SNP ID in VCF file with reference genome about discosnp

[pierrepeterlongo]

Hi Stephan

Just to be sure:

1. You computed variants from reads, wherever they come from, with discoSnp++.
2. You mapped the obtained .fasta file on reference genome A with VCF_creator
3. You mapped the same .fasta file on reference genome B, again with VCF_creator.

If this is the case, yes the variant ids are computed during step 1/. They appear in the prefix_[un]coherent.fa files. The downstream computations conserve those ids.

I hope this helps.

Pierre

[stefanerb89]

Thank you.
And for this case:

we have a more than a unique snp, syntax in fasta is higher_path_XXXXX....P1_pos....P2_pos

In VCF file the ID is then XXXXX_1 and XXXXX_2 (if both match)

But if the SNP matches on position 1 in reference genome A and the same SNP matches on position 2 in reference genome B, the ID names are no longer comparable. In both vcf files they are called XXXXX, and not XXXXX_1 and XXXXX_2. Am I correct in assuming?

[pierrepeterlongo]

Hi Stephan.

Sorry for the delay.

I'm not quite sure to understand your question.
After mapping the predicted variants on a reference, one obtains a vcf file that conserves the ids. For instance (this is in large_test directory), one finds this lines in the vcf:

humchr1 3483342 994_1   C       G       .       PASS    Ty=SNP;Rk=0;UL=977;UR=4834;CL=5893;CR=4834;Genome=.;Sd=1        GT:DP:PL:AD:HQ  0|1:43:229,17,409:26,17:0,0
humchr1 3483361 994_2   G       A       .       PASS    Ty=SNP;Rk=0;UL=977;UR=4834;CL=5893;CR=4834;Genome=.;Sd=1        GT:DP:PL:AD:HQ  0|1:43:229,17,409:26,17:0,0

Here we can see 994_1 and 994_2 mapped on the reference genome. If we map reads on another reference genome, one gets a second vcf file, with variant ids conserved.

This helps?