Format

Send to

Choose Destination
PLoS Comput Biol. 2015 Jan 8;11(1):e1004024. doi: 10.1371/journal.pcbi.1004024. eCollection 2015 Jan.

Analysis of individual protein regions provides novel insights on cancer pharmacogenomics.

Author information

1
Program on Bioinformatics and Systems Biology, Sanford-Burnham Medical Research Institute, La Jolla, California, United States of America.

Abstract

The promise of personalized cancer medicine cannot be fulfilled until we gain better understanding of the connections between the genomic makeup of a patient's tumor and its response to anticancer drugs. Several datasets that include both pharmacologic profiles of cancer cell lines as well as their genomic alterations have been recently developed and extensively analyzed. However, most analyses of these datasets assume that mutations in a gene will have the same consequences regardless of their location. While this assumption might be correct in some cases, such analyses may miss subtler, yet still relevant, effects mediated by mutations in specific protein regions. Here we study such perturbations by separating effects of mutations in different protein functional regions (PFRs), including protein domains and intrinsically disordered regions. Using this approach, we have been able to identify 171 novel associations between mutations in specific PFRs and changes in the activity of 24 drugs that couldn't be recovered by traditional gene-centric analyses. Our results demonstrate how focusing on individual protein regions can provide novel insights into the mechanisms underlying the drug sensitivity of cancer cell lines. Moreover, while these new correlations are identified using only data from cancer cell lines, we have been able to validate some of our predictions using data from actual cancer patients. Our findings highlight how gene-centric experiments (such as systematic knock-out or silencing of individual genes) are missing relevant effects mediated by perturbations of specific protein regions. All the associations described here are available from http://www.cancer3d.org.

PMID:
25568936
PMCID:
PMC4287345
DOI:
10.1371/journal.pcbi.1004024
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Public Library of Science Icon for PubMed Central
Loading ...
Support Center