scGeno: a Hidden Markov Model approach to denoise chromosome-scale genotypes from single-cell data

Area of Science:

• Genomics
• Computational Biology
• Single-cell analysis

Background:

• Accurate genotype determination is essential for single-cell studies of monoallelic expression and genomic imprinting.
• Challenges in single-cell RNA sequencing (scRNA-seq) data include technical noise, allelic dropout, and sparse expression, especially in heterogeneous populations.

Purpose of the Study:

• To develop a novel computational method for inferring chromosome-level genotype states from scRNA-seq data.
• To enable accurate genotype determination at the single-cell level, facilitating the study of allele-specific expression.

Main Methods:

• Introduced scGeno, a categorical Hidden Markov Model (HMM).
• Models sequential gene expression ratios from scRNA-seq data to infer genotype states along chromosomes.
• Leverages the sequential continuity of genotype states to address limitations of single-cell data.

Main Results:

• Generates chromosome-resolved, comprehensive genotype maps for individual samples.
• The probabilistic framework effectively accounts for technical noise, ensuring high accuracy in genotype assignment.
• Validated on experimental data, demonstrating robust performance in identifying distinct genotypic states.

Conclusions:

• scGeno provides a robust solution for genotype inference from scRNA-seq data.
• Enables systematic investigation of allele-specific expression patterns at single-cell resolution.
• The method overcomes key challenges associated with analyzing genetically heterogeneous single cells.