Lollipop plot showing a sorter protein that the mutations reported for the gene

[Julio Fernández Banet]

Hi,

I have a question regarding the lollipop plots. While randomly looking at some mutations I noticed that in a few cases the reported amino acid mutation falls outside of the "protein length" reported in the lollipop.

My initial guess is that the isoform used for the annotation and the one used for the lollipop are different.

Could this be the case? Can you provide some information on what is the criteria used to select the isoform used in the lollipop? If it is not based on the isoform with the longest CDS, what is the criteria? Is this something that the user could potentially modify to select which isoform to display in the lollipop.

Thanks a lot for the great work with the portal.

Best regards,

Julio

[Priti Kumari]

Hi Julio,

Thanks for writing. We use the canonical UniProt transcript in cBioPortal - you can see the list of transcript we use here: https://github.com/mskcc/vcf2maf/blob/master/data/isoform_overrides_uniprot

We currently do not have the ability to change the isoform to display in cBioPortal.

Can you point me to the study/query where you see the mutation fall outside the protein length?

Thank you,

Priti

On 11/1/18 6:48 PM, Julio Fernández Banet wrote:

        External Email - Use Caution        

--
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 post to this group, send email to cbiop...@googlegroups.com.
Visit this group at https://groups.google.com/group/cbioportal.
To view this discussion on the web visit https://groups.google.com/d/msgid/cbioportal/CA%2BvU8_07qZwa5wG6XSmSBtyk_J2X98cVAJqgTS6oBK9tB2gphA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[Ino de Bruijn]

Hi Julio,

Just to add to this. We do offer selection of isoform of interest in the mutation mapper:

http://www.cbioportal.org/mutation_mapper.jsp

We're working on enabling this feature on mutations tab as well

Best wishes,

Ino

To view this discussion on the web visit https://groups.google.com/d/msgid/cbioportal/14c36800-c2b5-81a4-c3c8-db6d091823b9%40jimmy.harvard.edu.

[Julio Fernández Banet]

Hi Ino,

Thanks a lot for the pointer to mutation mapper. That works percfectly and I was able to select an isoform that displays all the mutations.

Looking forward to have this feature implemented in the mutation tab.

Best regards,

Julio