SNP calling in bacterial genome-worflow

[Morgane_AUS]

I'm getting confused as to which workflow I should build up to get a list of SNPs I can be confident with. I have WGS data and a custom built reference genome and I'm using Galaxy main for now.

So far my workflow consists in mapping my reads, removing reads that didn't align, removing duplicates (rmdup_ I don't manage to get Markduplicates work for me). and transforming my SAM file into a BAM file.

from what I read that I should then realign my reads and recalibrate the quality score. I'm not sure why I need to do that and what do I use for that ?

What tool should I use to call SNPs? Seems like pileup is good enough for SNPs on small number of sample but doesn't generate a VCF file that I need for SNPs calling.

what criteria do you use to filter SNPs?

Sorry for this newbie question... and thanks for any advice

[cliffbeall]

The local realignment is carried out to avoid spurious SNPs that can appear near indels. Also samtools mpileup can generate vcf format with the -v option.

As to which program to use, there are at least samtools/bcftools, GATK, and freebayes. It seems to me that most tools are mainly used and optimized on haploid genomes and lowish coverage, probably neither of which apply to you. I know freebayes has a settable ploidy option which might be useful.