v{ver}.xlsx")
t_cf_clinical_v1_table_sva <- table_reader(input_xlsx)## [1] 14023 68input_xlsx <- glue("{cf_prefix}/Visits/t_clinical_v1_cf_sig_sva-v{ver}.xlsx")
t_cf_clinical_v1_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_clinical_v1_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")2.8.1 gProfiler
2.8.1.1 Increased in visit 1 cure/fail
t_cf_clinical_v1_sig_sva_up_gp <- simple_gprofiler(t_cf_clinical_v1_sig_sva_up)## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_clinical_v1_sig_sva_up_gp## Error in eval(expr, envir, enclos): object 't_cf_clinical_v1_sig_sva_up_gp' not found2.8.1.2 Increased in visit 1 fail/cure
t_cf_clinical_v1_sig_sva_down_gp <- simple_gprofiler(t_cf_clinical_v1_sig_sva_down)## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_clinical_v1_sig_sva_down_gp## Error in eval(expr, envir, enclos): object 't_cf_clinical_v1_sig_sva_down_gp' not foundenrichplot::dotplot(t_cf_clinical_v1_sig_sva_down_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_clinical_v1_sig_sva_down_gp' not found2.8.2 clusterProfiler
2.8.2.1 Increased visit 1 cure/fail
t_cf_clinical_v1_sig_sva_up_cp <- simple_cprofiler(
t_cf_clinical_v1_sig_sva_up,
de_table = t_cf_clinical_v1_table_sva,
orgdb = "org.Hs.eg.db")## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...t_cf_clinical_v1_sig_sva_up_cp## A set of ontologies produced by clusterprofiler.enrichplot::dotplot(t_cf_clinical_v1_sig_sva_up_cp[["enrich_objects"]][["BP_all"]])
2.8.2.2 Increased in visit 1 cure/fail
t_cf_clinical_v1_sig_sva_down_cp <- simple_cprofiler(
t_cf_clinical_v1_sig_sva_down,
orgdb = "org.Hs.eg.db")## Unable to find the fold-change column in the de table, not doing gsea.## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.2.8.2.3 GSEA
It appears there are too few results to perform the gsea plots.
2.9 Cure/Fail, Biopsies
The relevant xlsx output may be found at:
- “{cf_prefix}/Biopsies/t_biopsy_cf_table_sva-v{ver}.xlsx”
- “{cf_prefix}/Biopsies/t_cf_biopsy_sig_sva-v{ver}.xlsx”
input_xlsx <- glue("{cf_prefix}/Biopsies/t_biopsy_cf_table_sva-v{ver}.xlsx")
t_cf_biopsy_table_sva <- table_reader(input_xlsx, "outcome")## [1] 13513 68input_xlsx <- glue("{cf_prefix}/Biopsies/t_cf_biopsy_sig_sva-v{ver}.xlsx")
t_cf_biopsy_sig_sva_up <- sig_reader(input_xlsx, "outcome")## There are less than 20 rows in this significance table, it is unlikely to be interesting.t_cf_biopsy_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")## There are less than 20 rows in this significance table, it is unlikely to be interesting.t_cf_biopsy_sig_sva_both <- rbind.data.frame(t_cf_biopsy_sig_sva_up,
t_cf_biopsy_sig_sva_down)2.9.1 gProfiler
We only have 17 genes in the biopsies, but perhaps they are still interesting?
2.9.1.1 increased in cure vs fail
t_cf_biopsy_sig_sva_gp_up <- simple_gprofiler(
t_cf_biopsy_sig_sva_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_biopsy_sig_sva_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_biopsy_sig_sva_gp_up## Error in eval(expr, envir, enclos): object 't_cf_biopsy_sig_sva_gp_up' not foundenrichplot::dotplot(t_cf_biopsy_sig_sva_gp_up[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_biopsy_sig_sva_gp_up' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_biopsy_sig_sva_gp_up[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_biopsy_sig_sva_gp_up' not foundgo_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = glue("images/overrepresentation/t_cf_biopsy_up_gp_go-v{ver}.pdf"),
height = treeplot_height, width = treeplot_width)
go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not founddev.off()## png
## 2go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not found2.9.1.2 increased in fail vs cure
t_cf_biopsy_sig_sva_gp_down <- simple_gprofiler(
t_cf_biopsy_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_biopsy_sig_sva_down_gp-v{ver}.xlsx"))
t_cf_biopsy_sig_sva_gp_down## A set of ontologies produced by gprofiler using 11
## genes against the hsapiens annotations and significance cutoff 0.05.
## There are:
## 0 MF
## 0 BP
## 0 KEGG
## 0 REAC
## 0 WP
## 0 TF
## 0 MIRNA
## 0 HPA
## 0 CORUM
## 0 HP hits.## Not nearly as interesting2.9.2 clusterprofiler
Again, clusterprofiler version
2.9.2.1 increased in cure/fail
t_cf_biopsy_sig_sva_cp_up <- simple_cprofiler(
t_cf_biopsy_sig_sva_up, de_table = t_cf_biopsy_table_sva,
orgdb = "org.Hs.eg.db",
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_biopsy_sig_sva_up_cp-v{ver}.xlsx"))## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Writing a sheet containing the legend.
##
## Writing the BP data.
##
## Writing the MF data.
##
## Writing the CC data.
##
## Writing the KEGG data.
##
## Finished writing excel file.enrichplot::dotplot(t_cf_biopsy_sig_sva_cp_up[["enrich_objects"]][["BP_all"]])
2.9.2.2 GSEA of cure/fail
t_cf_biopsy_sig_topn_gsea <- plot_topn_gsea(t_cf_biopsy_sig_sva_cp_up)
t_cf_biopsy_sig_topn_gsea[["GO_outcome_up"]][[1]]## NULL2.10 Eosinophils cure/fail
input_xlsx <- glue("{cf_prefix}/Eosinophils/t_eosinophil_cf_table_sva-v{ver}.xlsx")
t_cf_eosinophil_table_sva <- table_reader(input_xlsx, "outcome")## [1] 10532 68input_xlsx <- glue("{cf_prefix}/Eosinophils/t_eosinophil_cf_sig_sva-v{ver}.xlsx")
t_cf_eosinophil_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_eosinophil_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_eosinophil_sig_sva_both <- rbind.data.frame(t_cf_eosinophil_sig_sva_up,
t_cf_eosinophil_sig_sva_down)2.10.1 gProfiler
2.10.1.1 increased in cure vs fail
t_cf_eosinophil_sig_sva_up_gp <- simple_gprofiler(
t_cf_eosinophil_sig_sva_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_eosinophil_sig_sva_up_gp## Error in eval(expr, envir, enclos): object 't_cf_eosinophil_sig_sva_up_gp' not foundenrichplot::dotplot(t_cf_eosinophil_sig_sva_up_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_eosinophil_sig_sva_up_gp' not foundenrichplot::dotplot(t_cf_eosinophil_sig_sva_up_gp[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_eosinophil_sig_sva_up_gp' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_up_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_eosinophil_sig_sva_up_gp' not foundgo_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = "figures/t_cf_eosinophil_up_gp_go.svg",
height = treeplot_height, width = treeplot_width)
go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not founddev.off()## png
## 2go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not foundreac_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_up_gp[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_eosinophil_sig_sva_up_gp' not foundreac_treeplot <- sm(treeplot(reac_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'reac_termsim' not foundpp(file = glue("images/overrepresentation/t_cf_eosinophil_up_gp_reac-v{ver}.pdf"),
height = treeplot_height, width = treeplot_width)
reac_treeplot## Error in eval(expr, envir, enclos): object 'reac_treeplot' not founddev.off()## png
## 22.10.1.2 increased in fail vs cure
t_cf_eosinophil_sig_sva_down_gp <- simple_gprofiler(
t_cf_eosinophil_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_down_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eeenrichplot::dotplot(t_cf_eosinophil_sig_sva_down_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_eosinophil_sig_sva_down_gp' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_down_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_eosinophil_sig_sva_down_gp' not foundgo_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = glue("images/overrepresentation/t_cf_eosinophil_down_gp_go-v{ver}.pdf"),
height = treeplot_height, width = treeplot_width)
go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not founddev.off()## png
## 2go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not found## There is only one reactome hit, so not plotting it.2.10.1.3 Both up and down
I evaluated this and the ‘up’ set next to each other and they are extremely similar:
up: 148 GO hits, 68 TF, 0 HPA both: 169 GO, 69 TF, and 2 HPA
Otherwise I think they are the same.
t_cf_eosinophil_sig_sva_both_gp <- simple_gprofiler(
t_cf_eosinophil_sig_sva_both,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_both_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_eosinophil_sig_sva_both_gp## Error in eval(expr, envir, enclos): object 't_cf_eosinophil_sig_sva_both_gp' not foundenrichplot::dotplot(t_cf_eosinophil_sig_sva_both_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_eosinophil_sig_sva_both_gp' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_both_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_eosinophil_sig_sva_both_gp' not foundgo_treeplot <- sm(treeplot(go_termsim, label_formap = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = glue("images/overrepresentation/t_cf_eosinophil_both_gp_go-v{ver}.pdf"),
height = treeplot_height, width = treeplot_width)
go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not founddev.off()## png
## 2go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not foundreac_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_both_gp[["REAC_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_eosinophil_sig_sva_both_gp' not foundreac_treeplot <- sm(treeplot(reac_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'reac_termsim' not foundpp(file = glue("images/overrepresentation/t_cf_eosinophil_both_gp_reac-v{ver}.pdf"),
height = treeplot_height, width = treeplot_width)
reac_treeplot## Error in eval(expr, envir, enclos): object 'reac_treeplot' not founddev.off()## png
## 22.10.2 clusterprofiler
2.10.2.1 increased in cure vs fail
t_cf_eosinophil_sig_sva_cp_up <- simple_cprofiler(
t_cf_eosinophil_sig_sva_up, de_table = t_cf_eosinophil_table_sva,
orgdb = "org.Hs.eg.db",
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_sig_sva_up_cp-v{ver}.xlsx"))## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Writing a sheet containing the legend.
##
## Writing the BP data.
##
## Writing the MF data.
##
## Writing the CC data.
##
## Writing the KEGG data.
##
## Finished writing excel file.enrichplot::dotplot(t_cf_eosinophil_sig_sva_cp_up[["enrich_objects"]][["BP_all"]])
2.10.2.2 increased in fail vs cure
t_cf_eosinophil_sig_sva_cp_down <- simple_cprofiler(
t_cf_eosinophil_sig_sva_down,
orgdb = "org.Hs.eg.db",
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_sig_sva_down_cp-v{ver}.xlsx"))## Unable to find the fold-change column in the de table, not doing gsea.## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Writing a sheet containing the legend.
