OncoSimulR

OncoSimulR simulates forward population genetic processes in asexual populations to model cancer progression via sequential accumulation and order-restricted dependencies of driver and passenger mutations.

Key Features:

• Simulation of Cancer Progression: Simulates sequential accumulation of mutations within tumors and models dependencies (order restrictions) among mutations.
• Oncogenetic Trees: Simulates random Directed Acyclic Graphs (DAGs) akin to Oncogenetic Trees to model evolutionary pathways and identify order restrictions, with a limitation in capturing dependencies that require conjunctions of multiple mutations.
• Conjunctive Bayesian Networks and Progression Networks: Supports simulations using Conjunctive Bayesian Networks and Progression Networks as alternative frameworks for tumor evolution.
• Simulation Plotting and Sampling: Plots and samples single or multiple simulation realizations, including single-cell sampling and visualization of true phylogenetic relationships among clones.
• Impact of Sampling Strategies: Enables exploration of sampling at different disease stages versus final-stage sampling to assess effects on detection of order restrictions.
• Incorporation of Passenger Mutations: Simulates passenger mutations alongside driver mutations to reflect their prevalence and evaluate challenges in filtering passengers when inferring order restrictions.

Scientific Applications:

• Cancer genomics and evolutionary biology: Facilitates simulation-based studies of tumor evolution to identify critical mutation sequences relevant to therapy and diagnosis.
• Method benchmarking and inference evaluation: Assesses performance of inferential methods for order restrictions under scenarios that include driver and passenger mutations.
• Experimental design and sampling evaluation: Informs design and interpretation of cross-sectional and single-cell sampling schemes by modeling the impact of different sampling strategies.

Methodology:

Simulates forward population genetic processes in asexual populations using evolutionary models that embed assumptions about mutation order restrictions and driver/passenger status, generates random DAGs (Oncogenetic Trees), implements Conjunctive Bayesian Networks and Progression Networks, and produces sampling and plotting outputs including single-cell realizations and true clone phylogenies.