SAMtools variant calling : running time and parallelization

[Kerkyra]

Hi,

I'm trying to call variants using samtools mpileup and bcftools on a mammal genome (genome size 3GB, 30X coverage), does anyone have an idea of how long this could take ?

The server here has 128GB RAM and 32 3GHz cores (although only one of them will be used by samtools).
SAMtools prints out the number of genomic sites processed, so I did an estimation based on the first few hours. It seems that it should take ~40 days, which seems awfully long ... And I can't seem to find a way to parallelize the process. Does it really take that long for a genome this size (e.g. human genome) ? If not, what could I be doing wrong? is there any way to parallelize this step ?

If someone has any experience on this, please let me know.

Thanks,

Leila

[aaronh]

Split the genome into 128 intervals and run each one separately, merging the VCF files after.