Dear Lukas. I think I need more experience about MBD seq analysis. I’m running MEDIPS on MeDIP parameter ‘False’ each MBD seq and hMedip. Thanks for giving me advice. Jong hwan kim From: Lukas Chavez [mailto:lukas.chavez.mailings@googlemail.com] Sent: Tuesday, January 07, 2014 4:04 AM To: 김종환 Cc: bioconductor@r-project.org Subject: Re: [BioC] [MEDIPS] Error in 1:max_signal_index : argument of length 0 - what does mean? Dear Jong hwan kim, it appears that your hMeDIP data cannot be modeled with the CpG dependent calibration curve and linear regression. Therefore, I recommend to set the MeDIP parameter of the MEDIPS.meth function to FALSE when you are processing hMeDIP data. Please note, you will not receive CpG density normalized rms data when MeDIP=FALSE, but you still get your genome wide read coverage (counts, rpkm), and you will be able to calculate differential coverage. Lukas On Mon, Jan 6, 2014 at 3:26 AM, 김종환 <kkjjhhk@naver.com> wrote: Dear Lukas and Bioconductors, I’m using MEDIPS version 1.12.0 and analyzing MBD seq data and hMeDIP data. Sample is a Mouse stem cell and have same Input data on MBD seq and hMeDIP seq. It was fastq file, paired - end and I’m mapping on mm10, using bwa. And I read about error on previous version, contain unmapped error on bam file, So I use -F 4 options and get removed unmapped read bam files. Same preprocessing on MBD seq and hMeDIP data. Using same script refer manuscript MEDIPS, and can get DMRs on MBD seq, but I cant get hMeDIP seq. I don’t know why these happening. It normally load data and also saturation, seqCoverage analysis doing well. But when I get a `mr.edgeR`, it show errors > mr.edgeR = MEDIPS.meth(MSet1 = CD4_MeDIP, MSet2 = CD8_MeDIP, CSet = CS, ISet1 = MBD_Input, ISet2 = MBD_Input, p.adj = "bonferroni", diff.method = "edgeR", prob.method = "poisson", MeDIP = T, CNV = F, type = "rpkm", minRowSum = 1) Calculating genomic coordinates... Creating Granges object for genome wide windows... Preprocessing MEDIPS SET 1 in MSet1... Calculating calibration curve... Performing linear regression... Error in 1:max_signal_index : argument of length 0 I’m trying to find errors, sort bam files, include bam index files or exclude index files, and mapping again using bwa. Why these happening? Thanks, Jong hwan kim. [[alternative HTML version deleted]] _______________________________________________ Bioconductor mailing list Bioconductor@r-project.org https://stat.ethz.ch/mailman/listinfo/bioconductor Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor [[alternative HTML version deleted]]