SNP calling vs emitting variants

[Rubal7]

Hello all,

I am using GATK unified genotyper to call snps from multiple low coverage samples. I have lowered the -stand_call_conf threshold to 4 as recommended under these circumstances. However should I also be lowering -stand_emit_conf to the same level? My uncertainty arises because I am not clear of the difference between calling snps and emitting snps. Could someone kindly clarify this for me or point me in the right direction?

Many thanks,

Rubal7

[jchoo]

the same question, need help too~~

[biomichael]

Hello all,

I am using GATK unified genotyper to call snps from multiple low coverage samples. I have lowered the -stand_call_conf threshold to 4 as recommended under these circumstances. However should I also be lowering -stand_emit_conf to the same level? My uncertainty arises because I am not clear of the difference between calling snps and emitting snps. Could someone kindly clarify this for me or point me in the right direction?

Many thanks,

Rubal7

I also encontered the same problem. Have you solved the problem?

[jchoo]

not yet now, maybe the forum of Broad may help……

[Rubal7]

No, not solved it yet

[biomichael]

I got some information in the following site: http://www.broadinstitute.org/gsa/ga...tand_call_conf

In the 1000 genomes project for pilot 2 (deep coverage of ~35x) we expect the raw Qscore > 50 variants to contain at least ~10% FP calls. We use extensive post-calling filters to eliminate most of these FPs. Variant Quality Score Recalibration is a tool to perform this filtering.

-stand_call_conf [50.0]
-stand_emit_conf 10.0

So, I guess the -stand_emit_conf parameter is to lower the FPs(false positive?) got by the -stand_call_conf parameter. if -stand_call_conf=50 and -stand_emit_conf=10, the variants between 10 and 50 will be emitted but also marked as filtered.

I don't know whether I understand rightly.

[ymc]

I got some information in the following site: http://www.broadinstitute.org/gsa/ga...tand_call_conf

In the 1000 genomes project for pilot 2 (deep coverage of ~35x) we expect the raw Qscore > 50 variants to contain at least ~10% FP calls. We use extensive post-calling filters to eliminate most of these FPs. Variant Quality Score Recalibration is a tool to perform this filtering.

-stand_call_conf [50.0]
-stand_emit_conf 10.0

So, I guess the -stand_emit_conf parameter is to lower the FPs(false positive?) got by the -stand_call_conf parameter. if -stand_call_conf=50 and -stand_emit_conf=10, the variants between 10 and 50 will be emitted but also marked as filtered.

I don't know whether I understand rightly.

Based on my experimentation, decreasing the -stand_emit_conf argument will simply increase number of LowQual calls but have virtually none effect on the PASS calls. So I don't think FP rate will change unless you also count LowQual calls as positives.

[blueskypy]

-stand_emit_conf 10.0 means that it won’t report any potential SNPs with a quality below 10.0; but unless they meet the quality threshold set by -stand_call_conf (50.0, in this case), they will be listed as failing the quality filter