Dear Pertille, I already tried, to consider Mset1=MeDIP_AV and ISet1=MeDIP_C, and a lot of other options... but still not working. The unique way to work is to considering MeDIP=F and CNV=F. Someone could explain me what is the implication of that? CpG density normalisation assumes a linear dependency of CpG density and MeDIP-seq signal in the low range of MeDIP-seq signals. If this is not observed in the data, MEDIPS fails to build a proper linear model. This is typically the case, if there was low enrichment or if the processed samples was an input sample. Sincerely, Lukas On 09 Feb 2016, at 16:34, pertille [bioc] <noreply@bioconductor.org<mailto:noreply@bioconductor.org>> wrote: Activity on a post you are following on support.bioconductor.org<https: support.bioconductor.org=""/> User pertille<https: support.bioconductor.org="" u="" 9104=""/> wrote Question: Error in MEDIPS.calibrationCurve looking for different methylation profiles<https: support.bioconductor.org="" p="" 78066=""/>: I am looking for different methylation profile between two different environmental conditions(control[MeDIP_C] and test(MeDIP_AV) in chickens. I also have a input that is a reduced genome sequencing from the same specie used in this experiment, but without methylation enrichment (I am not sure if I can use as input). When I do my callibration curve, everything works well using one of the animals from the "MeDIP_AV" animals, also works if I use the MeDIPS_C or even the input: > CS = MEDIPS.couplingVector(pattern = "CG", refObj = MeDIP_AV[[1]]) Get genomic sequence pattern positions... Searching in chr1 ...[ 1657074 ] found. Searching in chr2 ...[ 1258402 ] found. Searching in chr3 ...[ 967325 ] found. Searching in chr4 ...[ 821866 ] found. Searching in chr5 ...[ 617892 ] found. Searching in chr6 ...[ 387216 ] found. Searching in chr7 ...[ 404811 ] found. Searching in chr8 ...[ 344382 ] found. Searching in chr9 ...[ 295488 ] found. Searching in chr10 ...[ 273056 ] found. Searching in chr11 ...[ 248162 ] found. Searching in chr12 ...[ 260083 ] found. Searching in chr13 ...[ 254213 ] found. Searching in chr14 ...[ 236666 ] found. Searching in chr15 ...[ 193016 ] found. Searching in chr16 ...[ 11925 ] found. Searching in chr17 ...[ 189366 ] found. Searching in chr18 ...[ 186836 ] found. Searching in chr19 ...[ 177100 ] found. Searching in chr20 ...[ 217166 ] found. Searching in chr21 ...[ 124750 ] found. Searching in chr22 ...[ 65130 ] found. Searching in chr23 ...[ 128632 ] found. Searching in chr24 ...[ 122477 ] found. Searching in chr25 ...[ 65793 ] found. Searching in chr26 ...[ 121968 ] found. Searching in chr27 ...[ 115236 ] found. Searching in chr28 ...[ 109525 ] found. Searching in chrW ...[ 17866 ] found. Searching in chrZ ...[ 815671 ] found. Number of identified CG pattern: 10689093 Calculating genomic coordinates... Creating Granges object for genome wide windows... Counting the number of CG's in each window... When I try to use the mr.edgeR to calculate differencial coverage between two conditions, the follows error appear: > mr.edgeR = MEDIPS.meth(MSet1 = MeDIP_AV, MSet2 = MeDIP_C, CSet = CS, ISet1= MeDIP_allRJFGBS, ISellRJFGBS, p.adj = "bonferroni", diff.method = "edgeR", MeDIP = T, CNV = T, minRowSum = 10) Calculating genomic coordinates... Creating Granges object for genome wide windows... Preprocessing MEDIPS SET 1 in MSet1... Calculating calibration curve... Performing linear regression... Intercept: -0.0242126718892354 Slope: 0.0184639036661373 Calculating relative methylation score... Preprocessing MEDIPS SET 2 in MSet1... Calculating calibration curve... Error in MEDIPS.calibrationCurve(MSet = MSet1[[i]], CSet = CSet, input = F) : The dependency of coverage signals on sequence pattern (e.g. CpG) densities is different than expected. No linear model can be build, please check the calibration plot by providing the MSet object at ISet I already tried, to consider Mset1=MeDIP_AV and ISet1=MeDIP_C, and a lot of other options... but still not working. The unique way to work is to considering MeDIP=F and CNV=F. Someone could explain me what is the implication of that? Using, MEDIP=F, I found no significantly different window for methylation pattern between the two condition: > mr.edgeR.s = MEDIPS.selectSig(results = mr.edgeR, p.value = 0.1, adj = T, ratio = NULL, bg.countNV = F)L, C Total number of windows: 10030369 Number of windows tested for differential methylation: 130782 Remaining number of windows with adjusted p.value<=0.1: 0 Thank you ________________________________ Post tags: medips, R You may reply via email or visit Error in MEDIPS.calibrationCurve looking for different methylation profiles