Biostar Beta. Not for public use.
PSMC: fq2psmc conversion problems
0
Entering edit mode
11 months ago

I am in the process of using PSMC to look at my populations demographic history, but I am having issues with the conversion from fastq to psmcfa.

I followed the recommendations as best as I could on the github page (https://github.com/lh3/psmc), though I did my variant call with the entire cohort rather than one individual as they show. After the variant call, I extract the variants for each of my samples and filter for Q>20, and then convert to fastq using 'vcfutils.pl vcf2fq'.

All goes well, and my consensus fastq seems correct, though I have only lost a small portion (less than 1%) of my variants in the filter which seems a bit relaxed, but it's what they suggest. When I use PSMC's fq2psmcfa module though, the resulting file is 0 bytes, and thus completely blank inside. I used the following code;

utils/fq2psmcfa -q20 whole_genome_indivdual1.fq > whole_genome_individual1.psmcfa

Unfortunately, PSMC doesn't throw out any error messages, and I've tried with a smaller file size and unfiltered fastq files, but nothing happens.

Anybody have any ideas what is going wrong? I'm tempted to run the variant call once again just in case there was a mistake somewhere earlier in the pipeline I missed.

ADD COMMENTlink
0
Entering edit mode

PSMC is intended to infer population history from one diploid genome. Have a look at MSMC.

ADD REPLYlink
0
Entering edit mode

Does this mean I'm meant to call variants for each sample individually? I did the variant call altogether, and then extracted each sample from the vcf so each fastq is the whole genome consensus sequence for each individual.

I would use MSMC, but we didn't want to get phasing information since this was only meant to be a very small part of the study.

ADD REPLYlink
0
Entering edit mode

Hello miles! Have you found solution to your utils/fq2psmcfa conversion problem? I'am having the same issue now. Thanks!

ADD REPLYlink
1
Entering edit mode

Hi Miranda, before I pressed on too much more, we decided to phase the genomes we are using, and thus it would make more sense to use MSMC rather than PSMC. I'm in the final stages of getting that working - one more mask to make!.

Good luck.

ADD REPLYlink
0
Entering edit mode

Hi miles, I have the same problems with you. I just want to get some results by PSMC. Would you give some help to solve the problem? Thank you very much.

ADD REPLYlink
0
Entering edit mode

Please read the post you replied to. It explains that we didn't continue with PSMC.

ADD REPLYlink

Login before adding your answer.

Similar Posts
Loading Similar Posts
Powered by the version 2.1