ksd .tsv out

[ilaydagulmez]

Hi, thanks for the great tool!

I have a problem with .tsv file. Tool and command were successfully run and I got the plots. But when I tried the "peak" command, got the error. Then when I looked at my .tsv file there was no numeric counts:

Here is my input file looks like:

dmd command output:

Thanks for your time,

İlayda

[heche-psb]

Hi, thanks for your interest in using wgd v2! The NaN is a result of stripped alignment length as 0 for that family (i.e., after removing all gap-containing columns) because codeml will drop all gap-containing columns. This is often observed for big families, say, families with >200 gene members. If you want to get the Ks values for those families with stripped alignment length as 0 anyhow, you may add the --pairwise option which calculates the Ks in a pairwised fashion by subsetting the corresponding pairwise alignment from the whole familiy alightment, which is usually less gappy and thus Ks-calculatable. Could you please share the log and command information for your usage of wgd peak? I need a bit more details so as to look into it.

[ilaydagulmez]

Hi, thanks for your quick response.

I will try with the --pairwise option immediately, then with this out, try again wgd peak. If there is still an error I will add a log file and other details.

Thank you so much!

[ilaydagulmez]

Hi again! I tried with the command --pairwise but it doesn't work. Here is the fail log:

"IndexError: list index out of range"

Thanks for your time.

[heche-psb]

Hi, which family went error? Could you please share me with the cds and family file of those problematic families, I will try to reproduce your error.

[ilaydagulmez]

Hi, thanks for your quick response. Here are my files:

CDS file: cds.fasta.gz

family: cds.fasta.txt

log: slurm-1684903.out.txt

[heche-psb]

Hi, please first install the v2.0.38 and try again. There might be some bugs in the older versions which I have fixed already in the latest version.

[ilaydagulmez]

Hi, I didn't see the update sorry, thanks! I will try immediately.

[ilaydagulmez]

Hi again,

I used to try the latest version, and dmd succeeded but ksd wasn't with the parameter --pairwise. Here is the log:

When I run the without--pairwise it's completed but as I wrote before, I want to run--peak and --focus so with NaN values, it doesn't work.

The main problem is the time limit now, 3 days will not be enough for --pairwise but our server works max. 3 days.

Thank you for your time!