In Stefanos-Apostle/NixSEA-package:

Installing NixSEA from github

library(devtools)
install_github("Stefanos-Apostle/NixSEA-package")

library(NixSEA)

Dependencies

library(msigdbr)
library(fgsea)
library(ggplot2)
library(gplots)
library(viridis)
library(biomaRt)
library(tidyverse)
library(ggpubr)
library(biomaRt)

Reading in Rank file

v <- read.csv("/secondary/projects/mnp/stefanos/yang/RNAseq/nnat/gig_vs_wt_nopseudo.rnk", sep = "\t")
RNK_yang <- v[,2]
names(RNK_yang) <- v[,1]

ensembl = useMart( "ensembl", dataset = "mmusculus_gene_ensembl" )
genemap <- getBM( attributes = c("ensembl_gene_id", "entrezgene_id", "mgi_symbol"),
                  filters = "mgi_symbol",
                  values = names(RNK_yang),
                  mart = ensembl)

idx <- match( names(RNK_yang), toupper(genemap$mgi_symbol))
entrez_match <- genemap$entrezgene_id[idx]
names(RNK_yang) <- entrez_match

Grabbing all of MsigDB C2 database and formatting gene sets for GSEA of Rank file

msigdb <- msigdbr(species = "Mus musculus", category = "C2")
sets <- keyword_geneset(msigdb, "")
genesets <- geneset_list(sets)

gsea_res <- fgsea(genesets, RNK_yang, 10000)
gsea_res <- gsea_res[order(gsea_res$padj, decreasing = FALSE),]

Distribution of tested gene sets

geneset_dist(gsea_res)

Top significantly enriched gene sets in each condition

top_genesets(gsea_res=gsea_res, top_num = 5)

Analysis of top keywords for significantly enriched genesets

keyword_ann(gsea_res, top_keywords = 30)

Loading in all C2 gene sets with keywords CANCER and SIGNALING as seen in the top 10 keywords above

msigdb <- msigdbr(species = "Mus musculus", category = "C2")
sets <- keyword_geneset(msigdb, "LUNG")
genesets <- geneset_list(sets)

Enrichment plots of significant gene sets for LUNG keyword

figure1 <- enrichment_analysis(genesets, RNK_yang, sets, figure_header = "Significant Hallmark Geneset Enrichment")

Stefanos-Apostle/NixSEA-package documentation built on June 7, 2022, 3:37 a.m.