calculating within population nucleotide diversity for a virus population

Question

[vjmorley]

I'm interested in using my next-gen sequencing data to calculate within population nucleotide diversity in RNA virus populations. My samples were sequenced on Illumina Hiseq, and each sample represents a diverse virus population. I have come across several packages for calculating nucleotide diversity, but they all seem to assume that each sample represents one individual rather than a population of individuals. Can anyone recommend software to use? It would be great if I could input a .bam or .mpileup file and get back diversity statistics for the population.

Answer 1

[skbrimer]

Actually, trying to due the same thing. There are a couple of quasispecies reconstruction programs out there. For example there is shoRAH (here), vispa (here), and an older one VICUNA (here). I also believe the freebayes will also call haplotypes although I'm having some trouble with it going from my vcf file to a haplotype list.

[mark.rose]

Did you ever get freebayes to work for calling haplotypes. By the documentation and posts like this , it sure seems like it should be possible but I have yet to discover quite how and all my requests for help, though highly viewed, have gone unanswered.

[skbrimer]

Sadly, no. I tried playing around with the allele frequency since different populations would show up as low quality SNPs and you can do that with the -F flag I believe the default is 0.1 but I was trying as low as 0.005. Maybe I didn't go far enough? I'm sorry I'm not much help.

Also this project got pushed back for us so I haven't been working on it very much. I was going to try emailing Erik and ask for some guidance but never did.

Sorry again and good luck.

Answer 2

[apelin20]

I would recommend PoPoolation. Allows you to calculate Theta, Pi and Tajima's D from NGS per population (.mpileup) and even allows you to calculate differentiation between populations, see which genes are undergoing adaptation.

Answer 3

[vjmorley]

Update: In the end I found SNPGenie to be most useful for calculating Pi from RNA virus data. I tried PoPoolation, but found that it had trouble with the high coverage and large population sizes associated with RNA virus data.