Question regarding data harmonizing

[Hao-Kuen Lin]

Dear cBioPortal,

Hi, I'm new to cBioPortal and I have a general question regarding data harmonizing.

Do you apply a uniform pipeline to generate CNV?

For example, if there are several studies that have WGS, do you use the same algorithm to align reads, generate copy number segments and then use GISTIC 2.0 to generate gene-level CNVs? Or you retrieve the copy number segments provided by each study and generate gene-level CNVs?

Thank you.

Best,

Hao-Kuen

[Benjamin Gross]

Hi Hao-Kuen,

With the exception of expression - z-scores, in almost all cases we do not generate the data, including gene-level CNVs you find in cbioportal.org.  Datasets are provided to us by various centers and universities when they are published or studies are curated from publications by our data curation team.  In a small number of studies, those provided by the GDC, we have converted ASCAT CNV to GISTIC like data. 

Best,

Benjamin

--
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/47f6b963-5eee-4d42-b30e-37999510629dn%40googlegroups.com.

[Hao-Kuen Lin]

Thank you!