When I try to run the script, I get the following error:

ValueError: not enough values to unpack (expected 2, got 0)

Any suggestions?

Hi again - Well I think your script runs fine but I noticed that there might be one small issue that could potentially inflate counts of the invariant sites. In fact I've tested this with another in-house script (its not quite as fast and I did not write it). Lets assume in Col 1 or any col for that matter there is one or 2 samples with an N at one position and every other genome contain bases at that position. Will your script call this position as a monomorphic site? Because my understanding is even if there are bases at all the other positions, that site is not considered monomorphic.

Based on your example...will this:

ANAT
AART
AAAG
AA-G

Produce this?

AT
AT
AG
AG

Thanks

Hello I have some WGS samples and I ran gubbins following whole genome SNP alignment to remove recombinant sites which produces a phylip file. I used that as input and ran your script as stated in the readme python3 ./ascbias.py -p Gubbins.filtered_polymorphic_sites.phylip -o testout.phy

The output file just returns 0s....specifically 0 0 0 0.

Any clue and why that might be?

Hi,

I was using the ascbias.py script to remove the invariant sites. It seems to not work for me. I first converted my vcf file to phylip format using the script https://raw.githubusercontent.com/edgardomortiz/vcf2phylip/master/vcf2phylip.py, and then using this script to remove invariant sites for raxml phylogeny inference. But the phylip file after processing still contains IUPAC sites, and raxml was conplaining there are invariant sites in the input file that I shouldn't user the model GTR+ASC_LEWIS.

Do you know what could be the reason?

thanks,
Cui