##
## Writing the BP data.
##
## Writing the MF data.
##
## Writing the CC data.
##
## Writing the KEGG data.
##
## Finished writing excel file.enrichplot::dotplot(t_cf_eosinophil_sig_sva_cp_down[["enrich_objects"]][["BP_all"]])
2.10.2.3 GSEA
t_cf_eosinophil_sig_topn_gsea <- plot_topn_gsea(t_cf_eosinophil_sig_sva_cp_up)
t_cf_eosinophil_sig_topn_gsea[["GO_outcome_up"]][[1]]## NULL2.11 Monocytes cure/fail
input_xlsx <- glue("{cf_prefix}/Monocytes/t_monocyte_cf_table_sva-v{ver}.xlsx")
t_cf_monocyte_table_sva <- table_reader(input_xlsx, "outcome")## [1] 10862 68input_xlsx <- glue("{cf_prefix}/Monocytes/t_monocyte_cf_sig_sva-v{ver}.xlsx")
t_cf_monocyte_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_monocyte_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_monocyte_sig_sva_both <- rbind.data.frame(t_cf_monocyte_sig_sva_up,
t_cf_monocyte_sig_sva_down)2.11.1 gProfiler
Now that I am looking back over these results, I am not compeltely certain why I only did the gprofiler search for the sva data…
2.11.1.1 increased in cure vs fail
t_cf_monocyte_sig_sva_up_gp <- simple_gprofiler(
t_cf_monocyte_sig_sva_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_monocyte_sig_sva_up_gp## Error in eval(expr, envir, enclos): object 't_cf_monocyte_sig_sva_up_gp' not foundenrichplot::dotplot(t_cf_monocyte_sig_sva_up_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_monocyte_sig_sva_up_gp' not foundenrichplot::dotplot(t_cf_monocyte_sig_sva_up_gp[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_monocyte_sig_sva_up_gp' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_monocyte_sig_sva_up_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_monocyte_sig_sva_up_gp' not foundgo_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = "figures/overrepresentation/t_cf_monocyte_up_gp_go.svg",
height = treeplot_height, width = treeplot_width)## Warning in pp(file = "figures/overrepresentation/t_cf_monocyte_up_gp_go.svg", :
## The directory: figures/overrepresentation does not exist, will attempt to
## create it.go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not founddev.off()## png
## 2go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not found2.11.1.2 increased in fail vs cure
t_cf_monocyte_sig_sva_down_gp <- simple_gprofiler(
t_cf_monocyte_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_down_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eeenrichplot::dotplot(t_cf_monocyte_sig_sva_down_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_monocyte_sig_sva_down_gp' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_monocyte_sig_sva_down_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_monocyte_sig_sva_down_gp' not foundgo_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = glue("images/overrepresentation/t_cf_monocyte_down_gp_go-v{ver}.pdf"),
height = treeplot_height, width = treeplot_width)
go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not founddev.off()## png
## 2go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not found## Insufficient results to make a tree plot.2.11.1.3 Both up and down
t_cf_monocyte_sig_sva_both_gp <- simple_gprofiler(
t_cf_monocyte_sig_sva_both,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_both_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_monocyte_sig_sva_both_gp## Error in eval(expr, envir, enclos): object 't_cf_monocyte_sig_sva_both_gp' not foundenrichplot::dotplot(t_cf_monocyte_sig_sva_both_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_monocyte_sig_sva_both_gp' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_monocyte_sig_sva_both_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_monocyte_sig_sva_both_gp' not foundgo_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = glue("images/overrepresentation/t_cf_monocyte_both_gp_go-v{ver}.pdf"),
height = treeplot_height, width = treeplot_width)
go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not founddev.off()## png
## 2go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not found## Insufficient results for reactome
tf_termsim <- enrichplot::pairwise_termsim(t_cf_monocyte_sig_sva_both_gp[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_monocyte_sig_sva_both_gp' not foundtf_treeplot <- sm(treeplot(tf_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'tf_termsim' not foundpp(file = glue("images/overrepresentation/t_cf_monocyte_both_gp_tf-v{ver}.pdf"),
height = treeplot_height, width = treeplot_width)
tf_treeplot## Error in eval(expr, envir, enclos): object 'tf_treeplot' not founddev.off()## png
## 2tf_treeplot## Error in eval(expr, envir, enclos): object 'tf_treeplot' not found2.11.2 clusterprofiler
2.11.2.1 increased in cure vs fail
t_cf_monocyte_sig_sva_cp_up <- simple_cprofiler(
t_cf_monocyte_sig_sva_up, de_table = t_cf_monocyte_table_sva,
orgdb = "org.Hs.eg.db",
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_sig_sva_up_cp-v{ver}.xlsx"))## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Writing a sheet containing the legend.
##
## Writing the BP data.
##
## Writing the KEGG data.
##
## Finished writing excel file.enrichplot::dotplot(t_cf_monocyte_sig_sva_cp_up[["enrich_objects"]][["BP_all"]])
2.11.2.2 increased in fail vs cure
t_cf_monocyte_sig_sva_cp_down <- simple_cprofiler(
t_cf_monocyte_sig_sva_down,
orgdb = "org.Hs.eg.db",
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_sig_sva_down_cp-v{ver}.xlsx"))## Unable to find the fold-change column in the de table, not doing gsea.## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Writing a sheet containing the legend.
##
## Writing the BP data.
##
## Writing the MF data.
##
## Writing the CC data.
##
## Writing the KEGG data.
##
## Finished writing excel file.enrichplot::dotplot(t_cf_monocyte_sig_sva_cp_down[["enrich_objects"]][["BP_all"]])
2.11.2.3 GSEA
t_cf_monocyte_sig_topn_gsea <- plot_topn_gsea(t_cf_monocyte_sig_sva_cp_up)
t_cf_monocyte_sig_topn_gsea[["GO_outcome_up"]][[1]]## NULL2.12 Neutrophils cure/fail
input_xlsx <- glue("{cf_prefix}/Neutrophils/t_neutrophil_cf_table_sva-v{ver}.xlsx")
t_cf_neutrophil_table_sva <- table_reader(input_xlsx, "outcome")## [1] 9101 68input_xlsx <- glue("{cf_prefix}/Neutrophils/t_neutrophil_cf_sig_sva-v{ver}.xlsx")
t_cf_neutrophil_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_neutrophil_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_neutrophil_sig_sva_both <- rbind.data.frame(t_cf_neutrophil_sig_sva_up,
t_cf_neutrophil_sig_sva_down)2.12.1 gProfiler
Now that I am looking back over these results, I am not compeltely certain why I only did the gprofiler search for the sva data…
2.12.1.1 increased in cure vs fail
t_cf_neutrophil_sig_sva_up_gp <- simple_gprofiler(
t_cf_neutrophil_sig_sva_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_up_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_neutrophil_sig_sva_up_gp## Error in eval(expr, envir, enclos): object 't_cf_neutrophil_sig_sva_up_gp' not foundenrichplot::dotplot(t_cf_neutrophil_sig_sva_up_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_neutrophil_sig_sva_up_gp' not foundenrichplot::dotplot(t_cf_neutrophil_sig_sva_up_gp[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_neutrophil_sig_sva_up_gp' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_neutrophil_sig_sva_up_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_neutrophil_sig_sva_up_gp' not foundgo_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = "figures/t_cf_neutrophil_up_gp_go.svg",
height = treeplot_height, width = treeplot_width)
go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not founddev.off()## png
## 2go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not found2.12.1.2 increased in fail vs cure
t_cf_neutrophil_sig_sva_down_gp <- simple_gprofiler(
t_cf_neutrophil_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_neutrophil_sig_sva_down_gp## Error in eval(expr, envir, enclos): object 't_cf_neutrophil_sig_sva_down_gp' not found## Not much to work with here.2.12.1.3 Both up and down
t_cf_neutrophil_sig_sva_both_gp <- simple_gprofiler(
t_cf_neutrophil_sig_sva_both,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_both_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_neutrophil_sig_sva_both_gp## Error in eval(expr, envir, enclos): object 't_cf_neutrophil_sig_sva_both_gp' not foundenrichplot::dotplot(t_cf_neutrophil_sig_sva_both_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_neutrophil_sig_sva_both_gp' not foundenrichplot::dotplot(t_cf_neutrophil_sig_sva_both_gp[["TF_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'object' in selecting a method for function 'dotplot': object 't_cf_neutrophil_sig_sva_both_gp' not foundgo_termsim <- enrichplot::pairwise_termsim(t_cf_neutrophil_sig_sva_both_gp[["BP_enrich"]])## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'pairwise_termsim': object 't_cf_neutrophil_sig_sva_both_gp' not foundgo_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))## Error in h(simpleError(msg, call)): error in evaluating the argument 'x' in selecting a method for function 'treeplot': object 'go_termsim' not foundpp(file = glue("images/overrepresentation/t_cf_neutrophil_both_gp_go-v{ver}.pdf"),
height = treeplot_height, width = treeplot_width)
go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not founddev.off()## png
## 2go_treeplot## Error in eval(expr, envir, enclos): object 'go_treeplot' not found## Insufficient results for reactome2.12.2 clusterprofiler
2.12.2.1 increased in cure vs fail
t_cf_neutrophil_sig_sva_cp_up <- simple_cprofiler(
t_cf_neutrophil_sig_sva_up, de_table = t_cf_neutrophil_table_sva,
orgdb = "org.Hs.eg.db",
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_sig_sva_up_cp-v{ver}.xlsx"))## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Writing a sheet containing the legend.
##
## Writing the BP data.
##
## Writing the MF data.
##
## Writing the CC data.
##
## Finished writing excel file.enrichplot::dotplot(t_cf_neutrophil_sig_sva_cp_up[["enrich_objects"]][["BP_all"]])
2.12.2.2 increased in fail vs cure
t_cf_neutrophil_sig_sva_cp_down <- simple_cprofiler(
t_cf_neutrophil_sig_sva_down,
orgdb = "org.Hs.eg.db",
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_sig_sva_down_cp-v{ver}.xlsx"))## Unable to find the fold-change column in the de table, not doing gsea.## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Writing a sheet containing the legend.
##
## Writing the MF data.
##
## Writing the CC data.
##
## Writing the KEGG data.
##
## Finished writing excel file.enrichplot::dotplot(t_cf_neutrophil_sig_sva_cp_down[["enrich_objects"]][["BP_all"]])
2.12.2.3 GSEA
