IMPACT categories HIGH, MODERATE, LOW, MODIFIER - VEP vs snpEff vs ANNOVAR

[robertzeibich]

I am currently comparing the IMPACT categories HIGH, MODERATE, LOW and MODIFIER of VEP, snpEff and ANNOVAR.

perl $ANNOVAR/table_annovar.pl ${SCRATCH}/$FILEBASENAME/$interval.vcf.gz $ANNOVAR/humandb/ \
-buildver hg38 -out $FILEBASENAME_A/$interval -remove -protocol gnomad30_genome,dbnsfp30a,refGene -operation f,f,g -nastring . -vcfinput -polish -thread 2

Then I am using the MutationAssessor_pred output to obtain the following impact categories H=high/M=medium/L=low/N=neutral.

I noticed that InDels are not annotated with an impact category. ANNOVAR also annotates SNVs less frequently with an impact category than VEP or snpEff.

Am I missing something here? Is there a better way to obtain a more comparable impact category of ANNOVAR?

[kaichop]

ANNOVAR uses dbNSFP database to annotate these scores. You can use dbnsfp42a which is a lot more updated than dbnsfp30a. indels are not given predictions; they are generally not in databases such as dbnsfp (because there are way too many possibilities), and even if they are in databases it is better to just treat them as deleterious if being frameshift mutations, and then manually examine if they are common in population and if they impact non-essential genes and if they may be subject to non-sense mediated RNA decay, etc. They are just different from SNPs, and a scoring approach may not work well for them.

On Wed, Oct 19, 2022 at 10:21 PM robertzeibich @.***> wrote:

I am currently comparing the IMPACT categories HIGH, MODERATE, LOW and MODIFIER of VEP, snpEff and ANNOVAR.

perl ${ANNOVAR}/table_annovar.pl ${SCRATCH}/${FILEBASENAME}/${interval}.vcf.gz ${ANNOVAR}/humandb/ -buildver hg38 -out ${FILEBASENAME_A}/${interval} -remove -protocol gnomad30_genome,dbnsfp30a,refGene -operation f,f,g -nastring . -vcfinput -polish -thread 2

Then I am using the MutationAssessor_pred output to obtain the following impact categories H=high/M=medium/L=low/N=neutral.

I noticed that InDels are not annotated with an impact category. ANNOVAR also annotates SNVs less frequently with an impact category than VEP or snpEff.

Am I missing something here? Is there a better way to obtain a more comparable impact category of ANNOVAR?

— Reply to this email directly, view it on GitHub https://github.com/WGLab/doc-ANNOVAR/issues/202, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABNG3OA6NXNGFQMGBWXPO7TWECUBTANCNFSM6AAAAAARJVVMFQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[robertzeibich]

I just updated to dbnsfp42a. Info: For SNVs, I am still getting with VEP and snpEff higher numbers for the impact categories HIGH, MODERATE, LOW, MODIFIER/NEURTRAL and -. I like the InDel approach you are describing. Thanks for your help and response.