Identifying subpathway signatures for individualized anticancer drug response by integrating multi-omics data

Abstract

Background Individualized drug response prediction is vital for achieving personalized treatment of cancer and moving precision medicine forward. Large-scale multi-omics profiles provide unprecedented opportunities for precision cancer therapy. Methods In this study, we propose a pipeline to identify subpathway signatures for anticancer drug response of individuals by integrating the comprehensive contributions of multiple genetic and epigenetic (gene expression, copy number variation and DNA methylation) alterations. Results Totally, 46 subpathway signatures associated with individual responses to different anticancer drugs were identified based on five cancer-drug response datasets. We have validated the reliability of subpathway signatures in two independent datasets. Furthermore, we also demonstrated these multi-omics subpathway signatures could significantly improve the performance of anticancer drug response prediction. In-depth analysis of these 46 subpathway signatures uncovered the essential roles of three omics types and the functional associations underlying different anticancer drug responses. Patient stratification based on subpathway signatures involved in anticancer drug response identified subtypes with different clinical outcomes, implying their potential roles as prognostic biomarkers. In addition, a landscape of subpathways associated with cellular responses to 191 anticancer drugs from CellMiner was provided and the mechanism similarity of drug action was accurately unclosed based on these subpathways. Finally, we constructed a user-friendly web interface-CancerDAP ( http://bio-bigdata.hrbmu.edu.cn/CancerDAP/ ) available to explore 2751 subpathways relevant with 191 anticancer drugs response. Conclusions Taken together, our study identified and systematically characterized subpathway signatures for individualized anticancer drug response prediction, which may promote the precise treatment of cancer and the study for molecular mechanisms of drug actions. Electronic supplementary material The online version of this article (10.1186/s12967-019-2010-4) contains supplementary material, which is available to authorized users.

Survival Marker miRNA-Mediated Subpathways of Breast Invasive Carcinoma Derived From Activity Profile

“…Thus, there must be complicated relations among miRNAs, genes and pathways. Another view about risk survival markers is that, disease phenotypes are found to be highly associated with the key local subpathways, rather than entire pathways [25,68,69]. We believe that it is a promising way to understand the biological mechanism of cancer survival with deep mining of miRNA-mediated subpathways.…”

Incorporating Machine Learning into Established Bioinformatics Frameworks

“…With the recent increased availability of multiple, powerful omics techniques (that is, genomics, transcriptomics, proteomics, and metabolomics), a key emerging challenge is the integration of different omics platforms. Several methods have been developed for multi-omics integration using machine and deep learning techniques [133], including SVM [134,135], KNN [136,137], NMF [138], PCA [139] and CNN [140], for example, for cancer subtype and survival prediction [141][142][143] and for prediction of drug response [143,144], the paucity of studies systematically comparing different multi-omics integration methods is a serious bottleneck in the advancement of this field. Such systematic comparison was recently performed for a subset of the multi-omics techniques aimed at the prediction of tumor subtype [145].…”

Omics and Computational Modeling Approaches for the Effective Treatment of Drug-Resistant Cancer Cells

“…In general, machine learning models perform classification or regression, depending on a given problem. Recently, prediction of anticancer drug response was attempted by using various types of machine learning methods, such as logistic regression (Frejno et al, 2017;Yu et al, 2021), random forest (Xu et al, 2019) and deep neural network (DNN; e.g., multilayer perceptron) (Malik et al, 2021) on the basis of a range of omics and drug response data (Table 1). When developing these machine learning models, transcriptome (RNA-seq or mRNA microarray) was the most frequently adopted dataset, but other types of datasets were also considered, including genome (e.g., gene mutations) (Yu et al, 2021), proteome (Frejno et al, 2020), epigenome (Xu et al, 2019), mass spectrometry data (Liu R. et al, 2019) and molecular features of a target drug (Zhu et al, 2020).…”