t_cf_neutrophil_sig_topn_gsea <- plot_topn_gsea(t_cf_neutrophil_sig_sva_cp_up)
t_cf_neutrophil_sig_topn_gsea[["GO_outcome_up"]][[1]]## NULL2.13 Look at visit 1 and see if anything pops out
input_xlsx <- glue("{cf_prefix}/Monocytes/t_monocyte_v1_cf_table_sva-v{ver}.xlsx")
t_cf_monocyte_v1_table_sva <- table_reader(input_xlsx, "outcome")## [1] 10482 68input_xlsx <- glue("{cf_prefix}/Monocytes/t_monocyte_v1_cf_sig_sva-v{ver}.xlsx")
t_cf_monocyte_v1_sig_sva_up <- sig_reader(input_xlsx, "outcome")## There are less than 20 rows in this significance table, it is unlikely to be interesting.t_cf_monocyte_v1_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_monocyte_v1_sig_sva_both <- rbind.data.frame(t_cf_monocyte_v1_sig_sva_up,
t_cf_monocyte_v1_sig_sva_down)2.13.1 Gene Set Enrichment: gProfiler Monocytes by visit, V1
V1: Up: 14 genes; No categories V1: Down: 52 genes; 20 GO, 5 TF
t_cf_monocyte_v1_sig_sva_up_gp <- simple_gprofiler(
t_cf_monocyte_v1_sig_sva_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_up_gp-v{ver}.xlsx"))
t_cf_monocyte_v1_sig_sva_up_gp## A set of ontologies produced by gprofiler using 14
## genes against the hsapiens annotations and significance cutoff 0.05.
## There are:
## 0 MF
## 0 BP
## 0 KEGG
## 0 REAC
## 0 WP
## 0 TF
## 0 MIRNA
## 0 HPA
## 0 CORUM
## 0 HP hits.t_cf_monocyte_v1_sig_sva_down_gp <- simple_gprofiler(
t_cf_monocyte_v1_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_monocyte_v1_sig_sva_down_gp## Error in eval(expr, envir, enclos): object 't_cf_monocyte_v1_sig_sva_down_gp' not foundt_cf_monocyte_v1_sig_sva_both_gp <- simple_gprofiler(
t_cf_monocyte_v1_sig_sva_both,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_monocyte_v1_sig_sva_both_gp## Error in eval(expr, envir, enclos): object 't_cf_monocyte_v1_sig_sva_both_gp' not foundI like cats!
2.13.1.1 clusterProfiler
t_cf_monocyte_v1_sig_sva_up_cp <- simple_cprofiler(t_cf_monocyte_v1_sig_sva_up,
t_cf_monocyte_v1_table_sva,
orgdb = "org.Hs.eg.db")## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...enrichplot::dotplot(t_cf_monocyte_v1_sig_sva_up_cp[["enrich_objects"]][["BP_all"]])
t_cf_monocyte_v1_topn_gsea <- plot_topn_gsea(t_cf_monocyte_v1_sig_sva_up_cp)
t_cf_monocyte_v1_topn_gsea[["GO_outcome_up"]][[1]]## NULL2.13.2 Neutrophils v1
input_xlsx <- glue("{cf_prefix}/Neutrophils/t_neutrophil_v1_cf_table_sva-v{ver}.xlsx")
t_cf_neutrophil_v1_table_sva <- table_reader(input_xlsx, "outcome")## [1] 8717 68input_xlsx <- glue("{cf_prefix}/Neutrophils/t_neutrophil_v1_cf_sig_sva-v{ver}.xlsx")
t_cf_neutrophil_v1_sig_sva_up <- sig_reader(input_xlsx, "outcome")## There are less than 20 rows in this significance table, it is unlikely to be interesting.t_cf_neutrophil_v1_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")## There are less than 20 rows in this significance table, it is unlikely to be interesting.t_cf_neutrophil_v1_sig_sva_both <- rbind.data.frame(t_cf_neutrophil_v1_sig_sva_up,
t_cf_neutrophil_v1_sig_sva_down)2.13.3 Gene Set Enrichment: gProfiler Neutrophils by visit, V1
t_cf_neutrophil_v1_sig_sva_up_gp <- simple_gprofiler(
t_cf_neutrophil_v1_sig_sva_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_up_gp-v{ver}.xlsx"))## There are only, 5 returning null.t_cf_neutrophil_v1_sig_sva_up_gp## NULLt_cf_neutrophil_v1_sig_sva_down_gp <- simple_gprofiler(
t_cf_neutrophil_v1_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))## There are only, 8 returning null.t_cf_neutrophil_v1_sig_sva_down_gp## NULLt_cf_neutrophil_v1_sig_sva_both_gp <- simple_gprofiler(
t_cf_neutrophil_v1_sig_sva_both,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))
t_cf_neutrophil_v1_sig_sva_both_gp## A set of ontologies produced by gprofiler using 13
## genes against the hsapiens annotations and significance cutoff 0.05.
## There are:
## 0 MF
## 0 BP
## 0 KEGG
## 0 REAC
## 0 WP
## 0 TF
## 0 MIRNA
## 0 HPA
## 0 CORUM
## 0 HP hits.2.13.3.1 clusterProfiler
t_cf_neutrophil_v1_sig_sva_up_cp <- simple_cprofiler(t_cf_neutrophil_v1_sig_sva_up,
t_cf_neutrophil_v1_table_sva,
orgdb = "org.Hs.eg.db")## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.
## Scale for size is already present.
## Adding another scale for size, which will replace the existing scale.enrichplot::dotplot(t_cf_neutrophil_v1_sig_sva_up_cp[["enrich_objects"]][["BP_all"]])
t_cf_neutrophil_v1_topn_gsea <- plot_topn_gsea(t_cf_neutrophil_v1_sig_sva_up_cp)
t_cf_neutrophil_v1_topn_gsea[["GO_outcome_up"]][[1]]## NULL2.13.4 Eosinophils v1
input_xlsx <- glue("{cf_prefix}/Eosinophils/t_eosinophil_v1_cf_table_sva-v{ver}.xlsx")
t_cf_eosinophil_v1_table_sva <- table_reader(input_xlsx, "outcome")## [1] 9979 68input_xlsx <- glue("{cf_prefix}/Eosinophils/t_eosinophil_v1_cf_sig_sva-v{ver}.xlsx")
t_cf_eosinophil_v1_sig_sva_up <- sig_reader(input_xlsx, "outcome")## There are less than 20 rows in this significance table, it is unlikely to be interesting.t_cf_eosinophil_v1_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")## There are less than 20 rows in this significance table, it is unlikely to be interesting.t_cf_eosinophil_v1_sig_sva_both <- rbind.data.frame(t_cf_eosinophil_v1_sig_sva_up,
t_cf_eosinophil_v1_sig_sva_down)2.13.5 Gene Set Enrichment: gProfiler Eosinophils by visit, V1
t_cf_eosinophil_v1_sig_sva_up_gp <- simple_gprofiler(
t_cf_eosinophil_v1_sig_sva_up,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_up_gp-v{ver}.xlsx"))
t_cf_eosinophil_v1_sig_sva_up_gp## A set of ontologies produced by gprofiler using 13
## genes against the hsapiens annotations and significance cutoff 0.05.
## There are:
## 0 MF
## 0 BP
## 0 KEGG
## 0 REAC
## 0 WP
## 0 TF
## 0 MIRNA
## 0 HPA
## 0 CORUM
## 0 HP hits.t_cf_eosinophil_v1_sig_sva_down_gp <- simple_gprofiler(
t_cf_eosinophil_v1_sig_sva_down,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_down_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_eosinophil_v1_sig_sva_down_gp## Error in eval(expr, envir, enclos): object 't_cf_eosinophil_v1_sig_sva_down_gp' not foundt_cf_eosinophil_v1_sig_sva_both_gp <- simple_gprofiler(
t_cf_eosinophil_v1_sig_sva_both,
excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_down_gp-v{ver}.xlsx"))## Error in gprofiler_request(url, body): There's an issue with your request to g:Profiler.
## Error code: 503.
## Please double check your input. If this doesn't help, then check your internet connection or contact us with a reproducible example on biit.support@ut.eet_cf_eosinophil_v1_sig_sva_both_gp## Error in eval(expr, envir, enclos): object 't_cf_eosinophil_v1_sig_sva_both_gp' not found2.13.5.1 clusterProfiler
t_cf_eosinophil_v1_sig_sva_up_cp <- simple_cprofiler(t_cf_eosinophil_v1_sig_sva_up,
t_cf_eosinophil_v1_table_sva,
orgdb = "org.Hs.eg.db")## preparing geneSet collections...## GSEA analysis...## leading edge analysis...## done...enrichplot::dotplot(t_cf_eosinophil_v1_sig_sva_up_cp[["enrich_objects"]][["BP_all"]])
t_cf_eosinophil_v1_topn_gsea <- plot_topn_gsea(t_cf_eosinophil_v1_sig_sva_up_cp)
t_cf_eosinophil_v1_topn_gsea[["GO_outcome_up"]][[1]]## NULLBibliography
Ashburner, Michael, Catherine A. Ball, Judith A. Blake, David Botstein,
Heather Butler, J. Michael Cherry, Allan P. Davis, et al. 2000.
“Gene Ontology: Tool for the Unification of
Biology.” Nature Genetics 25 (1): 25–29. https://doi.org/10.1038/75556.
Croft, David, Gavin O’Kelly, Guanming Wu, Robin Haw, Marc Gillespie,
Lisa Matthews, Michael Caudy, et al. 2011. “Reactome: A Database
of Reactions, Pathways and Biological Processes.” Nucleic
Acids Research 39 (suppl_1): D691–97. https://doi.org/10.1093/nar/gkq1018.
Giurgiu, Madalina, Julian Reinhard, Barbara Brauner, Irmtraud
Dunger-Kaltenbach, Gisela Fobo, Goar Frishman, Corinna Montrone, and
Andreas Ruepp. 2019. “CORUM: The Comprehensive
Resource of Mammalian Protein Complexes—2019.” Nucleic Acids
Research 47 (D1): D559–63. https://doi.org/10.1093/nar/gky973.
Hsu, Sheng-Da, Feng-Mao Lin, Wei-Yun Wu, Chao Liang, Wei-Chih Huang,
Wen-Ling Chan, Wen-Ting Tsai, et al. 2011. “miRTarBase: A Database Curates Experimentally
Validated microRNA–Target
Interactions.” Nucleic Acids Research 39 (suppl_1):
D163–69. https://doi.org/10.1093/nar/gkq1107.
Kanehisa, Minoru, and Susumu Goto. 2000. “KEGG:
Kyoto Encyclopedia of Genes and
Genomes.” Nucleic Acids Research 28 (1):
27–30. https://doi.org/10.1093/nar/28.1.27.
Köhler, Sebastian, Nicole A. Vasilevsky, Mark Engelstad, Erin Foster,
Julie McMurry, Ségolène Aymé, Gareth Baynam, et al. 2017. “The
Human Phenotype Ontology in 2017.” Nucleic Acids
Research 45 (D1): D865–76. https://doi.org/10.1093/nar/gkw1039.
Kolberg, Liis, Uku Raudvere, Ivan Kuzmin, Jaak Vilo, and Hedi Peterson.
2020. “Gprofiler2 – an R Package for Gene List
Functional Enrichment Analysis and Namespace Conversion Toolset
g:Profiler.” F1000Research 9: ELIXIR–709.
https://doi.org/10.12688/f1000research.24956.2.
Kutmon, Martina, Anders Riutta, Nuno Nunes, Kristina Hanspers, Egon L.
Willighagen, Anwesha Bohler, Jonathan Mélius, et al. 2016.
“WikiPathways: Capturing the Full Diversity of
Pathway Knowledge.” Nucleic Acids Research 44 (D1):
D488–94. https://doi.org/10.1093/nar/gkv1024.
Pontén, F., J. M. Schwenk, A. Asplund, and P.-H. D. Edqvist. 2011.
“The Human Protein Atlas as a Proteomic Resource for
Biomarker Discovery.” Journal of Internal Medicine 270
(5): 428–46. https://doi.org/10.1111/j.1365-2796.2011.02427.x.
Wingender, E., P. Dietze, H. Karas, and R. Knüppel. 1996.
“TRANSFAC: A Database on
Transcription Factors and Their DNA Binding
Sites.” Nucleic Acids Research 24 (1): 238–41. https://doi.org/10.1093/nar/24.1.238.
Yu, Guangchuang. n.d. 📖 Introduction
Biomedical Knowledge Mining Using GOSemSim and
clusterProfiler. Accessed June 21,
2024.
Yu, Guangchuang, Li-Gen Wang, Yanyan Han, and Qing-Yu He. 2012.
“clusterProfiler: An R
Package for Comparing Biological Themes Among Gene Clusters.”
Omics : A Journal of Integrative Biology 16 (5): 284–87. https://doi.org/10.1089/omi.2011.0118.
---
title: "TMRC3 Enrichment analyses of the DE results: `r Sys.getenv('VERSION')`"
author: "atb abelew@gmail.com"
date: "`r Sys.Date()`"
bibliography: atb.bib
output:
  html_document:
    code_download: true
    code_folding: show
    fig_caption: true
    fig_height: 7
    fig_width: 7
    highlight: zenburn
    keep_md: false
    mode: selfcontained
    number_sections: true
    self_contained: true
    theme: readable
    toc: true
    toc_float:
      collapsed: false
      smooth_scroll: false
---

