FLASH-MM: fast and scalable single-cell differential expression analysis using linear mixed-effects models.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) provides high-resolution gene expression data.
• Analyzing scRNA-seq data for differential gene expression presents challenges in scalability, sample correlation, and individual variation.

Purpose of the Study:

• To develop a fast and scalable algorithm for single-cell differential expression analysis.
• To address computational complexity and memory usage in linear mixed-effects model (LMM) estimation for scRNA-seq data.

Main Methods:

• Developed FLASH-MM, a novel estimation algorithm for linear mixed-effects models (LMMs).
• Reformulated LMM estimation procedures to enhance computational speed and reduce memory requirements.
• Validated the algorithm using simulated scRNA-seq data and real-world datasets (tuberculosis immune cells, kidney cells).

Main Results:

• FLASH-MM demonstrates accuracy and computational efficiency in differential expression analysis.
• The algorithm effectively controls false positive rates and maintains high statistical power.
• FLASH-MM significantly accelerates single-cell differential expression analysis across various biological contexts.

Conclusions:

• FLASH-MM offers a powerful and efficient solution for analyzing large-scale scRNA-seq data.
• The algorithm's speed and accuracy make it valuable for diverse biological applications.
• FLASH-MM facilitates robust differential gene expression analysis in single-cell studies.