Classifying tumors by supervised network propagation
Open Access
- 27 June 2018
- journal article
- conference paper
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 34 (13) , i484-i493
Abstract
Network propagation has been widely used to aggregate and amplify the effects of tumor mutations using knowledge of molecular interaction networks. However, propagating mutations through interactions irrelevant to cancer leads to erosion of pathway signals and complicates the identification of cancer subtypes. To address this problem we introduce a propagation algorithm, Network-Based Supervised Stratification (NBS2), which learns the mutated subnetworks underlying tumor subtypes using a supervised approach. Given an annotated molecular network and reference tumor mutation profiles for which subtypes have been predefined, NBS2 is trained by adjusting the weights on interaction features such that network propagation best recovers the provided subtypes. After training, weights are fixed such that mutation profiles of new tumors can be accurately classified. We evaluate NBS2 on breast and glioblastoma tumors, demonstrating that it outperforms the best network-based approaches in classifying tumors to known subtypes for these diseases. By interpreting the interaction weights, we highlight characteristic molecular pathways driving selected subtypes. The NBS2 package is freely available at: https://github.com/wzhang1984/NBSS. Supplementary data are available at Bioinformatics online.Keywords
Funding Information
- NIH (U24CA184427, U54CA209891, P50GM085764, P41GM103504, R01GM070743)
This publication has 55 references indexed in Scilit:
- The Cancer Genome Atlas Pan-Cancer analysis projectNature Genetics, 2013
- Comprehensive molecular portraits of human breast tumoursNature, 2012
- Architecture of the human regulatory network derived from ENCODE dataNature, 2012
- Integrated genomic analyses of ovarian carcinomaNature, 2011
- Deconstructing the molecular portraits of breast cancerMolecular Oncology, 2010
- International network of cancer genome projectsNature, 2010
- Influence of pharmacogenetics on response and toxicity in breast cancer patients treated with doxorubicin and cyclophosphamideBritish Journal of Cancer, 2010
- Integrated Genomic Analysis Identifies Clinically Relevant Subtypes of Glioblastoma Characterized by Abnormalities in PDGFRA, IDH1, EGFR, and NF1Cancer Cell, 2010
- Epidemiology and etiology of gliomasActa Neuropathologica, 2005
- A census of human cancer genesNature Reviews Cancer, 2004