<style type="text/css">
body .main-container {
  max-width: 1600px;
}
body, td {
  font-size: 16px;
}
code.r{
  font-size: 16px;
}
pre {
  font-size: 16px
}
</style>

```{r, include=FALSE}
library(enrichplot)
library(ggstatsplot)
library(glue)
library(hpgltools)
knitr::opts_knit$set(progress = TRUE, verbose = TRUE, width = 90, echo = TRUE)
knitr::opts_chunk$set(
  error = TRUE, fig.width = 8, fig.height = 8, fig.retina = 2,
  out.width = "100%", dev = "png",
  dev.args = list(png = list(type = "cairo-png")))
old_options <- options(digits = 4, stringsAsFactors = FALSE, knitr.duplicate.label = "allow")
ggplot2::theme_set(ggplot2::theme_bw(base_size = 12))
ver <- Sys.getenv("VERSION")
rundate <- format(Sys.Date(), format = "%Y%m%d")
```

# Quick xlsx reader functions

My functions to perform the various overrepresentation/enrichment
analyses generally assume the result of extract_significant_genes() as
the input.  Sadly, I do not want to save those data structures to disk
because they can be quite monstrously large when serialized (because I
keep a reference to the input in them which gets dereferenced on
save()).  As a result, it is far simpler/safer to just read the xlsx
outputs produced by combine_de_tables and extract_significant_genes.
Ergo the following two little functions.

```{r}
table_reader <- function(xlsx, sheet = "outcome") {
  input_df <- openxlsx::readWorkbook(xlsx, sheet = sheet, startRow = 2)
  rownames(input_df) <- input_df[["row.names"]]
  input_df[["row.names"]] <- NULL
  print(dim(input_df))
  if (nrow(input_df) < 5000) {
    warning("Something appears wrong with the de table: ", input_xlsx)
  }
  return(input_df)
}

sig_reader <- function(xlsx, sheet = "outcome", direction = "up") {
  this_sheet <- paste0(direction, "_deseq_", sheet)
  input_df <- openxlsx::readWorkbook(xlsx, sheet = this_sheet, startRow = 2)
  rownames(input_df) <- input_df[["row.names"]]
  input_df[["row.names"]] <- NULL
  dim(input_df)
  if (nrow(input_df) < 20) {
    message("There are less than 20 rows in this significance table, it is unlikely to be interesting.")
  }
  return(input_df)
}
```

I have been having some difficulties getting the treeplot fonts to
match our readability goals.  The set of things I have tried so far
are:

1.  Setting a geom_text(size = x): this failed because it didn't know
    to which element(s) to apply the new size.
2.  Changing the options for theme_bw(): this does change some font
    elements, but not the ones that I wanted.  I tried more than a
    few, most importantly theme(text = element_text(size = x))
3.  Changing the default for theme_bw via theme_update(text =
    element_text(size = x)): this also changes some text, but not all.
4.  Invoking update_geom_defaults("text", aes(size = 10)) at the stop
    of my document.  This does not seem to do much of anything.
5.  Playing with options in enrichplot::treeplot(), notably the cex
    and fontsize options.  These also help with some, but not all
    fonts.  I also spent a fair amount of time reading the code for
    treeplot() and decided to try #6 before I got too frustrated.
6.  Just changing the canvas size of my output pdf.  This worked
    great, though it may require setting the wordwrap options in
    treeplot.

With that in mind, the following two parameters are going to set the
width and height of the output pdf documents which will be sent to
Maria Adelaida in order to arrange into the final figures via
inkscape.

```{r}
treeplot_height <- 6
treeplot_width <- 12
wrap_width <- 24
desired_go_ont <- "BP"
```

# Introduction

I once again moved the ontology analyses to their own document.  This
is primarily because I want a fresh notebook to play around with these
and get some of the resulting clutter out of the DE notebooks.

The con of doing this is that some of my enrichment methods are smart
enough to directly take the output from
combine_de_tables()/extract_significant_genes().  I guess I can
dispatch a method to take the xlsx file as input...

Thus, for each enrichment analysis that _was_ in the differential
expression documents, there will now be a few stanzas here, one
which loads the appropriate xlsx file, one which performs gProfiler,
and one which uses clusterProfiler.  On occasion there may be another
with random stuff like me poking at transcription factors or other
side interests...

In addition, I am going to do the Tumaco-only analyses first, because
they are what are actually in the paper.

## Gene Set Enrichment

The gene set enrichment will follow each DE analysis during this
document.  I am adding a series of explicitly GSEA analyses in this
most recent iteration, in these I will pass the full DE table and
check the distribution of logFC values against the genes in each
category as opposed to the simpler over-enrichment of the high/low DE
genes.

Most (all?) of the gene set enrichment analyses performed in this
document are a combination of gProfiler2
(@kolbergGprofiler2PackageGene2020) and clusterProfiler
(@yuClusterProfilerPackageComparing2012).  There are functions
available in hpgltools to also perform a few other methods, but for
the moment I am sticking to only these two.  Oh yeah, I need to wrap
clusterProfiler's tree (taken from topGO) plotter because it spams
plots _everywhere_...

```{r}
input_xlsx <- glue("analyses/3_cali_and_tumaco/DE_Cure_Fail/Clinical_Samples/tc_clinical_cf_table_sva-v{ver}.xlsx")
all_de_cf_table <- table_reader(input_xlsx, "outcome")

all_de_cf_up <- sig_reader(glue("analyses/3_cali_and_tumaco/DE_Cure_Fail/Clinical_Samples/tc_clinical_cf_sig_sva-v{ver}.xlsx"), "outcome", "up")
all_de_cf_down <- sig_reader(glue("analyses/3_cali_and_tumaco/DE_Cure_Fail/Clinical_Samples/tc_clinical_cf_sig_sva-v{ver}.xlsx"), "outcome", "down")
all_de_cf <- rbind.data.frame(all_de_cf_up, all_de_cf_down)

tumaco_xlsx <- glue("analyses/4_tumaco/DE_Cure_Fail/t_all_visitcf_sig_sva-v{ver}.xlsx")
t_de_cf <- openxlsx::readWorkbook(tumaco_xlsx, sheet = 3, startRow = 2)
rownames(t_de_cf) <- t_de_cf[["row.names"]]
t_de_cf[["row.names"]] <- NULL
t_de_cf_up <- t_de_cf[, c("deseq_logfc", "deseq_adjp", "deseq_basemean", "deseq_lfcse")]
t_de_cf <- openxlsx::readWorkbook(tumaco_xlsx, sheet = 4, startRow = 2)
rownames(t_de_cf) <- t_de_cf[["row.names"]]
t_de_cf[["row.names"]] <- NULL
t_de_cf_down <- t_de_cf[, c("deseq_logfc", "deseq_adjp", "deseq_basemean", "deseq_lfcse")]
t_de_cf <- rbind.data.frame(t_de_cf_up, t_de_cf_down)
```

## gProfiler search of all Tumaco samples

The following gProfiler searches use the all_gprofiler() function
instead of simple_gprofiler().  As a result, the results are separated
by {contrast}_{direction}.  Thus 'outcome_down'.

The same plots are available as the previous gProfiler searches, but
in many of the following runs, I used the dotplot() function to get a
slightly different view of the results.

## Shared paths

These are the paths from the DE document and which will be used to get
the inputs for gprofiler/clusterprofiler.

```{r}
xlsx_prefix <- "analyses/4_tumaco"
cf_prefix <- glue("{xlsx_prefix}/DE_Cure_Fail")
```

## All Tumaco clinical samples cure/fail

The xlsx files are:

* "{cf_prefix}/All_Samples/t_clinical_cf_table_sva-v{ver}.xlsx"
* "{cf_prefix}/All_Samples/t_clinical_cf_sig_sva-v{ver}.xlsx"

```{r}
input_xlsx <- glue("{cf_prefix}/All_Samples/t_clinical_cf_table_sva-v{ver}.xlsx")
t_clinical_cf_table_sva <- table_reader(input_xlsx, "outcome")

input_xlsx <- glue("{cf_prefix}/All_Samples/t_clinical_cf_sig_sva-v{ver}.xlsx")
t_clinical_cf_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_clinical_cf_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
```

And without biopsies

```{r}
input_xlsx <- glue("{cf_prefix}/All_Samples/t_clinical_nobiop_cf_table_sva-v{ver}.xlsx")
t_clinicalnb_cf_table_sva <- table_reader(input_xlsx, "outcome")

input_xlsx <- glue("{cf_prefix}/All_Samples/t_clinical_nobiop_cf_sig_sva-v{ver}.xlsx")
t_clinicalnb_cf_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_clinicalnb_cf_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_clinicalnb_cf_sig_sva_both <- rbind.data.frame(t_clinicalnb_cf_sig_sva_up,
                                                 t_clinicalnb_cf_sig_sva_down)
```

