A Bayesian approach to model uncertainty in single-cell genomic data

TL;DR

This paper presents a Bayesian framework for clustering single-cell genomic data that models uncertainty, captures transitional cellular states, and improves analysis of development and disease processes.

Contribution

It introduces a variational Bayesian Gaussian mixture model for probabilistic cell clustering, enabling uncertainty quantification and dynamic cellular state analysis.

Findings

Captures cellular transitions during differentiation and disease progression.

Provides biologically coherent insights into neurogenesis and breast cancer.

Introduces a new metric for clustering performance evaluation.

Abstract

Network models provide a powerful framework for analysing single-cell count data, facilitating the characterisation of cellular identities, disease mechanisms, and developmental trajectories. However, uncertainty modeling in unsupervised learning with genomic data remains insufficiently explored. Conventional clustering methods assign a singular identity to each cell, potentially obscuring transitional states during differentiation or mutation. This study introduces a variational Bayesian framework for clustering and analysing single-cell genomic data, employing a Bayesian Gaussian mixture model to estimate the probabilistic association of cells with distinct clusters. This approach captures cellular transitions, yielding biologically coherent insights into neurogenesis and breast cancer progression. The inferred clustering probabilities enable further analyses, including Differential Expression Analysis and pseudotime analysis. Furthermore, we propose utilising the misclustering rate and Area Under the Curve in clustering scRNA-seq data as an innovative metric to quantitatively evaluate overall clustering performance. This methodological advancement enhances the resolution of single-cell data analysis, enabling a more nuanced characterisation of dynamic cellular identities in development and disease.