BIMSBbioinfo / pigx_bsseq

bisulfite sequencing pipeline from fastq to methylation reports
https://bioinformatics.mdc-berlin.de/pigx/
GNU General Public License v3.0
9 stars 4 forks source link

Error in getTabixByChunk #190

Open fafaris39 opened 1 year ago

alexg9010 commented 1 year ago

Hi @fafaris39,

The error indicates that the tiled object is empty, probably since there are not enough reads left after the minimum coverage threshold. It is basically the same reason why also your call to unite was not successful (https://github.com/al2na/methylKit/issues/278).

If you are running a tiling analysis you may lower the coverage threshold to 3 or 5, but you should still check that enough CpGs are covered to continue your analysis.

Best, Alex

fafaris39 @.***> schrieb am Fr., 31. März 2023, 09:58:

I am using the function tileMethylCounts() to get DMRs.

library("methylKit")

file.list1 = list("/test1-TechRep_1-BioRep_1.deduplicated.CX_report.txt", "/test2-TechRep_1-BioRep_2.deduplicated.CX_report.txt", "/ctrl1-TechRep_1-BioRep_3.deduplicated.CX_report.txt", "/ctrl2-TechRep_1-BioRep_2.deduplicated.CX_report.txt")

readingReports1=methRead(file.list, sample.id=list("test1","test2","ctrl1","ctrl2"), assembly="bismark", treatment=c(1,1,0,0), context="CpG", mincov = 10, pipeline= "bismarkCytosineReport", dbtype = "tabix", dbdir = "methylDB" )

meth1=methylKit::unite(readingReports, destrand=FALSE)

meth1

getCoverageStats(readingReports1[[4]],plot=TRUE)

tiles = tileMethylCounts(readingReports1,win.size=1000,step.size=1000,cov.bases =10)

and I'm getting an error message

checking wether tabix file already exists: /home/ffaris/mnt/biseps/test_data/Resultats_Bismark_LG15/methylDB/test1_tiled.txt.bgz tabix file is new. continuing now ...

compressing the file with bgzip... making tabix index... Error in getTabixByChunk(tbxFile, chunk.size = nrow, return.type) : the tabix file seems to be empty. stopping here.

any idea?

— Reply to this email directly, view it on GitHub https://github.com/BIMSBbioinfo/pigx_bsseq/issues/190, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADK7JD62SMRNP55JPIF7K6TW62FBLANCNFSM6AAAAAAWOJU474 . You are receiving this because you are subscribed to this thread.Message ID: @.***>