NMFClustering: Accessible NMF-based clustering utilizing GPU acceleration.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Non-negative Matrix Factorization (NMF) is a dimensionality reduction technique for analyzing gene expression data.
• The computational intensity of NMF has limited its application to large datasets like single-cell RNA sequencing (scRNA-seq).

Approach:

• Implemented NMF-based clustering utilizing high-performance GPU computing with CuPy and Message Passing Interface (MPI).
• Optimized NMF for accelerated analysis of large-scale RNA-Seq and scRNA-seq count matrices.

Key Points:

• Achieved up to a three-orders-of-magnitude reduction in computation time.
• Enabled practical NMF clustering analysis for large genomic datasets.
• Integrated the method into the GenePattern gateway for public access and multi-omic data analysis.

Conclusions:

• Accelerated NMF clustering makes large-scale gene expression analysis feasible.
• The GenePattern gateway provides accessible, reproducible bioinformatics tools for researchers.
• Facilitates in silico research for non-programmers through user-friendly interfaces and HPC integration.