What is the best way for bacterial variant calling?

Question

[Bertalan_Takacs]

I am trying to analyze Illumina sequenced bacterial data from known species to find mutations and I am stuck. Which tool I should use? It seems like half of the tools hugely over-represents the number of variants, even with strict parameters (Breseq, Freebayes) and the other half finds barely any variants (Snippy, VarScan) even with very loose parameters. I read several papers about this and tried a lot of software but still I feel like all the tools belong to either category. What can I do?

Answer 1

[Istvan Albert]

In general, the approach is to filter the variant calls yourself, rather than expect the variant caller to find just the right ones.

Freebayes does indeed find a lot of variants but filtering them after will allow you to hone in the ones that are most reliable and credible:

https://thegenomefactory.blogspot.com/2018/10/a-unix-one-liner-to-call-bacterial.html