Drug sensitivity prediction for cancer cell lines based on pairwise kernels and miRNA profiles

Computational models for drug sensitivity prediction have the potential to revolutionise personalized cancer medicine. Drug sensitivity assays, as well as profiling of cancer cell lines and drugs becomes increasingly available for training such models. Machine learning methods for drug sensitivity prediction must be optimized for: (i) leveraging the wealth of information about both cancer cell lines and drugs, (ii) predictive performance and (iii) interpretability. Multiple methods were proposed for predicting drug sensitivity from cancer cell line features, some in a multi-task fashion. So far, no such model leveraged drug inhibition profiles. Recent neural network-based recommender systems arise as models capable of predicting cancer cell line response to drugs from their biological features with high prediction accuracy. These models, however, require a tailored approach to model interpretability. In this work, we develop a neural network recommender system for kinase inhibitor sensitivity prediction called DEERS. The model utilizes molecular features of the cancer cell lines and kinase inhibition profiles of the drugs. DEERS incorporates two autoencoders to project cell line and drug features into 10-dimensional hidden representations and a feed-forward neural network to combine them into response prediction. We propose a novel model interpretability approach offering the widest possible assessment of the specific genes and biological processes that underlie the action of the drugs on the cell lines. The approach considers also such genes and processes that were not included in the set of modeled features. Our approach outperforms simpler matrix factorization models, achieving R=0.82 correlation between true and predicted response for the unseen cell lines. Using the interpretability analysis, we evaluate correlation of all human genes with each of the hidden cell line dimensions. Subsequently, we identify 67 biological processes associated with these dimensions. Combined with drug response data, these associations point at the processes that drive the cell line sensitivity to particular compounds. Detailed case studies are shown for PHA-793887, XMD14-99 and Dabrafenib. Our framework provides an expressive, multitask neural network model with a custom interpretability approach for inferring underlying biological factors and explaining cancer cell response to drugs.

Download Full-text

Drug Sensitivity Prediction From Cell Line-Based Pharmacogenomics Data: Guidelines for Developing Machine Learning Models

10.1101/2021.04.09.439076 ◽

2021 ◽

Author(s):

Hossein Sharifi-Noghabi ◽

Soheil Jahangiri-Tazehkand ◽

Casey Hon ◽

Petr Smirnov ◽

Anthony Mammoliti ◽

...

Keyword(s):

Machine Learning ◽

Cell Line ◽

Cell Lines ◽

Cancer Cell ◽

Drug Sensitivity ◽

Cancer Cell Lines ◽

Precision Oncology ◽

Learning Models ◽

Drug Sensitivity Prediction ◽

Machine Learning Models

ABSTRACTThe goal of precision oncology is to tailor treatment for patients individually using the genomic profile of their tumors. Pharmacogenomics datasets such as cancer cell lines are among the most valuable resources for drug sensitivity prediction, a crucial task of precision oncology. Machine learning methods have been employed to predict drug sensitivity based on the multiple omics data available for large panels of cancer cell lines. However, there are no comprehensive guidelines on how to properly train and validate such machine learning models for drug sensitivity prediction. In this paper, we introduce a set of guidelines for different aspects of training a predictor using cell line datasets. These guidelines provide extensive analysis of the generalization of drug sensitivity predictors, and challenge many current practices in the community including the choice of training dataset and measure of drug sensitivity. Application of these guidelines in future studies will enable the development of more robust preclinical biomarkers.

Download Full-text

Network-based clustering for drug sensitivity prediction in cancer cell lines

10.1101/764043 ◽

2019 ◽

Author(s):

Maryam Pouryahya ◽

Jung Hun Oh ◽

James C. Mathews ◽

Zehor Belkhatir ◽

Caroline Moosmüller ◽

...

Keyword(s):

Cell Line ◽

Cell Lines ◽

Cancer Cell ◽

Large Scale ◽

Computational Models ◽

Drug Sensitivity ◽

Cancer Cell Lines ◽

Machine Learning Techniques ◽

Learning Techniques ◽

Drug Sensitivity Prediction

AbstractThe study of large-scale pharmacogenomics provides an unprecedented opportunity to develop computational models that can accurately predict large cohorts of cell lines and drugs. In this work, we present a novel method for predicting drug sensitivity in cancer cell lines which considers both cell line genomic features and drug chemical features. Our network-based approach combines the theory of optimal mass transport (OMT) with machine learning techniques. It starts with unsupervised clustering of both cell line and drug data, followed by the prediction of drug sensitivity in the paired cluster of cell lines and drugs. We show that prior clustering of the heterogenous cell lines and structurally diverse drugs significantly improves the accuracy of the prediction. In addition, it facilities the interpretability of the results and identification of molecular biomarkers which are significant for both clustering of the cell lines and predicting the drug response.

Download Full-text

Predicting tumor response to drugs based on gene-expression biomarkers of sensitivity learned from cancer cell lines

BMC Genomics ◽

10.1186/s12864-021-07581-7 ◽

2021 ◽

Vol 22 (1) ◽

Author(s):

Yuanyuan Li ◽

David M. Umbach ◽

Juno M. Krahn ◽

Igor Shats ◽

Xiaoling Li ◽

...

