Identify Marker Features in Bulk Expression Data — find_markers_in

Identifies informative marker features across groups from bulk gene expression or signature score matrices using Seurat workflows. Performs feature selection, scaling, PCA, clustering, and marker discovery.

Usage

find_markers_in_bulk(
  pdata,
  eset,
  group,
  id_pdata = "ID",
  nfeatures = 2000,
  top_n = 20,
  thresh.use = 0.25,
  only.pos = TRUE,
  min.pct = 0.25,
  npcs = 30
)

Arguments

pdata: Data frame. Sample metadata.
eset: Matrix. Gene expression or signature score matrix.
group: Character. Column name in pdata specifying grouping variable.
id_pdata: Character. Column name for sample IDs. Default is "ID".
nfeatures: Integer. Number of top variable features to select. Default is 2000.
top_n: Integer. Number of top markers to retain per cluster. Default is 20.
thresh.use: Numeric. Threshold for marker selection. Default is 0.25.
only.pos: Logical. Whether to retain only positive markers. Default is TRUE.
min.pct: Numeric. Minimum expression percentage threshold. Default is 0.25.
npcs: Integer. Number of principal components to use. Default is 30.

Value

List with components: `sce` (Seurat object), `markers` (all markers), `top_markers` (top markers per group).

Examples

eset_tme_stad <- load_data("eset_tme_stad")
colnames(eset_tme_stad) <- substring(colnames(eset_tme_stad), 1, 12)
pdata_sig_tme <- load_data("pdata_sig_tme")
# \donttest{
res <- find_markers_in_bulk(
  pdata = pdata_sig_tme, eset = eset_tme_stad,
  group = "TMEcluster"
)
#> Using Seurat v5+ workflow
#> Normalizing layer: counts
#> Normalizing layer: counts
#> Centering and scaling data matrix
#> Finding variable features for layer counts
#> PC_ 1 
#> Positive:  ACTA2, TAGLN, VIM, TNS1, AXL, CNN1, ZEB1, ACTG2, CTGF, TWIST2 
#> 	   TPM1, ZEB2, FOXF1, ROR2, ADAM19, TGFBI, PER1, FSTL3, FAM101B, HSPB1 
#> 	   PXDC1, IGFBP3, COL4A1, PDCD1LG2, B2M, HAVCR2, SH3PXD2A, TWIST1, CD8A, FAP 
#> Negative:  FANCI, POLQ, FANCD2, EME1, XRCC2, EXO1, RAD54L, BRIP1, GEN1, BRCA2 
#> 	   BLM, FANCA, RAD54B, CHEK1, RECQL4, RAD51, MSH2, FANCB, TOPBP1, BRCA1 
#> 	   POLE, CHEK2, FEN1, NEIL3, LIG1, RDM1, TDP1, FANCM, FANCG, RAD18 
#> PC_ 2 
#> Positive:  ATM, REV3L, SHPRH, TP53BP1, ERCC4, RECQL, FAN1, POLK, ERCC6, ATR 
#> 	   RIF1, ZEB2, WRN, POLI, FANCM, MLH3, PDCD1LG2, UVSSA, ADAM19, ERCC8 
#> 	   ATRIP, ZEB1, XPC, RAD52, HELQ, MSH6, POLH, LIG3, CTLA4, FAP 
#> Negative:  H2AFX, APEX1, PRPF19, RAD23A, NUDT1, PCNA, HLA-A, RPA3, APEX2, NTHL1 
#> 	   HLA-C, UBE2N, HLA-B, XRCC6, UBE2A, RAD23B, HSPB1, SHFM1, MPG, XRCC5 
#> 	   SMUG1, XRCC1, UNG, B2M, CETN2, ERCC1, CLDN7, FEN1, MAD2L2, ALKBH2 
#> PC_ 3 
#> Positive:  LAG3, GZMB, PRF1, CXCL10, IFNG, TBX21, CXCL9, PDCD1, TIGIT, PDCD1LG2 
#> 	   CD8A, HAVCR2, CD274, GZMA, CTLA4, TAP1, RAD54L, TAP2, ADAM12, FANCA 
#> 	   RAD51, LIG1, POLD1, CHAF1A, TWIST1, EXO1, HIST1H2BL, ADAM19, FAP, SEMA7A 
#> Negative:  XRCC5, RAD17, LIG4, RAD50, HUS1, NEIL1, GTF2H1, ERCC5, XPA, CLK2 
#> 	   FAN1, UBE2V2, CETN2, MSH3, RAD23B, DDB1, UBE2B, POLK, MMS19, POLM 
#> 	   POLI, MLH3, PMS1, XRCC6, TDP2, ERCC6, MRE11A, TPM1, ERCC3, SPRTN 
#> PC_ 4 
#> Positive:  ERCC2, POLG, MUS81, MDC1, RECQL5, POLD1, LIG3, PNKP, XRCC3, ATRIP 
#> 	   SH3PXD2A, POLM, HIST2H2BF, FANCE, SEMA7A, XAB2, FOXF1, ROR2, RAD9A, CLDN3 
#> 	   TNS1, OGG1, FSTL3, POLL, DDB1, NTHL1, RECQL4, HIST1H2BL, PRPF19, MMS19 
#> Negative:  CCNH, UBE2B, GZMA, B2M, CDK7, IFNG, RRM2B, RAD17, POLB, GZMB 
#> 	   NBN, FANCL, CD274, CXCL10, UBE2N, DUT, CXCL9, TAP2, PMS1, TDP2 
#> 	   MDM2, XRCC4, ERCC5, RB1, GTF2H3, PCNA, UBE2V2, MNAT1, GTF2H5, ERCC8 
#> PC_ 5 
#> Positive:  XPC, POLL, RNF4, XAB2, PARP3, DDB1, POLG, MMS19, DDB2, RPA1 
#> 	   NHEJ1, RAD23A, MPG, ERCC2, PDCD1, XRCC1, XRCC6, DCLRE1B, MDM2, PRF1 
#> 	   TP53, PARP1, POLH, LIG1, OGG1, CD8A, TAP1, TDP1, TP53BP1, POLD1 
#> Negative:  RPA4, CDKN2A, SHFM1, FAP, MSH5, CCNE1, RAD54B, RAD1, TWIST1, MRE11A 
#> 	   ADAM12, CLDN3, HLTF, GTF2H5, E2F3, CETN2, GTF2H4, NUDT1, LOXL2, ZEB1 
#> 	   WNT5A, COL4A1, RPA3, TWIST2, ACTA2, CLK2, XRCC2, UBE2V2, FSTL3, BLM 
#> Calculating cluster IA
#> For a (much!) faster implementation of the Wilcoxon Rank Sum Test,
#> (default method for FindMarkers) please install the presto package
#> --------------------------------------------
#> install.packages('devtools')
#> devtools::install_github('immunogenomics/presto')
#> --------------------------------------------
#> After installation of presto, Seurat will automatically use the more 
#> efficient implementation (no further action necessary).
#> This message will be shown once per session
#> Calculating cluster IE
#> Calculating cluster IS
# Extract top markers per cluster using base R
top_markers <- res$top_markers
# }