### gProfiler

After most of the DE analyses, the set of significantly DE genes gets
passed to gProfiler2 and clusterProfiler.  One slightly neat thing in
my gprofiler (and goseq/topGO/gostats) function(s): it coerces the
result into the same datastructure produced by clusterProfiler, thus
one may play with the various plotting functions in the enrichplot
(@yuIntroductionBiomedicalKnowledge) package.  This is kind of fun
because gProfiler2 provides easy access to a few datasets:

1. GO: (@ashburnerGeneOntologyTool2000a)
2. KEGG: (@kanehisaKEGGKyotoEncyclopedia2000)
3. Reactome: (@croftReactomeDatabaseReactions2011)
4. WikiPathways: (@kutmonWikiPathwaysCapturingFull2016)
5. Transfac: (@wingenderTRANSFACDatabaseTranscription1996)
6. miRTarBase: (@hsuMiRTarBaseDatabaseCurates2011)
7. The Human Protein Atlas: (@pontenHumanProteinAtlas2011)
8. Corum: (@giurgiuCORUMComprehensiveResource2019)
9. Human phenotype ontology: (@kohlerHumanPhenotypeOntology2017)

#### Genes higher in cure

My general sense is that the comparisons of primary interest are
reactome and one or more GO.  I suspect that if there are lots of
transcription factors, that might prove interesting.

```{r}
t_cf_clinical_gp_up <- simple_gprofiler(
  t_clinical_cf_sig_sva_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinical_cure_up_gp-v{ver}.xlsx"))
t_cf_clinical_gp_up

go_termsim <- enrichplot::pairwise_termsim(t_cf_clinical_gp_up[["BP_enrich"]])
t_cf_clinical_gp_go_up_tree <- sm(enrichplot::treeplot(go_termsim,
                                                       label_format = wrap_width))
pp(file = "images/overrepresentation/t_cf_clinical_gp_up_tree.pdf",
   width = treeplot_width, height = treeplot_height)
t_cf_clinical_gp_go_up_tree
dev.off()
t_cf_clinical_gp_go_up_tree
enrichplot::dotplot(t_cf_clinical_gp_up[["BP_enrich"]])

reactome_termsim <- enrichplot::pairwise_termsim(t_cf_clinical_gp_up[["REAC_enrich"]])
t_cf_clinical_gp_reac_up_tree <- sm(enrichplot::treeplot(reactome_termsim,
                                                         label_format = wrap_width))
pp(file = "images/overrepresentation/t_cf_clinical_gp_up_tree.pdf",
   width = treeplot_width, height = treeplot_height)
t_cf_clinical_gp_reac_up_tree
dev.off()
t_cf_clinical_gp_reac_up_tree
pp(file = "images/overrepresentation/t_cf_clinical_gp_up_dot.pdf",
   width = treeplot_width, height = treeplot_height)
enrichplot::dotplot(t_cf_clinical_gp_up[["REAC_enrich"]])
dev.off()
enrichplot::dotplot(t_cf_clinical_gp_up[["REAC_enrich"]])

tf_termsim <- enrichplot::pairwise_termsim(t_cf_clinical_gp_up[["TF_enrich"]])
## The treeplot fails for this for some reason?
##t_cf_clinical_gp_tf_up_tree <- sm(enrichplot::treeplot(tf_termsim))
enrichplot::dotplot(t_cf_clinical_gp_up[["TF_enrich"]])
enrichplot::dotplot(t_cf_clinical_gp_up[["WP_enrich"]])
```

No biopsies!

```{r}
t_cf_clinicalnb_gp_up <- simple_gprofiler(
  t_clinicalnb_cf_sig_sva_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinicalnb_cure_up_gp-v{ver}.xlsx"))
t_cf_clinicalnb_gp_up

go_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_up[["BP_enrich"]])
t_cf_clinicalnb_gp_go_up_tree <- sm(enrichplot::treeplot(go_termsim,
                                                         label_format = wrap_width))
pp(file = "figures/t_cf_clinicalnb_gp_up_tree.svg",
   width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_go_up_tree
dev.off()
t_cf_clinicalnb_gp_go_up_tree
enrichplot::dotplot(t_cf_clinicalnb_gp_up[["BP_enrich"]])

reactome_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_up[["REAC_enrich"]])
t_cf_clinicalnb_gp_reac_up_tree <- sm(enrichplot::treeplot(reactome_termsim,
                                                           label_format = wrap_width))
pp(file = "images/overrepresentation/t_cf_clinicalnb_gp_up_tree.pdf",
   width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_reac_up_tree
dev.off()
t_cf_clinicalnb_gp_reac_up_tree
pp(file = "images/overrepresentation/t_cf_clinicalnb_gp_up_dot.pdf",
   width = treeplot_width, height = treeplot_height)
enrichplot::dotplot(t_cf_clinicalnb_gp_up[["REAC_enrich"]])
dev.off()
enrichplot::dotplot(t_cf_clinicalnb_gp_up[["REAC_enrich"]])

tf_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_up[["TF_enrich"]])
## The treeplot fails for this for some reason?
##t_cf_clinicalnb_gp_tf_up_tree <- sm(enrichplot::treeplot(tf_termsim))
enrichplot::dotplot(t_cf_clinicalnb_gp_up[["TF_enrich"]])
enrichplot::dotplot(t_cf_clinicalnb_gp_up[["WP_enrich"]])
```

#### Genes higher in fail

The only significant results for this group appear to be GO.

```{r}
t_cf_clinical_gp_down <- simple_gprofiler(
  t_clinical_cf_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinical_fail_up_gp-v{ver}.xlsx"))
t_cf_clinical_gp_down

go_termsim <- enrichplot::pairwise_termsim(t_cf_clinical_gp_down[["BP_enrich"]])
t_cf_clinical_gp_go_down_tree <- sm(enrichplot::treeplot(go_termsim))
pp(file = "images/overrepresentation/t_cf_clinical_gp_down_tree.pdf",
   width = treeplot_width, height = treeplot_height)
t_cf_clinical_gp_go_down_tree
dev.off()
```

No biopsies!


```{r}
t_cf_clinicalnb_gp_down <- simple_gprofiler(
  t_clinicalnb_cf_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinicalnb_fail_up_gp-v{ver}.xlsx"))
t_cf_clinicalnb_gp_down

go_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_down[["BP_enrich"]])
t_cf_clinicalnb_gp_go_down_tree <- sm(enrichplot::treeplot(go_termsim))
pp(file = "images/overrepresentation/t_cf_clinicalnb_gp_down_tree.pdf",
   width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_go_down_tree
dev.off()
t_cf_clinicalnb_gp_go_down_tree

## And both, this is not usually where I put this, but the
## clinical samples are a bit of a special case.
t_cf_clinicalnb_gp_both <- simple_gprofiler(
  t_clinicalnb_cf_sig_sva_both,
  excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinicalnb_fail_up_gp-v{ver}.xlsx"))
t_cf_clinicalnb_gp_both

go_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_both[["BP_enrich"]])
t_cf_clinicalnb_gp_go_both_tree <- sm(enrichplot::treeplot(go_termsim))
pp(file = "images/overrepresentation/t_cf_clinicalnb_gp_both_tree.pdf",
   width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_go_both_tree
dev.off()
t_cf_clinicalnb_gp_go_both_tree

reac_termsim <- enrichplot::pairwise_termsim(t_cf_clinicalnb_gp_both[["REAC_enrich"]])
t_cf_clinicalnb_gp_reac_both_tree <- sm(enrichplot::treeplot(reac_termsim))
pp(file = "images/overrepresentation/t_cf_clinicalnb_gp_both_reac_tree.pdf",
   width = treeplot_width, height = treeplot_height)
t_cf_clinicalnb_gp_reac_both_tree
dev.off()
t_cf_clinicalnb_gp_reac_both_tree
```

### clusterProfiler

The following essentially repeats the gProfiler2 invocation using
clusterProfiler.  I have a couple of functions which compare the GO
results from various methods, perhaps I should dig it out and see how
similar the results are using these two tools.  My assumption is that
the primary differences should arise from the fact that gProfiler
theoretically is updating their GO data over time and cProfiler uses
the information in org.Hs.eg.db, which afaik has not changed in quite
a while.

#### Genes higher in cure

```{r}
t_cf_clinical_cp_up <- simple_clusterprofiler(
  t_clinical_cf_sig_sva_up, de_table = t_clinical_cf_table_sva,
  excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinical_cure_up_cp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_clinical_cp_up[["enrich_objects"]][["BP_all"]])
```

#### Genes higher in fail

```{r}
t_cf_clinical_cp_down <- simple_clusterprofiler(
  t_clinical_cf_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Overrepresentation/clinical_fail_up_cp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_clinical_cp_down[["enrich_objects"]][["BP_all"]])
```

#### GSEA

GSEA is not associated with either up nor down, it takes the rank
order of genes with respect (in this case) to fold-change.

```{r}
t_cf_clinical_topn_gsea <- plot_topn_gsea(t_cf_clinical_cp_up)
t_cf_clinical_topn_gsea[["GO"]][[1]]
t_cf_clinical_topn_gsea[["GO"]][[2]]
t_cf_clinical_topn_gsea[["GO"]][[3]]
t_cf_clinical_topn_gsea[["GO"]][[4]]
t_cf_clinical_topn_gsea[["GO"]][[5]]
```

## Visit 1 vs. other visits

The relevant xlsx files are:

* "{xlsx_prefix}/DE_Visits/tv1_vs_later_tables-v{ver}.xlsx"
* "{xlsx_prefix}/DE_Visits/tv1_vs_later_sig-v{ver}.xlsx"

```{r}
input_xlsx <- glue("{xlsx_prefix}/DE_Visits/tv1_vs_later_tables-v{ver}.xlsx")
tv1_vs_later_table <- table_reader(input_xlsx, "later_vs_first")

input_xlsx <- glue("{xlsx_prefix}/DE_Visits/tv1_vs_later_sig-v{ver}.xlsx")
tv1_vs_later_up_sig <- sig_reader(input_xlsx, "later_vs_first")
tv1_vs_later_down_sig <- sig_reader(input_xlsx, "later_vs_first", "down")
```

### Increased in visit 1

I am not likely to do the decreased in visit 1, there are only 7 genes.

