Mathews, J. C. | bioRxiv Science

bioRxiv2019

Fingerprints of cancer by persistent homology

We have carried out a topological data analysis of gene expressions for different databases based on the Fermat distance between the z scores of different tissue samples. There is a critical value of the filtration parameter at which all clusters collapse in a single one. This critical value for healthy samples is gapless and smaller than that for cancerous ones. After collapse in a single cluster, topological holes persist for larger filtration parameter values in cancerous samples. Barcodes, persistence diagrams and Betti numbers as functions of the filtration parameter are different for different types of cancer and constitute fingerprints thereof.

cancer biology↗

bioRxiv2019

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

The 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.

bioinformatics↗