DepInfeR

DepInfeR infers tumor-specific molecular protein dependencies by integrating ex-vivo drug sensitivity profiles with drug-protein affinity data through regularized multivariate linear regression.

Key Features:

• Data Integration: DepInfeR integrates drug sensitivity profiles of cancer cell lines or primary tumors ex-vivo (X) with drug affinities of a set of proteins (Y) to infer a matrix of molecular protein dependencies of the cancers (ß).
• Deconvolution of Polypharmacological Effects: DepInfeR deconvolutes the protein inhibition effect on the viability phenotype using regularized multivariate linear regression.
• Dependence Coefficient Assignment: DepInfeR assigns a "dependence coefficient" to each protein and each sample.
• Known Dependency Identification: DepInfeR correctly identified known dependencies including EGFR dependence in Her2+ breast cancer cell line, FLT3 dependence in AML tumors with FLT3-ITD mutations, and differential dependencies on the B-cell receptor pathway in two major subtypes of chronic lymphocytic leukemia (CLL).
• Novel Dependency Discovery: DepInfeR uncovered new subgroup-specific dependencies, including a previously unreported dependence of high-risk CLL on Checkpoint kinase 1 (CHEK1).

Scientific Applications:

• Functional Precision Oncology: DepInfeR guides the choice of pharmacological intervention for a specific cancer type, sub-type, or an individual patient.
• Genomic Aberration Interpretation: DepInfeR enables causal and accurate understanding of functional consequences of genomic aberrations in heterogeneous diseases.
• Heterogeneous Cancer Sample Analysis: DepInfeR produces accurate maps of molecular dependencies in heterogeneous cancer sample sets, as demonstrated with 117 CLL samples.

Methodology:

DepInfeR uses regularized multivariate linear regression to deconvolute protein inhibition effects on viability phenotypes from integrated drug sensitivity and drug-protein affinity data matrices.