#### enrichment: gProfiler

```{r}
tv1later_up_gp <- simple_gprofiler(
  tv1_vs_later_up_sig,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/tv1_vs_later_up_gp-v{ver}.xlsx"))
tv1later_up_gp

enrichplot::dotplot(tv1later_up_gp[["BP_enrich"]])
enrichplot::dotplot(tv1later_up_gp[["REAC_enrich"]])
```

#### enrichment: clusterProfiler

Note to self: make some classes for plot_topn_gsea so it can handle
the various inputs it is likely to receive.

```{r}
tv1later_up_cp <- simple_clusterprofiler(
  tv1_vs_later_up_sig, de_table = tv1_vs_later_table,
  orgdb = "org.Hs.eg.db", kegg_prefix = "hs", do_kegg = TRUE)
enrichplot::dotplot(tv1later_up_cp[["enrich_objects"]][["MF_all"]])
enrichplot::dotplot(tv1later_up_cp[["enrich_objects"]][["BP_all"]])

written <- write_cp_data(
  tv1later_up_cp,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/tv1_vs_later_up_cp-v{ver}.xlsx"))
```

#### GSEA: clusterProfiler

```{r}
tv1later_topn_gsea <- plot_topn_gsea(tv1later_up_cp)
tv1later_topn_gsea[["GO"]][[1]]
tv1later_topn_gsea[["GO"]][[2]]
```

## Patient Sex

The relevant input xlsx files are:

* {xlsx_prefix}/DE_Sex/t_sex_table-v{ver}.xlsx"))
* {xlsx_prefix}/DE_Sex/t_sex_sig-v{ver}.xlsx"))

Oh, I messed up and put the DE results in the GSEA directory!

Let us see if we observe general male/female differences in the data.
This has an important caveat: there are few female failures in the
dataset and so the results may reflect that.

```{r}
input_xlsx <- glue("{xlsx_prefix}/DE_Sex/t_sex_cure_table-v{ver}.xlsx")
t_sex_table <- table_reader(input_xlsx, "male_vs_female")

input_xlsx <- glue("{xlsx_prefix}/DE_Sex/t_sex_cure_sig-v{ver}.xlsx")
t_sex_up_sig <- sig_reader(input_xlsx, "male_vs_female")
t_sex_down_sig <- sig_reader(input_xlsx, "male_vs_female", "down")
```

### gProfiler

#### Increased in men

```{r}
t_sex_up_gp <- simple_gprofiler(
  t_sex_up_sig,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/sex_male_up_gp-v{ver}.xlsx"))
t_sex_up_gp

enrichplot::dotplot(t_sex_up_gp[["BP_enrich"]])
enrichplot::dotplot(t_sex_up_gp[["REAC_enrich"]])
enrichplot::dotplot(t_sex_up_gp[["TF_enrich"]])
```

#### Increased in women

```{r}
t_sex_down_gp <- simple_gprofiler(
  t_sex_down_sig,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/sex_female_up_gp-v{ver}.xlsx"))
t_sex_down_gp
```

Given the results from gProfiler, I do not have high expectations for
clusterProfiler, so I am not likely to take the time to write them out.
It seems like male/female is confounded with cure/fail.

### clusterProfiler

#### Increased in men

```{r}
t_sex_up_cp <- simple_clusterprofiler(
  t_sex_up_sig, de_table = t_sex_table,
  orgdb = "org.Hs.eg.db", kegg_prefix = "hs", do_kegg = TRUE)
enrichplot::dotplot(t_sex_up_cp[["enrich_objects"]][["BP_all"]])
```

#### Increased in women

```{r}
t_sex_down_cp <- simple_clusterprofiler(
  t_sex_down_sig, de_table = t_sex_table,
  orgdb = "org.Hs.eg.db", kegg_prefix = "hs", do_kegg = TRUE)
enrichplot::dotplot(t_sex_down_cp[["enrich_objects"]][["BP_all"]])
```

#### GSEA

```{r}
t_sex_topn_gsea <- plot_topn_gsea(t_sex_up_cp)
t_sex_topn_gsea[["GO"]][[1]]
```

## Ethnicity

Once again, the relevant files to load:

* "{xlsx_prefix}/DE_Ethnicity/t_ethnicity_table-v{ver}.xlsx"
* "{xlsx_prefix}/DE_Ethnicity/t_ethnicity_sig-v{ver}.xlsx"

```{r}
input_xlsx <- glue("{xlsx_prefix}/DE_Ethnicity/t_ethnicity_table-v{ver}.xlsx")
t_ethnicity_mestizo_indigenous <- table_reader(input_xlsx, "mestizo_indigenous")
t_ethnicity_mestizo_afrocol <- table_reader(input_xlsx, "mestizo_afrocol")
t_ethnicity_indigenous_afrocol <- table_reader(input_xlsx, "indigenous_afrocol")

input_xlsx <- glue("{xlsx_prefix}/DE_Ethnicity/t_ethnicity_sig-v{ver}.xlsx")
t_ethnicity_mestizo_indigenous_up <- sig_reader(input_xlsx, "mestizo_indigenous")
t_ethnicity_mestizo_indigenous_down <- sig_reader(input_xlsx, "mestizo_indigenous", "down")
t_ethnicity_mestizo_afrocol_up <- sig_reader(input_xlsx, "mestizo_afrocol")
t_ethnicity_mestizo_afrocol_down <- sig_reader(input_xlsx, "mestizo_afrocol", "down")
t_ethnicity_indigenous_afrocol_up <- sig_reader(input_xlsx, "indigenous_afrocol")
t_ethnicity_indigenous_afrocol_down <- sig_reader(input_xlsx, "indigenous_afrocol", "down")
```

### gProfiler

#### Increased in mestizo vs indigenous

```{r}
mestizo_indigenous_up_gp <- simple_gprofiler(
  t_ethnicity_mestizo_indigenous_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_mi_up_gp-v{ver}.xlsx"))
mestizo_indigenous_up_gp
```

#### Increased in indigenous vs mestizo

```{r}
indigenous_mestizo_up_gp <- simple_gprofiler(
  t_ethnicity_mestizo_indigenous_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_im_up_gp-v{ver}.xlsx"))
indigenous_mestizo_up_gp
```

#### Increased in mestizo vs afrocolombian

```{r}
mestizo_afrocol_up_gp <- simple_gprofiler(
  t_ethnicity_mestizo_afrocol_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_ma_up_gp-v{ver}.xlsx"))
mestizo_afrocol_up_gp
```

#### Increased in afrocolombian vs mestizo

```{r}
afrocol_mestizo_up_gp <- simple_gprofiler(
  t_ethnicity_mestizo_afrocol_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_am_up_gp-v{ver}.xlsx"))
afrocol_mestizo_up_gp
```

#### Increased in indigenous vs afrocolombian

```{r}
indigenous_afrocol_up_gp <- simple_gprofiler(
  t_ethnicity_indigenous_afrocol_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_ia_up_gp-v{ver}.xlsx"))
indigenous_afrocol_up_gp
```

#### Increased in afrocolombian vs indigenous

```{r}
afrocol_indigenous_up_gp <- simple_gprofiler(
  t_ethnicity_indigenous_afrocol_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/ethnicity_ai_up_gp-v{ver}.xlsx"))
afrocol_indigenous_up_gp
```

### clusterProfiler

#### Increased in mestizo vs indigenous

```{r}
mestizo_indigenous_up_cp <- simple_clusterprofiler(
  t_ethnicity_mestizo_indigenous_up,
  de_table = t_ethnicity_mestizo_indigenous)
mestizo_indigenous_up_cp
```

#### Increased in indigenous vs mestizo

```{r}
indigenous_mestizo_up_cp <- simple_clusterprofiler(
  t_ethnicity_mestizo_indigenous_down)
indigenous_mestizo_up_cp
```

#### Increased in mestizo vs afrocolombian

```{r}
mestizo_afrocol_up_cp <- simple_clusterprofiler(
  t_ethnicity_mestizo_afrocol_up,
  de_table = t_ethnicity_mestizo_afrocol)
mestizo_afrocol_up_cp
```

#### Increased in afrocolombian vs mestizo

```{r}
afrocol_mestizo_up_cp <- simple_clusterprofiler(t_ethnicity_mestizo_afrocol_down)
afrocol_mestizo_up_cp
```

#### Increased in indigenous vs afrocolombian

```{r}
indigenous_afrocol_up_cp <- simple_clusterprofiler(
  t_ethnicity_indigenous_afrocol_up,
  de_table = t_ethnicity_indigenous_afrocol)
indigenous_afrocol_up_cp
```

#### Increased in afrocolombian vs indigenous

```{r}
afrocol_indigenous_up_cp <- simple_clusterprofiler(t_ethnicity_indigenous_afrocol_down)
afrocol_indigenous_up_cp
```

## Visit 1 cure/fail

It looks like there are very few groups in the visit 1 significant
genes.

The relevant xlsx files are:

* "{cf_prefix}/Visits/t_clinical_v1_cf_table_sva-v{ver}.xlsx"
* "{cf_prefix}/Visits/t_clinical_v1_cf_sig_sva-v{ver}.xlsx"

```{r}
input_xlsx <- glue("{cf_prefix}/Visits/t_clinical_v1_cf_table_sva-v{ver}.xlsx")
t_cf_clinical_v1_table_sva <- table_reader(input_xlsx)

input_xlsx <- glue("{cf_prefix}/Visits/t_clinical_v1_cf_sig_sva-v{ver}.xlsx")
t_cf_clinical_v1_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_clinical_v1_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
```

### gProfiler

#### Increased in visit 1 cure/fail

```{r}
t_cf_clinical_v1_sig_sva_up_gp <- simple_gprofiler(t_cf_clinical_v1_sig_sva_up)
t_cf_clinical_v1_sig_sva_up_gp
```

#### Increased in visit 1 fail/cure

```{r}
t_cf_clinical_v1_sig_sva_down_gp <- simple_gprofiler(t_cf_clinical_v1_sig_sva_down)
t_cf_clinical_v1_sig_sva_down_gp
enrichplot::dotplot(t_cf_clinical_v1_sig_sva_down_gp[["BP_enrich"]])
```

### clusterProfiler

#### Increased visit 1 cure/fail

```{r}
t_cf_clinical_v1_sig_sva_up_cp <- simple_cprofiler(
  t_cf_clinical_v1_sig_sva_up,
  de_table = t_cf_clinical_v1_table_sva,
  orgdb = "org.Hs.eg.db")
t_cf_clinical_v1_sig_sva_up_cp

enrichplot::dotplot(t_cf_clinical_v1_sig_sva_up_cp[["enrich_objects"]][["BP_all"]])
```

#### Increased in visit 1 cure/fail

```{r}
t_cf_clinical_v1_sig_sva_down_cp <- simple_cprofiler(
  t_cf_clinical_v1_sig_sva_down,
  orgdb = "org.Hs.eg.db")
```

#### GSEA

It appears there are too few results to perform the gsea plots.

## Cure/Fail, Biopsies

The relevant xlsx output may be found at:

* "{cf_prefix}/Biopsies/t_biopsy_cf_table_sva-v{ver}.xlsx"
* "{cf_prefix}/Biopsies/t_cf_biopsy_sig_sva-v{ver}.xlsx"

```{r}
input_xlsx <- glue("{cf_prefix}/Biopsies/t_biopsy_cf_table_sva-v{ver}.xlsx")
t_cf_biopsy_table_sva <- table_reader(input_xlsx, "outcome")

input_xlsx <- glue("{cf_prefix}/Biopsies/t_cf_biopsy_sig_sva-v{ver}.xlsx")
t_cf_biopsy_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_biopsy_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_biopsy_sig_sva_both <- rbind.data.frame(t_cf_biopsy_sig_sva_up,
                                             t_cf_biopsy_sig_sva_down)
```

### gProfiler

We only have 17 genes in the biopsies, but perhaps they are still interesting?

