ToMExO: A probabilistic tree-structured model for cancer progression

Abstract

Identifying the interrelations among cancer driver genes and the patterns in which the driver genes get mutated is critical for understanding cancer. In this paper, we study cross-sectional data from cohorts of tumors to identify the cancer-type (or subtype) specific process in which the cancer driver genes accumulate critical mutations. We model this mutation accumulation process using a tree, where each node includes a driver gene or a set of driver genes. A mutation in each node enables its children to have a chance of mutating. This model simultaneously explains the mutual exclusivity patterns observed in mutations in specific cancer genes (by its nodes) and the temporal order of events (by its edges). We introduce a computationally efficient dynamic programming procedure for calculating the likelihood of our noisy datasets and use it to build our Markov Chain Monte Carlo (MCMC) inference algorithm, ToMExO. Together with a set of engineered MCMC moves, our fast likelihood calculations enable us to work with datasets with hundreds of genes and thousands of tumors, which cannot be dealt with using available cancer progression analysis methods. We demonstrate our method’s performance on several synthetic datasets covering various scenarios for cancer progression dynamics. Then, a comparison against two state-of-the-art methods on a moderate-size biological dataset shows the merits of our algorithm in identifying significant and valid patterns. Finally, we present our analyses of several large biological datasets, including colorectal cancer, glioblastoma, and pancreatic cancer. In all the analyses, we validate the results using a set of method-independent metrics testing the causality and significance of the relations identified by ToMExO or competing methods.

C-ToMExO: Learning Cancer Progression Dynamics from Clonal Composition of Tumors

“…We use a simplified version of the progression model originally introduced in ToMExO [12] as our core model. This model is designed to explain the progression and mutual exclusivity relations among cancer driver genes and is used at the level of tumors in its corresponding publication.…”

“…As an example, a loss-of-function mutation in a tumor suppressor gene 𝒜 might be the key to igniting a specific biological process, increasing the chance of mutation in another driver gene ℬ. In such a scenario, a so-called progression relation [12] from gene 𝒜 to gene ℬ is expected to emerge in the observations gathered from a cohort of tumors. In other words, we expect to have an over-representation of mutations in gene ℬ among the tumors that have a mutation in gene 𝒜.…”

“…Following our previous example, suppose gene 𝒞 plays a similar role as gene 𝒜. In that case, we expect to see a so-called mutual exclusivity relation [12] among 𝒞 and 𝒜, meaning that while mutations in any one of these genes 𝒜 or 𝒞 improves the evolutionary fitness of the harboring cells, having a mutation in one of them is enough (in the sense of selective advantage). Moreover, the progression relation from the pathway including both 𝒜 and 𝒞 to gene ℬ might be statistically stronger than the progression relation from 𝒜 to ℬ or from 𝒞 to ℬ.…”

“…Moreover, the progression relation from the pathway including both 𝒜 and 𝒞 to gene ℬ might be statistically stronger than the progression relation from 𝒜 to ℬ or from 𝒞 to ℬ. Various methods with different mathematical models and inference techniques are introduced to discover such pathway-level patterns in the driver genes [12–16].…”

“…In this paper, we focus on a recently introduced CPM method called ToMExO [12]. This method uses a probabilistic model for cancer progression.…”