Abstract
Drug screening studies typically involve assaying the sensitivity of a range of cancer cell lines across an array of anti-cancer therapeutics. Alongside these sensitivity measurements high dimensional molecular characterizations of the cell lines are typically available, including gene expression, copy number variation and genomic mutations. We propose a sparse multitask regression model which learns discriminative latent characteristics that predict drug sensitivity and are associated with specific molecular features. We use ideas from Bayesian nonparametrics to automatically infer the appropriate number of these latent characteristics. The resulting analysis couples high predictive performance with interpretability since each latent characteristic involves a typically small set of drugs, cell lines and genomic features. Our model uncovers a number of drug-gene sensitivity associations missed by single gene analyses. We functionally validate one such novel association: that increased expression of the cell-cycle regulator C/EBPδ decreases sensitivity to the histone deacetylase (HDAC) inhibitor panobinostat.
Highlights
Several drug screening studies have assayed the sensitivity of a library of cancer cell lines to an array of anti-cancer compounds
We know which drugs have the same molecular target, and which features are related to the same gene, e.g. gene expression, copy number variation (CNV) and mutation status will all typically be assayed for a given gene
We have found active area to be more predictable from molecular profiles than IC50: 10-fold cross-validation on CCLE using group LASSO explains 27.5% of heldout variance in active area scores across drugs, compared to only 14.4% for IC50
Summary
Several drug screening studies have assayed the sensitivity of a library of cancer cell lines to an array of anti-cancer compounds. CCLE includes gene expression microarrays, copy number variation (CNV), and oncogene mutation status assays. These data have the potential to both help understand the key differences between cancers and cancer subtypes that drive resistance to specific drugs, and to one day help choose the appropriate drug (or combination of drugs) for an individual patient, the core idea of precision medicine. The existing analyses of these datasets involve simple per drug regressions, such as elastic net [7] While these methods are able to pick out the strongest signals in the data, they suffer from not taking advantage of known relationships between drugs and between genomic features. We know which drugs have the same molecular target, and which features are related to the same gene, e.g. gene expression, CNV and mutation status will all typically be assayed for a given gene
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
