cancerSimCraft: A Multi-resolution Cancer Genome Simulator with Comprehensive Ground Truth Tracking

Cancer evolution follows complex trajectories involving diverse genomic alterations and clonal dynamics, making it challenging to validate computational methods for single-cell DNA sequencing analysis. Here we present cancerSimCraft, a comprehensive framework for simulating cancer genome data at both clonal and single-cell resolution. cancerSimCraft combines deterministic rules with stochastic processes to model various genomic events including CNVs, SNVs, and WGDs. The framework enables integration of real cancer genome patterns with user-defined parameters, supporting customizable simulation designs that reflect both empirical data and theoretical models. Through systematic benchmarking, we demonstrate cancerSimCrafts utility in evaluating computational methods under various conditions, particularly focusing on the impact of dataset size and parameter sensitivity. We further demonstrated its application in exploring clonal evolution and mutation patterns through controlled in silico experiments. These comprehensive simulation capabilities make cancerSimCraft a valuable resource for both computational method development and theoretical studies in cancer genomics.