search

sigven / pcgr

Personal Cancer Genome Reporter (PCGR)
https://sigven.github.io/pcgr
MIT License
251 stars 47 forks source link

Star allele and CPIC guidlines #81

Closed kokyriakidis closed 5 years ago

kokyriakidis commented 5 years ago

Do you have in mind including Star allele pipeline and CPIC guidlines?

sigven commented 5 years ago

Hi, Thanks for your suggestion. Could you include a pointer/URL for what this refers to?

kokyriakidis commented 5 years ago

This is an implementation from PharmGKB that uses CPIC guidelines and suggests dosage changes on drugs that are affected by different alleles by their metabolized genes. This is so cool because of the report it generates!

http://pharmcat.org/

https://ascpt.onlinelibrary.wiley.com/doi/epdf/10.1002/cpt.1568

CPIC guidlines are the “lighthouse” guiding PGx. They are trusted, validated and clinically important information.

https://cpicpgx.org/

There are other implementations (maybe better ones) but do not produce a nice visual output.

https://stargazer.gs.washington.edu/stargazerweb/

So I thought you could work on that because I really like your reports :)

ALSO I tried to see how you implement "Drug Sensitivity" but I did not find some output in the 2 example reports you have. Can you elaborate it a bit more? Because I am really interested in this.

I see some references about "TARGETED_DRUGS" on "Proto-oncogenes subject to copy number amplifications" but there are entries only about the first 2 oncogenes affected. Why is that?

sigven commented 5 years ago

Dear @kokyriakidis,

Thanks a lot for your input, highly appreciated. And sorry for the truly late response. First, regarding the CPIC guidelines. Clearly, this is a relevant and exciting topic, and from looking at the PharmCAT project, that tool looks really promising. Although these guidelines and implementations clearly is at the core of genomic medicine, I am not convinced that an implementation of these guidelines is within the scope of PCGR (at least how it has been designed currently). Keep in mind that PCGR is intended for the analysis and prioritisation of clinically relevant findings of acquired, somatic aberrations in a given tumor. The input for CPIC guidelines constitute germline variants from a given case/patient, which is not currently considered for interpretation by PCGR. The other tool in development, the Cancer Predisposition Sequencing Reporter, CPSR is focusing on the germline, but largely related to the analysis of cancer susceptibility alleles, and not currently dealing with pharmacogenomics aspects.

With respect to targeted drugs in PCGR reports, there are a number of aspects related to this. Without going too much into details, PCGR is primarily integrating knowledge on targeted cancer drugs (and associated drug sensitivity markers) from two sources: targetvalidation.org, and CIViC. The reason that there were only two genes that are targetable is that are are no cancer drugs (at late stage in clinical development) that are targeted towards the other gene amplifications present in the given case. You can browse the different drugs that are targeted towards ERBB2 here (in the context of breast cancer):

https://www.targetvalidation.org/evidence/ENSG00000141736/EFO_0000305?view=sec:known_drug

Hopefully this was somewhat helpful,

thanks again for your interest, closing this for now. Sigve