Tao Liu (τν)

Because you used this options: "-t AndrewCD38LowR1.BedUtils.bed -f BED --format BAMPE" You have a bed file but you asked MACS2 to read it as BAM file. Try to remove "--format...

@bioinfograd Thanks for asking! Yes. You can use the peaks bed/xls generated from `callpeak --SPMR` since the option `--SPMR` only affects the bedGraph output. Regarding your follow-up question on `bdgdiff`,...

Hi @huizhen2014, the algorithm in MACS2 is different from the one described in the 2008 paper. Please refer to the help message from your MACS2 program -- `macs2 callpeak -h`....

@zeus19900814 You can find the compiled package from bioconda for macs2: https://anaconda.org/bioconda/macs2 Aa for macs3, please keep an eye on our own conda channel: https://anaconda.org/macs3/repo We will have precompiled package...

There are many changes over years including fixing bugs related to overflow issues https://github.com/macs3-project/MACS/blob/master/ChangeLog. I am not sure exactly which one causes the difference in your data analysis. Could you...

@sklasfeld Thanks for sharing! I have checked the codes between 2.1.1.20160309 and the most recent 2.2.7 and don't see big difference on peak calling part, and I tested the output...

It is really strange that MACS found the average insertion size over 10k from both treatment and control. Do you mind share with me part of the bam file so...

@AIBio Interesting. Could you provide more detail of the header before and after `reheader`? Also the number of reads mapped to each chromosome before and after `reheader`?

@eregenyi You can check if any of the three reasons are true. 1) the control has a lot of tags in the surrounding area (lambda 1k, 5k, 10k) or in...

@eregenyi Yes. That's the lag between the Watson and Crick tags. As I mentioned before, I would think the prediction of the fragment length as a step for quality control,...