MAF File Validation Warning - Variant_Classification Mapping Issues

25 views
Skip to first unread message

Ji Hen Lau

unread,
Jun 10, 2025, 7:48:12 AM6/10/25
to cBioPortal for Cancer Genomics Discussion Group
Hi,
 
I'm currently validating my MAF file for upload and have encountered a warning regarding the Variant_Classification column. The warning states:

"Given value for Variant_Classification column is not one of the expected values. This can result in mapping issues and subsequent missing features in the mutation view UI, such as missing COSMIC information."

Current Situation:
My MAF file contains variant classifications that appear to differ from the expected controlled vocabulary. Based on the documentation, I see the expected values include terms like:
  • Frame_Shift_Del, Frame_Shift_Ins, In_Frame_Del, In_Frame_Ins
  • Missense_Mutation, Nonsense_Mutation, Silent, Splice_Site
  • Translation_Start_Site, Nonstop_Mutation
  • UTR and flanking region terms
  • etc.
However, my data includes terms like:
  • Start_Codon_Ins, Start_Codon_Del, Start_Codon_SNP, Stop_Codon_Del
  • lincRNA
My Questions:
  1. Should I remap my variant classifications to match the expected controlled vocabulary exactly?
  2. Is there a comprehensive mapping guide available for converting common variant annotation terms to cBioPortal's expected format?
  3. How critical is this warning? Will it significantly impact the functionality of my data in cBioPortal, particularly regarding mutation visualization and COSMIC annotations as stated?
Any guidance on best practices for handling this would be greatly appreciated.
Thank you for your time and support.

Best regards,
Ji Hen

Prasanna Jagannathan

unread,
Jun 10, 2025, 11:30:02 AM6/10/25
to Ji Hen Lau, cBioPortal for Cancer Genomics Discussion Group
Hi Ji Hen,

To see these variants in cBioPortal, add the below line to the Mutation data Meta file:

variant_classification_filter:

By leaving this field empty, all variants will be imported into the cBioPortal database and all variants will be visible on cBioPortal.

The full instructions are available in the link below:
https://docs.cbioportal.org/file-formats/#mutation-data

Please reply only to the <cbiop...@googlegroups.com> email.

thanks
Jag


--
You received this message because you are subscribed to the Google Groups "cBioPortal for Cancer Genomics Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cbioportal+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cbioportal/17e38948-8ac6-4150-a0ad-eca798884e43n%40googlegroups.com.

Ji Hen Lau

unread,
Jun 10, 2025, 9:28:56 PM6/10/25
to Prasanna Jagannathan, cBioPortal for Cancer Genomics Discussion Group
Hi Jag!

Yes, I already left the field empty as mentioned in the docs! But the issue I face is this warning:

"Given value for Variant_Classification column is not one of the expected values. This can result in mapping issues and subsequent missing features in the mutation view UI, such as missing COSMIC information."

And the value in my data that raises the warning is Start_Codon_Ins, Start_Codon_Del, Start_Codon_SNP, Stop_Codon_Del, lincRNA, which I believe is not listed in the docs.

So my question is:
  1. Should I remap my variant classifications to match the expected controlled vocabulary exactly?
  2. Is there a comprehensive mapping guide available for converting common variant annotation terms to cBioPortal's expected format?
  3. How critical is this warning? Will it significantly impact the functionality of my data in cBioPortal, particularly regarding mutation visualization and COSMIC annotations as stated?
Regards,
Ji Hen

Prasanna Jagannathan

unread,
Jun 11, 2025, 11:46:13 AM6/11/25
to Ji Hen Lau, cBioPortal for Cancer Genomics Discussion Group
Hi Ji Hen

The easiest way is to remap the variant classifications to match the expected controlled vocabulary exactly. The tool vcf2maf is used as standard to match the cBioPortal expected format.


Please reply only to the <cbiop...@googlegroups.com> email.

thanks
Jag

Ji Hen Lau

unread,
Jun 11, 2025, 8:27:10 PM6/11/25
to cBioPortal for Cancer Genomics Discussion Group
Thanks for the clarification !
Reply all
Reply to author
Forward
0 new messages