Vardict cannot correctly count the reads in the normal sample

[shenzhongjun]

Hi,

Thank you for providing such a good software, its sensitivity fully meets the actual needs.

My question is, thereis a deletion chr12:40875389:TGGACCATCAGCTGGAGTGACAGTGACATCC:T which called in vcf is marked "StrongSomatic" and "DP=1544;VD=1544;AF=1", with normal VD and AF is 0.

But in IGV, the actual situation is that there are a lot of reads in the normal sample, and the mutation frequency is close to 1.

Please help me check out what needs to be improved here，thank you.

Related files have been added to the compressed package MUC19_vaf1.zip
The Vardict-java is v1.8.3
The tumor bam is MUC19_vaf1.bam
The normal bam is MUC19_vaf1_dz.bam
The test shell is MUC19_vaf1.sh
The bed file is MUC19_vaf1.bed