Edinburgh Research Explorer

Detecting repeated cancer evolution from multi-region tumor sequencing data

Research output: Contribution to journalArticle

Related Edinburgh Organisations

Open Access permissions

Open

Documents

https://www.nature.com/articles/s41592-018-0108-x
Original languageEnglish
Pages (from-to)707-714
Number of pages8
JournalNature Methods
Volume15
Early online date31 Aug 2018
DOIs
StateE-pub ahead of print - 31 Aug 2018

Abstract

Recurrent successions of genomic changes, both within and between patients, reflect repeated evolutionary processes that are valuable for anticipating cancer progression. Multi-region sequencing allows the temporal order of some genomic changes to be inferred within a tumour, but the robust identification of repeated evolution across patients remains an unmet challenge. Here we present a machine learning method based on Transfer Learning that overcomes the stochastic effects of cancer evolution and noise in the data, and identifies hidden evolutionary patterns in cancer cohorts. When applied to multi-region sequencing datasets from lung, breast, renal and colorectal cancer (768 samples from 178 patients), our method detected repeated evolutionary trajectories in subgroups of patients, which reproduced in single-sample cohorts (n=2,935). Our method provides ways to classify patients based on how their tumour evolved, with implications for anticipating cancer evolution.

ID: 70385842