Keyword(s):

Gene Expression ◽

Cell Lines ◽

Cancer Cell ◽

Predictive Models ◽

Drug Sensitivity ◽

Human Cancer ◽

Cancer Cell Lines ◽

The Cancer Genome Atlas ◽

Differential Sensitivity ◽

Tumor Type

Abstract Background Human cancer cell line profiling and drug sensitivity studies provide valuable information about the therapeutic potential of drugs and their possible mechanisms of action. The goal of those studies is to translate the findings from in vitro studies of cancer cell lines into in vivo therapeutic relevance and, eventually, patients’ care. Tremendous progress has been made. Results In this work, we built predictive models for 453 drugs using data on gene expression and drug sensitivity (IC50) from cancer cell lines. We identified many known drug-gene interactions and uncovered several potentially novel drug-gene associations. Importantly, we further applied these predictive models to ~ 17,000 bulk RNA-seq samples from The Cancer Genome Atlas (TCGA) and the Genotype-Tissue Expression (GTEx) database to predict drug sensitivity for both normal and tumor tissues. We created a web site for users to visualize and download our predicted data (https://manticore.niehs.nih.gov/cancerRxTissue). Using trametinib as an example, we showed that our approach can faithfully recapitulate the known tumor specificity of the drug. Conclusions We demonstrated that our approach can predict drugs that 1) are tumor-type specific; 2) elicit higher sensitivity from tumor compared to corresponding normal tissue; 3) elicit differential sensitivity across breast cancer subtypes. If validated, our prediction could have relevance for preclinical drug testing and in phase I clinical design.

Download Full-text

A pan-cancer analysis of progression mechanisms and drug sensitivity in cancer cell lines

Molecular Omics ◽

10.1039/c9mo00119k ◽

2019 ◽

Vol 15 (6) ◽

pp. 399-405 ◽

Cited By ~ 1

Author(s):

Julia L. Fleck ◽

Ana B. Pavel ◽

Christos G. Cassandras

Keyword(s):

Cell Lines ◽

Cancer Cell ◽

Late Stage ◽

Drug Sensitivity ◽

Cancer Cell Lines ◽

Resistance Mechanisms ◽

General Effect ◽

Tumor Types ◽

Pan Cancer ◽

Genetic Events

Sequences of genetic events were identified that may help explain common patterns of oncogenesis across 22 tumor types. The general effect of late-stage mutations on drug sensitivity and resistance mechanisms in cancer cell lines was evaluated.

Download Full-text

Widespread Molecular Patterns Associated with Drug Sensitivity in Breast Cancer Cell Lines, with Implications for Human Tumors

PLoS ONE ◽

10.1371/journal.pone.0071158 ◽

2013 ◽

Vol 8 (12) ◽

pp. e71158 ◽

Cited By ~ 4

Author(s):

Chad J. Creighton

Keyword(s):

Breast Cancer ◽

Cell Lines ◽

Cancer Cell ◽

Breast Cancer Cell ◽

Drug Sensitivity ◽

Cancer Cell Lines ◽

Human Tumors ◽

Breast Cancer Cell Lines ◽

Molecular Patterns

Download Full-text

MicroRNAs related to intrinsic resistance to oxaliplatin and the irinotecan metabolite SN-38 in 10 colorectal cancer cell lines.

Journal of Clinical Oncology ◽

10.1200/jco.2012.30.4_suppl.524 ◽

2012 ◽

Vol 30 (4_suppl) ◽

pp. 524-524 ◽

Cited By ~ 2

Author(s):

Niels Frank Jensen ◽

Rolf Soekilde ◽

Jan Stenvang ◽

Birgitte Sander Nielsen ◽

Thomas Litman ◽

...

Keyword(s):

Colorectal Cancer ◽

Cell Lines ◽

Cancer Cell ◽

Drug Sensitivity ◽

Predictive Biomarkers ◽

Cancer Cell Lines ◽

Colorectal Cancer Cell ◽

Intrinsic Resistance ◽

Short Term ◽

Colorectal Cancer Cell Lines

524 Background: Chemotherapy of metastatic colorectal cancer is based on 5-flourouracil combined with either oxaliplatin or irinotecan (active metabolite: SN-38). Identification of predictive biomarkers of drug response is needed to provide a better personalized treatment. In this study we aimed to identify microRNAs related to intrinsic resistance to oxaliplatin or irinotecan in a panel of ten colorectal cancer cell lines. Methods: Drug sensitivity towards oxaliplatin and SN-38 was determined for ten colorectal cancer cell lines (Colo-205, DLD-1, HCC-2998, HCT-15, HCT-116, HT-29, KM12, LoVo, LS-174T, and SW620), using the cell viability MTT assay and the cell death LDH assay. In addition, two cell lines (DLD-1 and LoVo) were exposed to the drugs for 6, 24 or 48 hours. MicroRNA expression profiles were generated using the Exiqon miRCURY LNA microarray platform (including 840 microRNAs), and four differentially expressed microRNAs were validated by independent qRT-PCR measurements. Results: The drug sensitivities of the ten colorectal cancer cell lines varied about 50 times between the least and most sensitive cell lines. Correlation of drug sensitivity data to microRNA expression data across the ten cell lines yielded about 25 microRNA biomarker candidates, for each of the drug/assay combinations. Following short-term drug treatment 10-20 microRNAs were altered for each drug/cell line combination. Validation by qRT-PCR showed a very good correlation to the microarray data. MicroRNAs identified by correlation to drug sensitivity and by short-term treatment were compared, and less than 10% were identified by both approaches, perhaps representing the most promising candidates. These candidates are for SN-38 miR-15a, miR-22, miR-24, miR-98, miR-142-3p, miR-1290, and let-7b, and for oxaliplatin miR-23b, miR-27a, miR-192, miR-200a, miR-222, miR-886-5p, and miR-1308. Conclusions: In the present study we identified a number of microRNAs that are potentially involved in intrinsic resistance and/or could be predictive biomarkers for either irinotecan or oxaliplatin.

Download Full-text