Machine Learning Detects Pan-cancer Ras Pathway Activation in The Cancer Genome Atlas

Abstract

SUMMARY Precision oncology uses genomic evidence to match patients with treatment but often fails to identify all patients who may respond. The transcriptome of these “hidden responders” may reveal responsive molecular states. We describe and evaluate a machine-learning approach to classify aberrant pathway activity in tumors, which may aid in hidden responder identification. The algorithm integrates RNA-seq, copy number, and mutations from 33 different cancer types across The Cancer Genome Atlas (TCGA) PanCanAtlas project to predict aberrant molecular states in tumors. Applied to the Ras pathway, the method detects Ras activation across cancer types and identifies phenocopying variants. The model, trained on human tumors, can predict response to MEK inhibitors in wild-type Ras cell lines. We also present data that suggest that multiple hits in the Ras pathway confer increased Ras activity. The transcriptome is underused in precision oncology and, combined with machine learning, can aid in the identification of hidden responders.

Sequential compression of gene expression across dimensionalities and methods reveals no single best method or dimensionality

“…The classifiers were controlled for mutation 725 burden. More details about the specific implementation are described in Way et al 2018 [67]. 726…”

OncoNetExplainer: Explainable Predictions of Cancer Types Based on Gene Expression Data

“…In particular, GE profiles about 33 prevalent tumor type for 9,074 samples are used in our approach. This dataset has been used widely as prior knowledge to generate tumor-specific biomarkers [18][19][20]. These data are hybridized by the Affymetrix 6.0 , which allows us to examine the largest number of cases along with the highest probe density [21].…”

Multi-view Factorization AutoEncoder with Network Constraints for Multi-omic Integrative Analysis

“…Most the work had been focused on comprehensive molecular characterization of individual cancer types Hutter & Zenklusen (2018); Shen et al (2018), which mainly employed statistical analysis of molecular features associated with clinical outcomes. Machine learning approaches also have been applied to study individual -omic data types Malta et al (2018) and integrate multi-omic data Way et al (2018); Angione et al (2016). These approaches mainly employ traditional machine learning techniques, for example, logistic regression Malta et al (2018), random forest Way et al (2018), and similarity network fusion Angione et al (2016).…”

“…Machine learning approaches also have been applied to study individual -omic data types Malta et al (2018) and integrate multi-omic data Way et al (2018); Angione et al (2016). These approaches mainly employ traditional machine learning techniques, for example, logistic regression Malta et al (2018), random forest Way et al (2018), and similarity network fusion Angione et al (2016).…”