geneset_shifts: Visualizes geneset "activity" across a number of...
In facilebio/FacileIncubator: Incubator for half-baked ideas

Description Usage Arguments Details Examples

View source: R/geneset_shifts.R

This funciton is used primarily to generate geneshift plots to compare the AppSAA model vs 5XFAD, but why not generalize ...

geneset_shifts(
  x,
  genesets,
  gsea.method = NULL,
  colors = NULL,
  ymin = 0.005,
  ymax = 0.025,
  bw.bg = 0.7,
  bw.gs = 0.9,
  ribbon_wrap_n = 18,
  legend.position = "none",
  columns = c("genesets", "results"),
  xlims = NULL,
  facet_scales = "free",
  trim = 0.02,
  with_stats = c(pval = "pvalue", padj.by.collection = "FDR"),
  ...
)

`x`	A single FacileTtestFseaAnalysisResult or a named list of them.
`genesets`	character vector of geneset names to extract from the GeneSetDb objects from each results.
`colors`	aaaaarrrghhhhhhhhh
`with_stats`	A named character vector that specifies the gsea statistics to print in each of the results. `names()` correspond to the column names from the result table of the individual GSEA result, and the values are the labels you want to use for that statistic. How do you know what columns are available for printing? Look at the column names in the table returned by `tidy(ffsea.resut, name = gsea.method)`. By default, the nominal pvalue and FDR are printed. To disable, set to `NULL`.
`...`	arbitrary number of named ffsea results
`stats`	the name of the gsea result to pull pvalues from

We assume that the ffsea results each have a GeneSetDb that has genesets by the same name in them.

The row and column order are determined by the order in which ffsea results and genesets are provided in the respective variables.

This will eventually go into the FacileAnalysis package

# We'll setup two ffsea (GSEA) results and plot geneset effects from each
efds <- FacileData::exampleFacileDataSet()
gdb.h <- sparrow::getMSigGeneSetDb("H", "human", id.type = "entrez")
dge <- list(
  crc = efds %>%
    FacileData::filter_samples(indication == "CRC") %>%
    FacileAnalysis::flm_def(
      covariate = "sample_type", numer = "tumor", denom = "normal",
      batch = "sex") %>%
    FacileAnalysis::fdge(method = "voom"),
 blca = efds %>%
    FacileData::filter_samples(indication == "BLCA") %>%
    FacileAnalysis::flm_def(
      covariate = "sample_type", numer = "tumor", denom = "normal",
      batch = "sex") %>%
    FacileAnalysis::fdge(method = "voom"))
gsea <- lapply(dge, FacileAnalysis::ffsea, gdb.h, "fgsea")

gs.1 <- geneset_shifts(
  gsea,
  c("beta cells" = "HALLMARK_PANCREAS_BETA_CELLS",
    "angiogenesis" = "HALLMARK_ANGIOGENESIS"),
  gsea.method = "fgsea",
  columns = "genesets",
  with_stats = c(pval = "pvalue", padj = "FDR", NES = "NES"))
gs.1$plot

gs.2 <- geneset_shifts(
  gsea,
  c("beta cells" = "HALLMARK_PANCREAS_BETA_CELLS",
    "angiogenesis" = "HALLMARK_ANGIOGENESIS"),
  gsea.method = "fgsea",
  columns = "results",
  with_stats = c(pval = "pvalue", padj = "FDR", NES = "NES"))
gs.2$plot