Echidna: A Bayesian framework for quantifying gene dosage effect impacting phenotypic plasticity

Abstract

Phenotypic plasticity, the ability of cells to adapt their behavior in response to genetic or environmental changes, is a fundamental biological process that drives cellular diversity in both normal and pathological contexts, including in tumor evolution. While chromosomal instability and somatic copy number alterations (CNAs) are known to influence cellular states, it remains difficult to separate genetic from cell non-autonomous mechanisms that govern transcriptional variability. Here, we present Echidna, a Bayesian hierarchical model that integrates single-cell RNA sequencing (scRNA-seq) and bulk whole-genome sequencing (WGS) data to quantify the impact of CNAs on gene expression dynamics. By jointly inferring clone-specific CNA profiles and uncovering clonal dependencies, Echidna bridges genomic and transcriptomic landscapes within and across multiple time points, enabling the decoupling of gene dosage effects from cell-extrinsic factors on phenotypic plasticity. Applying Echidna to patient tumor specimens, we demonstrate its superior performance in clonal reconstruction and derive insights into resistance mechanisms.