Using the "Hidden" Genome to Improve Classification of Cancer Types

TL;DR

This paper introduces a multi-level meta-feature regression method that leverages rare and previously unobserved somatic mutations in tumor genomes to improve the accuracy of cancer type classification.

Contribution

It proposes a scalable, high-dimensional modeling approach combining feature screening and group-lasso penalization to extract diagnostic information from the hidden genome.

Findings

Method accurately classifies cancer types using whole-exome sequencing data.

Harnesses diagnostic information from rare and unobserved variants.

Demonstrates effectiveness on TCGA dataset with 3702 samples.

Abstract

It is increasingly common clinically for cancer specimens to be examined using techniques that identify somatic mutations. In principle these mutational profiles can be used to diagnose the tissue of origin, a critical task for the 3-5% of tumors that have an unknown primary site. Diagnosis of primary site is also critical for screening tests that employ circulating DNA. However, most mutations observed in any new tumor are very rarely occurring mutations, and indeed the preponderance of these may never have been observed in any previous recorded tumor. To create a viable diagnostic tool we need to harness the information content in this "hidden genome" of variants for which no direct information is available. To accomplish this we propose a multi-level meta-feature regression to extract the critical information from rare variants in the training data in a way that permits us to also extract diagnostic information from any previously unobserved variants in the new tumor sample. A scalable implementation of the model is obtained by combining a high-dimensional feature screening approach with a group-lasso penalized maximum likelihood approach based on an equivalent mixed-effect representation of the multilevel model. We apply the method to the Cancer Genome Atlas whole-exome sequencing data set including 3702 tumor samples across 7 common cancer sites. Results show that our multi-level approach can harness substantial diagnostic information from the hidden genome.