#### increased in cure vs fail

```{r}
t_cf_biopsy_sig_sva_gp_up <- simple_gprofiler(
  t_cf_biopsy_sig_sva_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_biopsy_sig_sva_up_gp-v{ver}.xlsx"))
t_cf_biopsy_sig_sva_gp_up

enrichplot::dotplot(t_cf_biopsy_sig_sva_gp_up[["BP_enrich"]])
go_termsim <- enrichplot::pairwise_termsim(t_cf_biopsy_sig_sva_gp_up[["BP_enrich"]])
go_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))
pp(file = glue("images/overrepresentation/t_cf_biopsy_up_gp_go-v{ver}.pdf"),
   height = treeplot_height, width = treeplot_width)
go_treeplot
dev.off()
go_treeplot
```

#### increased in fail vs cure

```{r}
t_cf_biopsy_sig_sva_gp_down <- simple_gprofiler(
  t_cf_biopsy_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_biopsy_sig_sva_down_gp-v{ver}.xlsx"))
t_cf_biopsy_sig_sva_gp_down

## Not nearly as interesting
```

### clusterprofiler

Again, clusterprofiler version

#### increased in cure/fail

```{r}
t_cf_biopsy_sig_sva_cp_up <- simple_cprofiler(
  t_cf_biopsy_sig_sva_up, de_table = t_cf_biopsy_table_sva,
  orgdb = "org.Hs.eg.db",
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_biopsy_sig_sva_up_cp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_biopsy_sig_sva_cp_up[["enrich_objects"]][["BP_all"]])
```

#### GSEA of cure/fail

```{r}
t_cf_biopsy_sig_topn_gsea <- plot_topn_gsea(t_cf_biopsy_sig_sva_cp_up)
t_cf_biopsy_sig_topn_gsea[["GO_outcome_up"]][[1]]
```

## Eosinophils cure/fail

```{r}
input_xlsx <- glue("{cf_prefix}/Eosinophils/t_eosinophil_cf_table_sva-v{ver}.xlsx")
t_cf_eosinophil_table_sva <- table_reader(input_xlsx, "outcome")

input_xlsx <- glue("{cf_prefix}/Eosinophils/t_eosinophil_cf_sig_sva-v{ver}.xlsx")
t_cf_eosinophil_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_eosinophil_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_eosinophil_sig_sva_both <- rbind.data.frame(t_cf_eosinophil_sig_sva_up,
                                                 t_cf_eosinophil_sig_sva_down)
```

### gProfiler

#### increased in cure vs fail

```{r}
t_cf_eosinophil_sig_sva_up_gp <- simple_gprofiler(
  t_cf_eosinophil_sig_sva_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_up_gp-v{ver}.xlsx"))
t_cf_eosinophil_sig_sva_up_gp
enrichplot::dotplot(t_cf_eosinophil_sig_sva_up_gp[["BP_enrich"]])
enrichplot::dotplot(t_cf_eosinophil_sig_sva_up_gp[["TF_enrich"]])

go_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_up_gp[["BP_enrich"]])
go_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))
pp(file = "figures/t_cf_eosinophil_up_gp_go.svg",
   height = treeplot_height, width = treeplot_width)
go_treeplot
dev.off()
go_treeplot

reac_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_up_gp[["REAC_enrich"]])
reac_treeplot <- sm(treeplot(reac_termsim, label_format = wrap_width))
pp(file = glue("images/overrepresentation/t_cf_eosinophil_up_gp_reac-v{ver}.pdf"),
   height = treeplot_height, width = treeplot_width)
reac_treeplot
dev.off()
```

#### increased in fail vs cure

```{r}
t_cf_eosinophil_sig_sva_down_gp <- simple_gprofiler(
  t_cf_eosinophil_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_down_gp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_eosinophil_sig_sva_down_gp[["BP_enrich"]])

go_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_down_gp[["BP_enrich"]])
go_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))
pp(file = glue("images/overrepresentation/t_cf_eosinophil_down_gp_go-v{ver}.pdf"),
   height = treeplot_height, width = treeplot_width)
go_treeplot
dev.off()
go_treeplot

## There is only one reactome hit, so not plotting it.
```

#### Both up and down

I evaluated this and the 'up' set next to each other and they are
extremely similar:

up: 148 GO hits, 68 TF, 0 HPA
both: 169 GO, 69 TF, and 2 HPA

Otherwise I think they are the same.

```{r}
t_cf_eosinophil_sig_sva_both_gp <- simple_gprofiler(
  t_cf_eosinophil_sig_sva_both,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_both_gp-v{ver}.xlsx"))
t_cf_eosinophil_sig_sva_both_gp
enrichplot::dotplot(t_cf_eosinophil_sig_sva_both_gp[["BP_enrich"]])

go_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_both_gp[["BP_enrich"]])
go_treeplot <- sm(treeplot(go_termsim, label_formap = wrap_width))
pp(file = glue("images/overrepresentation/t_cf_eosinophil_both_gp_go-v{ver}.pdf"),
   height = treeplot_height, width = treeplot_width)
go_treeplot
dev.off()
go_treeplot

reac_termsim <- enrichplot::pairwise_termsim(t_cf_eosinophil_sig_sva_both_gp[["REAC_enrich"]])
reac_treeplot <- sm(treeplot(reac_termsim, label_format = wrap_width))
pp(file = glue("images/overrepresentation/t_cf_eosinophil_both_gp_reac-v{ver}.pdf"),
   height = treeplot_height, width = treeplot_width)
reac_treeplot
dev.off()
```

### clusterprofiler

#### increased in cure vs fail

```{r}
t_cf_eosinophil_sig_sva_cp_up <- simple_cprofiler(
  t_cf_eosinophil_sig_sva_up, de_table = t_cf_eosinophil_table_sva,
  orgdb = "org.Hs.eg.db",
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_sig_sva_up_cp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_eosinophil_sig_sva_cp_up[["enrich_objects"]][["BP_all"]])
```

#### increased in fail vs cure

```{r}
t_cf_eosinophil_sig_sva_cp_down <- simple_cprofiler(
  t_cf_eosinophil_sig_sva_down,
  orgdb = "org.Hs.eg.db",
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_sig_sva_down_cp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_eosinophil_sig_sva_cp_down[["enrich_objects"]][["BP_all"]])
```

#### GSEA

```{r}
t_cf_eosinophil_sig_topn_gsea <- plot_topn_gsea(t_cf_eosinophil_sig_sva_cp_up)
t_cf_eosinophil_sig_topn_gsea[["GO_outcome_up"]][[1]]
```

## Monocytes cure/fail

```{r}
input_xlsx <- glue("{cf_prefix}/Monocytes/t_monocyte_cf_table_sva-v{ver}.xlsx")
t_cf_monocyte_table_sva <- table_reader(input_xlsx, "outcome")

input_xlsx <- glue("{cf_prefix}/Monocytes/t_monocyte_cf_sig_sva-v{ver}.xlsx")
t_cf_monocyte_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_monocyte_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_monocyte_sig_sva_both <- rbind.data.frame(t_cf_monocyte_sig_sva_up,
                                               t_cf_monocyte_sig_sva_down)
```

### gProfiler

Now that I am looking back over these results, I am not compeltely
certain why I only did the gprofiler search for the sva data...

#### increased in cure vs fail

```{r}
t_cf_monocyte_sig_sva_up_gp <- simple_gprofiler(
  t_cf_monocyte_sig_sva_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_up_gp-v{ver}.xlsx"))
t_cf_monocyte_sig_sva_up_gp
enrichplot::dotplot(t_cf_monocyte_sig_sva_up_gp[["BP_enrich"]])
enrichplot::dotplot(t_cf_monocyte_sig_sva_up_gp[["TF_enrich"]])

go_termsim <- enrichplot::pairwise_termsim(t_cf_monocyte_sig_sva_up_gp[["BP_enrich"]])
go_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))
pp(file = "figures/overrepresentation/t_cf_monocyte_up_gp_go.svg",
   height = treeplot_height, width = treeplot_width)
go_treeplot
dev.off()
go_treeplot
```

#### increased in fail vs cure

```{r}
t_cf_monocyte_sig_sva_down_gp <- simple_gprofiler(
  t_cf_monocyte_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_down_gp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_monocyte_sig_sva_down_gp[["BP_enrich"]])

go_termsim <- enrichplot::pairwise_termsim(t_cf_monocyte_sig_sva_down_gp[["BP_enrich"]])
go_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))
pp(file = glue("images/overrepresentation/t_cf_monocyte_down_gp_go-v{ver}.pdf"),
   height = treeplot_height, width = treeplot_width)
go_treeplot
dev.off()
go_treeplot

## Insufficient results to make a tree plot.
```

#### Both up and down

```{r}
t_cf_monocyte_sig_sva_both_gp <- simple_gprofiler(
  t_cf_monocyte_sig_sva_both,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_both_gp-v{ver}.xlsx"))
t_cf_monocyte_sig_sva_both_gp
enrichplot::dotplot(t_cf_monocyte_sig_sva_both_gp[["BP_enrich"]])

go_termsim <- enrichplot::pairwise_termsim(t_cf_monocyte_sig_sva_both_gp[["BP_enrich"]])
go_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))
pp(file = glue("images/overrepresentation/t_cf_monocyte_both_gp_go-v{ver}.pdf"),
   height = treeplot_height, width = treeplot_width)
go_treeplot
dev.off()
go_treeplot

## Insufficient results for reactome

tf_termsim <- enrichplot::pairwise_termsim(t_cf_monocyte_sig_sva_both_gp[["TF_enrich"]])
tf_treeplot <- sm(treeplot(tf_termsim, label_format = wrap_width))
pp(file = glue("images/overrepresentation/t_cf_monocyte_both_gp_tf-v{ver}.pdf"),
   height = treeplot_height, width = treeplot_width)
tf_treeplot
dev.off()
tf_treeplot
```

### clusterprofiler

#### increased in cure vs fail

```{r}
t_cf_monocyte_sig_sva_cp_up <- simple_cprofiler(
  t_cf_monocyte_sig_sva_up, de_table = t_cf_monocyte_table_sva,
  orgdb = "org.Hs.eg.db",
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_sig_sva_up_cp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_monocyte_sig_sva_cp_up[["enrich_objects"]][["BP_all"]])
```

#### increased in fail vs cure

```{r}
t_cf_monocyte_sig_sva_cp_down <- simple_cprofiler(
  t_cf_monocyte_sig_sva_down,
  orgdb = "org.Hs.eg.db",
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_monocyte_sig_sva_down_cp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_monocyte_sig_sva_cp_down[["enrich_objects"]][["BP_all"]])
```

#### GSEA

```{r}
t_cf_monocyte_sig_topn_gsea <- plot_topn_gsea(t_cf_monocyte_sig_sva_cp_up)
t_cf_monocyte_sig_topn_gsea[["GO_outcome_up"]][[1]]
```

## Neutrophils cure/fail

```{r}
input_xlsx <- glue("{cf_prefix}/Neutrophils/t_neutrophil_cf_table_sva-v{ver}.xlsx")
t_cf_neutrophil_table_sva <- table_reader(input_xlsx, "outcome")

input_xlsx <- glue("{cf_prefix}/Neutrophils/t_neutrophil_cf_sig_sva-v{ver}.xlsx")
t_cf_neutrophil_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_neutrophil_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_neutrophil_sig_sva_both <- rbind.data.frame(t_cf_neutrophil_sig_sva_up,
                                               t_cf_neutrophil_sig_sva_down)
```

### gProfiler

Now that I am looking back over these results, I am not compeltely
certain why I only did the gprofiler search for the sva data...

