Quantum Generative Modeling of Single-Cell transcriptomes: Capturing Gene-Gene and Cell-Cell Interactions

Abstract

Single-cell RNA sequencing (scRNA-seq) data simulation is limited by classical methods relying on linear correlations, failing to capture nonlinear dependencies. No existing simulator jointly models gene-gene regulatory interactions and cell-cell communication. We introduce qSimCells, a quantum computing-based simulator that uses entanglement to model intra- and inter-cellular interactions, generating realistic single-cell transcriptomic data from heterogeneous cell populations. Its quantum kernel uses a parameterized circuit with CNOT gates to encode gene regulatory networks (GRNs) and cell-cell communication topologies. By programming the entanglement architecture, the simulator establishes a known generative ground truth for both regulatory and communication pathways. The resulting synthetic data exhibits dependencies arising from the joint probability structure of the quantum circuit. Notably, standard correlation-based analyses (Pearson and Spearman) fail to recover the programmed causal relationships and instead report spurious associations driven by high baseline gene-expression probabilities. Applying cell-cell communication detection serves as an internal consistency check: CellChat correctly identifies the true ligand-receptor pairs when inter-state entanglement is active, revealing a robust, up to ~98-fold relative increase in inferred communication probability. These results demonstrate that the quantum kernel produces high-fidelity benchmark datasets with known ground truth, highlighting the limitations of correlation-based inference and the need for approaches capable of capturing complex structural dependencies underlying gene regulation and cell-cell communication.