CADD is reporting SNV as intergenic when it is CDS

[feltus]

Hi.

1. I ran CADD v1.6 on a VCF file using hg38 as the reference like this:

./CADD-scripts/CADD.sh -c 6 -a -g GRCh38 -o split.M.vcf.cadd.gz split.M.vcf.gz

2. Everything seemed OK, but I looked a critical SNV at position 11253 on chrM and CADD reported it as Intergenic.

M 11253 T C SNV 0 Intergenic INTERGENIC 0 intergenic 0.477 0.04 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA 10000000 10000000 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA 1.92 0.46 1.34 0.57 1.20 0.31 2.54 0.49 2.25 0.70 1.36 0.48 1.96 0.36 1.16 0.35 1.49 0.29 1.77 0.34 2.49 1.10 32624.20 6828.83 342.64 134.35 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA 204 412 0.475676 6.248

3. However, this is clearly a CDS SNV that is associated with a known phenotype. Here is the link to this position in the UCSC browser.

https://genome.ucsc.edu/cgi-bin/hgTracks?db=hg38&lastVirtModeType=default&lastVirtModeExtraState=&virtModeType=default&virtMode=0&nonVirtPosition=&position=chrM%3A11252%2D11254&hgsid=1629186917_hNnPacm0uiOrlOAhCGcWeEHn6xy7

Could CADD be reporting this wrong or did I do something wrong?

Thanks, Alex

[makirc]

Dear Alex, we stopped the support of the mitochondrial genome with CADD v1.4 (https://cadd.bihealth.org/static/ReleaseNotes_CADD_v1.4.pdf). The mitochondrial genome is really a different beast and many annotations that CADD uses do not work for the MT. If you still want to us CADD on mitochondrial variants, you have to go back to v1.3 and the Cambridge Reference Sequence that we include with GRCh37. Sorry, that we do not announce that more clearly on our website. Best, Martin