the significant results means tested species is a hybrid or just have gene flow?

[CrawlingSponge]

Dear professor,
Great software!
I have an issue about the results about the hybrid.
Here is the situation of my tests. I use there species of A, B, C and D with their species tree is (((A,B),C),D) to test that whether B is a hybrid of A and C, while the results shows that C is a hybrid of A and B with significance (P =0).
My question is that is this results available? or this results just represent C maybe have gene flow with both A and B?

Another question is that, I use the population "VCF" file to transform into sequences, but the sequences is weird with non-ATCG.
My question is how to use the vcf file transfer into sequences? is there any softwares?

Thank you very much and hoping for you reply!

best wish

Liu

[pblischak]

Hi Liu -- thanks for using HyDe!

Depending on how big the estimate of gamma is can be a better indicator of if something is a hybrid or if there is just gene flow occurring. What was the estimated gamma in your case?

For converting VCF files, if the heterozygous sites are recoded into the IUPAC ambiguity codes then HyDe can handle those no problem. If you're looking for another piece of software to use for the conversion then I would suggest using vcf2phylip: https://github.com/edgardomortiz/vcf2phylip.

[CrawlingSponge]

Dear professor,
Thanks very much for your kind reply!
In my case of hyde tests, I got the a gamma about 0.70, so the other one should be 0.30 theoretically. But the candidate hybrid species should not be a hybrid in most of our previously studies, technically the hybrid contains several species, which are very close in phylogeny. Also, from my understanding of the genetic information of the hybrid should be composed by both a half of its parents, which means the parents both provide 50% of their genetic information to the hybrid theoretically, and the percentage should not be quickly changed in a short evolutionary history (in my case, they are diverged since from less than 7 mya).

Thanks again for your methods to transfer vcf to fasta sequences and I will try that software soon. And I also reproduced the vcf by firstly phased and imputation in a standard ways instand of the quick way with samtools/bcftools, and I hope that could be avialable.

In additaion, I find a former issue asked by another user said that gamma represent P1 and hybrid, againt to the P2 and hybrid? Since I did not see the original codes. Hope for your reply!

Really appreciate for your kind help!
Great and convinient software

Best wishses
Yours,
Liu

[pblischak]

Okay gotcha -- sounds like a cool question! One thing to check would be to make sure that the individuals in the input data file are in the same order as they are in the map file. Individuals in the same population also need to be grouped together in both input files. If things are not arranged this way, then the results could be incorrect

[CrawlingSponge]

Dear professor,
Really appreciate for your kind remind! I used the population data to test the hybrid and found the the gamma is very small (less than 5%).
I also filtered the vcf and reperformed the hyde to test, and there is no significant hybrid in the results.

Thanks again for your carefull and kind reply!
Best regards!
yours,
Liu