#### increased in cure vs fail

```{r}
t_cf_neutrophil_sig_sva_up_gp <- simple_gprofiler(
  t_cf_neutrophil_sig_sva_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_up_gp-v{ver}.xlsx"))
t_cf_neutrophil_sig_sva_up_gp
enrichplot::dotplot(t_cf_neutrophil_sig_sva_up_gp[["BP_enrich"]])
enrichplot::dotplot(t_cf_neutrophil_sig_sva_up_gp[["TF_enrich"]])

go_termsim <- enrichplot::pairwise_termsim(t_cf_neutrophil_sig_sva_up_gp[["BP_enrich"]])
go_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))
pp(file = "figures/t_cf_neutrophil_up_gp_go.svg",
   height = treeplot_height, width = treeplot_width)
go_treeplot
dev.off()
go_treeplot
```

#### increased in fail vs cure

```{r}
t_cf_neutrophil_sig_sva_down_gp <- simple_gprofiler(
  t_cf_neutrophil_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))
t_cf_neutrophil_sig_sva_down_gp

## Not much to work with here.
```

#### Both up and down

```{r}
t_cf_neutrophil_sig_sva_both_gp <- simple_gprofiler(
  t_cf_neutrophil_sig_sva_both,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_both_gp-v{ver}.xlsx"))
t_cf_neutrophil_sig_sva_both_gp
enrichplot::dotplot(t_cf_neutrophil_sig_sva_both_gp[["BP_enrich"]])
enrichplot::dotplot(t_cf_neutrophil_sig_sva_both_gp[["TF_enrich"]])

go_termsim <- enrichplot::pairwise_termsim(t_cf_neutrophil_sig_sva_both_gp[["BP_enrich"]])
go_treeplot <- sm(treeplot(go_termsim, label_format = wrap_width))
pp(file = glue("images/overrepresentation/t_cf_neutrophil_both_gp_go-v{ver}.pdf"),
   height = treeplot_height, width = treeplot_width)
go_treeplot
dev.off()
go_treeplot

## Insufficient results for reactome
```

### clusterprofiler

#### increased in cure vs fail

```{r}
t_cf_neutrophil_sig_sva_cp_up <- simple_cprofiler(
  t_cf_neutrophil_sig_sva_up, de_table = t_cf_neutrophil_table_sva,
  orgdb = "org.Hs.eg.db",
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_sig_sva_up_cp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_neutrophil_sig_sva_cp_up[["enrich_objects"]][["BP_all"]])
```

#### increased in fail vs cure

```{r}
t_cf_neutrophil_sig_sva_cp_down <- simple_cprofiler(
  t_cf_neutrophil_sig_sva_down,
  orgdb = "org.Hs.eg.db",
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_sig_sva_down_cp-v{ver}.xlsx"))
enrichplot::dotplot(t_cf_neutrophil_sig_sva_cp_down[["enrich_objects"]][["BP_all"]])
```

#### GSEA

```{r}
t_cf_neutrophil_sig_topn_gsea <- plot_topn_gsea(t_cf_neutrophil_sig_sva_cp_up)
t_cf_neutrophil_sig_topn_gsea[["GO_outcome_up"]][[1]]
```

## Look at visit 1 and see if anything pops out

```{r}
input_xlsx <- glue("{cf_prefix}/Monocytes/t_monocyte_v1_cf_table_sva-v{ver}.xlsx")
t_cf_monocyte_v1_table_sva <- table_reader(input_xlsx, "outcome")

input_xlsx <- glue("{cf_prefix}/Monocytes/t_monocyte_v1_cf_sig_sva-v{ver}.xlsx")
t_cf_monocyte_v1_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_monocyte_v1_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_monocyte_v1_sig_sva_both <- rbind.data.frame(t_cf_monocyte_v1_sig_sva_up,
                                                  t_cf_monocyte_v1_sig_sva_down)
```

### Gene Set Enrichment: gProfiler Monocytes by visit, V1

V1: Up: 14 genes; No categories
V1: Down: 52 genes; 20 GO, 5 TF

```{r}
t_cf_monocyte_v1_sig_sva_up_gp <- simple_gprofiler(
  t_cf_monocyte_v1_sig_sva_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_up_gp-v{ver}.xlsx"))
t_cf_monocyte_v1_sig_sva_up_gp
t_cf_monocyte_v1_sig_sva_down_gp <- simple_gprofiler(
  t_cf_monocyte_v1_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))
t_cf_monocyte_v1_sig_sva_down_gp
t_cf_monocyte_v1_sig_sva_both_gp <- simple_gprofiler(
  t_cf_monocyte_v1_sig_sva_both,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))
t_cf_monocyte_v1_sig_sva_both_gp
```

I like cats!

#### clusterProfiler

```{r}
t_cf_monocyte_v1_sig_sva_up_cp <- simple_cprofiler(t_cf_monocyte_v1_sig_sva_up,
                                                   t_cf_monocyte_v1_table_sva,
                                                   orgdb = "org.Hs.eg.db")
enrichplot::dotplot(t_cf_monocyte_v1_sig_sva_up_cp[["enrich_objects"]][["BP_all"]])
t_cf_monocyte_v1_topn_gsea <- plot_topn_gsea(t_cf_monocyte_v1_sig_sva_up_cp)
t_cf_monocyte_v1_topn_gsea[["GO_outcome_up"]][[1]]
```

### Neutrophils v1

```{r}
input_xlsx <- glue("{cf_prefix}/Neutrophils/t_neutrophil_v1_cf_table_sva-v{ver}.xlsx")
t_cf_neutrophil_v1_table_sva <- table_reader(input_xlsx, "outcome")

input_xlsx <- glue("{cf_prefix}/Neutrophils/t_neutrophil_v1_cf_sig_sva-v{ver}.xlsx")
t_cf_neutrophil_v1_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_neutrophil_v1_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_neutrophil_v1_sig_sva_both <- rbind.data.frame(t_cf_neutrophil_v1_sig_sva_up,
                                                  t_cf_neutrophil_v1_sig_sva_down)
```

### Gene Set Enrichment: gProfiler Neutrophils by visit, V1

```{r}
t_cf_neutrophil_v1_sig_sva_up_gp <- simple_gprofiler(
  t_cf_neutrophil_v1_sig_sva_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_up_gp-v{ver}.xlsx"))
t_cf_neutrophil_v1_sig_sva_up_gp
t_cf_neutrophil_v1_sig_sva_down_gp <- simple_gprofiler(
  t_cf_neutrophil_v1_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))
t_cf_neutrophil_v1_sig_sva_down_gp
t_cf_neutrophil_v1_sig_sva_both_gp <- simple_gprofiler(
  t_cf_neutrophil_v1_sig_sva_both,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_neutrophil_down_gp-v{ver}.xlsx"))
t_cf_neutrophil_v1_sig_sva_both_gp
```

#### clusterProfiler

```{r}
t_cf_neutrophil_v1_sig_sva_up_cp <- simple_cprofiler(t_cf_neutrophil_v1_sig_sva_up,
                                                     t_cf_neutrophil_v1_table_sva,
                                                     orgdb = "org.Hs.eg.db")
enrichplot::dotplot(t_cf_neutrophil_v1_sig_sva_up_cp[["enrich_objects"]][["BP_all"]])
t_cf_neutrophil_v1_topn_gsea <- plot_topn_gsea(t_cf_neutrophil_v1_sig_sva_up_cp)
t_cf_neutrophil_v1_topn_gsea[["GO_outcome_up"]][[1]]
```

### Eosinophils v1

```{r}
input_xlsx <- glue("{cf_prefix}/Eosinophils/t_eosinophil_v1_cf_table_sva-v{ver}.xlsx")
t_cf_eosinophil_v1_table_sva <- table_reader(input_xlsx, "outcome")

input_xlsx <- glue("{cf_prefix}/Eosinophils/t_eosinophil_v1_cf_sig_sva-v{ver}.xlsx")
t_cf_eosinophil_v1_sig_sva_up <- sig_reader(input_xlsx, "outcome")
t_cf_eosinophil_v1_sig_sva_down <- sig_reader(input_xlsx, "outcome", "down")
t_cf_eosinophil_v1_sig_sva_both <- rbind.data.frame(t_cf_eosinophil_v1_sig_sva_up,
                                                  t_cf_eosinophil_v1_sig_sva_down)
```

### Gene Set Enrichment: gProfiler Eosinophils by visit, V1

```{r}
t_cf_eosinophil_v1_sig_sva_up_gp <- simple_gprofiler(
  t_cf_eosinophil_v1_sig_sva_up,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_up_gp-v{ver}.xlsx"))
t_cf_eosinophil_v1_sig_sva_up_gp
t_cf_eosinophil_v1_sig_sva_down_gp <- simple_gprofiler(
  t_cf_eosinophil_v1_sig_sva_down,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_down_gp-v{ver}.xlsx"))
t_cf_eosinophil_v1_sig_sva_down_gp
t_cf_eosinophil_v1_sig_sva_both_gp <- simple_gprofiler(
  t_cf_eosinophil_v1_sig_sva_both,
  excel = glue("{xlsx_prefix}/Gene_Set_Enrichment/t_cf_eosinophil_down_gp-v{ver}.xlsx"))
t_cf_eosinophil_v1_sig_sva_both_gp
```

#### clusterProfiler

```{r}
t_cf_eosinophil_v1_sig_sva_up_cp <- simple_cprofiler(t_cf_eosinophil_v1_sig_sva_up,
                                                     t_cf_eosinophil_v1_table_sva,
                                                     orgdb = "org.Hs.eg.db")
enrichplot::dotplot(t_cf_eosinophil_v1_sig_sva_up_cp[["enrich_objects"]][["BP_all"]])
t_cf_eosinophil_v1_topn_gsea <- plot_topn_gsea(t_cf_eosinophil_v1_sig_sva_up_cp)
t_cf_eosinophil_v1_topn_gsea[["GO_outcome_up"]][[1]]
```

# Bibliography
