Entering edit mode
@jananisundaresan-19303
Last seen 5.9 years ago
How can I correct continuous variable (eg: age) in diffbind? I find that in my data age is a covariate that I need to correct for. The column TISSUE is used for AGE in my data. I tried by altering the contrast and then running dba.analyze but I get 0 peaks after blocking AGE.
library(DiffBind)
library(DESeq2)
library(edgeR)
library(GenomicAlignments)
> chip1
21 Samples, 200401 sites in matrix (367587 total):
> dba(chip1, minOverlap=2,
+ sampleSheet="/scratch/CHIP/DATA/NBB/CHIP_NBB.csv",
+ config=data.frame(RunParallel=TRUE, reportInit="chip1", DataType=chip_DATA_GRANGES,
+ AnalysisMethod=chip_DESEQ2, minQCth=15, fragmentSize=125,
+ bCorPlot=TRUE, th=0.05, bUsePval=FALSE),
+ peakCaller="macs")
21 Samples, 200401 sites in matrix (367587 total):
> chip1 <- dba.count(chip1, summits=150)
Re-centering peaks...
> chip1 <- dba.contrast(chip1, categories=DBA_CONDITION,block = DBA_TISSUE)
> chip1 <- dba.analyze(chip1)
1 Contrast:
Group1 Members1 Group2 Members2 Block1Val InBlock1 Block2Val InBlock2 Block3Val InBlock3 Block4Val InBlock4 Block5Val InBlock5
1 CASE 10 CONTROL 11 102 1 73 1 75 2 76 1 77 1
Block6Val InBlock6 Block7Val InBlock7 Block8Val InBlock8 Block9Val InBlock9 Block10Val InBlock10 Block11Val InBlock11 Block12Val
1 78 1 80 1 81 1 84 4 85 1 86 1 88
InBlock12 Block13Val InBlock13 Block14Val InBlock14 Block15Val InBlock15 Block16Val InBlock16 Block17Val InBlock17 DB.DESeq2
1 1 89 1 91 1 93 1 95 1 96 1 1632
DB.DESeq2.block
1 0