Abstract
With the advancement of high-throughput biotechnologies, we increasingly accumulate biomedical data about diseases, especially cancer. There is a need for computational models and methods to sift through, integrate, and extract new knowledge from the diverse available data, to improve the mechanistic understanding of diseases and patient care. To uncover molecular mechanisms and drug indications for specific cancer types, we develop an integrative framework able to harness a wide range of diverse molecular and pan-cancer data. We show that our approach outperforms the competing methods and can identify new associations. Furthermore, it captures the underlying biology predictive of drug response. Through the joint integration of data sources, our framework can also uncover links between cancer types and molecular entities for which no prior knowledge is available. Our new framework is flexible and can be easily reformulated to study any biomedical problem.
Highlights
Over 18 million new cases of cancer and 9 million deaths were recorded worldwide in 2018.1 This makes cancer one of the leading causes of death
Cancer is a multi-faceted, complex disease arising from an accumulation of somatic mutations within the genome of normal cells that eventually leads to loss of normal cellular functioning and appearance of tumors that can spread across the body
In the first validation step, we systematically evaluate the performance of our approach with a 10-fold crossvalidation using both the area under the receiver operating characteristic (AUROC) and the area under the precision recall curve (AUPRC) and compare our results to state-of-the-art methods for link prediction
Summary
Over 18 million new cases of cancer and 9 million deaths were recorded worldwide in 2018.1 This makes cancer one of the leading causes of death. Cancer projects, including The Cancer Genome Atlas (TCGA) and the International Cancer Genome Consortium (ICGC), have made publicly available wide-ranging, multi-modal, multi-omics cancer data, such as patient whole slide images, genome alterations, transcriptome, and epigenome.[4,5] Free access to these large-scale, diverse databases has dramatically facilitated studies of the biological mechanisms of specific cancer types.[4,6,7] The available data have enabled pan-cancer analyses that study cancer in general to identify common mechanisms and differences across cancer types.[7,8] Recently, the Pan-Cancer Analysis of Whole Genome (PCAWG) project[7] has informed that our knowledge about cancer is far from complete, as 5% of their cohort was without any known cancer driver mutations These large databases have paved the way for the field of Precision Medicine, whose overarching aim is to improve medical care for patients by tailoring treatment to their individual molecular profiles.[9] Precision medicine has diverse intermediary objectives, for instance, uncovering diagnostic and prognostic biomarkers. This is especially relevant to a heterogeneous disease, such as cancer, which manifests uniquely in every patient
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
