Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Issues with angsd -sites function #466

Closed
williamamarsh opened this issue Mar 16, 2022 · 4 comments · Fixed by #533
Closed

Issues with angsd -sites function #466

williamamarsh opened this issue Mar 16, 2022 · 4 comments · Fixed by #533

Comments

@williamamarsh
Copy link

williamamarsh commented Mar 16, 2022

Hi,

I have recently updated my workflow to include conda, and as a result have begun to use angsd installed using conda, in this case f0.935.

However, I no longer get the required output using the command below:

angsd -b [xxxx] -gl 2 -domajorminor 3 -sites [xxx] -domaf 1 -minmaf 0.05 -doGlf 3 -out test.snp -nthreads 4

which worked perfectly well previously using exactly the same input, with angsd v0.929.

The issue seems to be in the filtering step, as seen below:

for angsd v.0935

-> Wed Mar 16 16:27:57 2022
	-> Arguments and parameters for all analysis are located in .arg file
	-> Total number of sites analyzed: 641039227
	-> Number of sites retained after filtering: 0 
	[ALL done] cpu-time used =  581.59 sec
	[ALL done] walltime used =  219.00 sec

for angsd v.0929

> Mon Oct  4 17:56:14 2021
	-> Arguments and parameters for all analysis are located in .arg file
	-> Total number of sites analyzed: 2710295196
	-> Number of sites retained after filtering: 414575 
	[ALL done] cpu-time used =  3324.33 sec
	[ALL done] walltime used =  3327.00 sec

Is this a known issue/how best may I go about remedying it?

Best,

William

@wenluo711
Copy link

This is so interesting. I'm having another issue and was trying the latest version. Running 2 different versions of angsd using same syntax same input but the latest version gave me 0 sites after filtering while the previous version makes more sense.
My version upgrade is from 0.935-46-ga3e53f3-dirty to 0.937 though...

@zjnolen
Copy link

zjnolen commented Mar 28, 2022

I can also add another instance to this issue, I've tried using the -sites option when creating an SFS with realSFS in both 0.933 and 0.937, and find that if I use it on 0.933, it properly filters to the site list, but on 0.937 it removes all sites. Interestingly, I've not been having issues with -r or -rf when making my saf files, so I wonder if converting the -sites filter to an -rf filter (should be possible by just converting the \tin the sites file to a :) might work in the versions we are having issues with.

@zjnolen
Copy link

zjnolen commented Apr 8, 2022

After playing a bit with this, it seems to be an issue with the conda install specifically (unless @wenluo711 is seeing it without the conda install). Using 0.937 with htslib 1.15.1 on conda and with an install from source, the -sites option works using the source install, but returns 0 sites in the conda install. Conda versions of ANGSD have seemed to have different issues like this when I've tried them in the past (such as #396), but based on #460 there may be an official conda version coming, which potentially could resolve some of these issues. For now, I've switched to using ANGSD in a singularity container rather than conda in my workflows.

@wenluo711
Copy link

Yes, I did used conda install. I may also try singularity as alternative solution.

Thanks!

isinaltinkaya added a commit that referenced this issue Oct 11, 2022
Add function aio::doAssert to replace asserts
Did not use aio::assert as name  since aio.h
namespace complains due to assert being a macro
Fixes the major bug explained in #527
Fixes issues #520 #474 #466 #420 #405 #396 #385
Possibly others; other issues should rerun the
commands using the latest version.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants