Re: a question about mcomp

[Sun, Deqiang]

Hello Egor,

Thanks for your interest in MOABS. I actually just corrected a bug. Please download again. 

For your question, please add -m wt.G.bed -m ko.G.bed in your command line

Note this option: -m [ —mergedRatiosFiles ] arg If —ratiosFiles is ‘,’ separated, then this option must be set;

So try

~/moabs/bin/mcomp -r wt_r1.G.bed,wt_r2.G.bed,wt_r3.G.bed -r ko_r1.G.bed,ko_r2.G.bed,ko_r3.G.bed -m wt.G.bed -m ko.G.bed -c comp.wt.vs.ko.txt

Let me know if you have any other question. By the way, please send your future question to moabs_...@googlegroups.com.

Best,

Deqiang

From: Egor Dolzhenko <dolz...@usc.edu>

Date: Mon, 3 Mar 2014 20:16:04 -0600
To: deqiang sun <deqi...@bcm.edu>
Subject: a question about mcomp

Hello Deqiang,

I am writing to ask a quick question about your DMR method that just got publied in Genome Biology. (By the way, very nice paper!)

I have some count files generated by our methylation analysis pipeline which I am currently trying to run the mcomp on. Here is what I did.

First, I converted the files to the format described on p. 17 of the manual. Here are the first few lines:

#chrom startend ratiototalC methCstrand nextPlus totalCmethC MinustotalC methC

chr1 3000573     3000575 0.483871 3115 +G +31 15- 00

chr1 3000725    30007270.642857 149 +G +14 9- 00

chr1 3000900    30009020.5 63 +G +6 3- 00

As you can see, I am not doing strand based analysis so all of the counts are on the plus strand. I have attached the first 10K lines of each input file.

So, when I run

~/moabs/bin/mcomp -r wt_r1.G.bed,wt_r2.G.bed,wt_r3.G.bed -r ko_r1.G.bed,ko_r2.G.bed,ko_r3.G.bed -c comp.wt.vs.ko.txt

I get this output:

Options are saved in file run.config and printed here:

| compFile=comp.wt.vs.ko.txt

| dmrMethods=7

| doComp=1

| doDmcScan=1

| doDmrScan=1

| doMergeRatioFiles=0

| doStrandSpecifiMeth=0

| filterCredibleDif=-10

| lmFit=1

| maxDistConsDmcs=300

| mergeNotIntersect=1

| minCredibleDif=0.2

| minDepthForComp=3

| minDmcsInDmr=3

| minNominalDif=0.3333

| pFetDmc=0.05

| pFetDmr=0.05

| pSimDmc=1

| pSimDmr=1

| precision=3

| ratiosFiles=wt_r1.G.bed,wt_r2.G.bed,wt_r3.G.bed ko_r1.G.bed,ko_r2.G.bed,ko_r3.G.bed 

| threads=6

| topRankByCDif=1

| topRankByPSim=1

| You request to merge files before comparison so please specify the filename for the merged file

Thus, it looks like mcomp thinks I am trying to merge files whereas I want to run differential methylation analysis.​

By the way, I am running precompiled binaries on a 64 bit xubunty machine. 

Could you please let me know what's wrong? 

Thanks a lot for looking into this stuff!

Egor

--

Egor Dolzhenko

Postdoctoral Research Associate

Department of Computational Biology and Bioinformatics

University of Southern California

[Egor Dolzhenko]

Thanks,​ Deqiang. I've just downloaded and tried running the updated version. For some reason I get this error:

~/moabs/bin/mcomp 

terminate called after throwing an instance of 'std::runtime_error'

  what():  locale::facet::_S_create_c_locale name not valid

Aborted (core dumped)

I get this problem both on CentOS and XUbuntu. I've tried tweaking the locale options, but nothing I tried worked. Do you know what the problem might be?

Thanks again,

Egor

---

From: Sun, Deqiang <deqi...@bcm.edu>
Sent: Tuesday, March 4, 2014 10:31 PM
To: Egor Dolzhenko
Cc: Sun, Deqiang; moabs_...@googlegroups.com
Subject: Re: a question about mcomp

 

[Deqiang Sun]

Hello Egor,

I have compiled against boost version 1.49.0. Could you download again and see it that works? 

Best,

Deqiang

--
You received this message because you are subscribed to the Google Groups "MOABS Discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to moabs_msuite...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Egor Dolzhenko]

Deqiang,

Yup, it works. Thanks!

Egor

---

From: Deqiang Sun <sunnyi...@gmail.com>
Sent: Wednesday, March 5, 2014 2:38 PM
To: Egor Dolzhenko
Cc: moabs_...@googlegroups.com

[Rafet Al]

Hi,

where I could find  version 1.49.0 because I have the same error. Github has  version 1.3.9.6

Thanks