omicsGMF: a multi-tool for dimensionality reduction, batch correction and imputation applied to bulk- and single cell...

Area of Science:

• Proteomics
• Bioinformatics
• Computational Biology

Background:

• Mass spectrometry (MS) enables large-scale proteomics but faces challenges with batch effects and missing data.
• Current workflows involve sequential, less efficient data processing steps.

Purpose of the Study:

• Introduce omicsGMF, a scalable matrix factorization method for proteomics data.
• Address limitations in dimensionality reduction, batch correction, and imputation for MS data.

Main Methods:

• Developed omicsGMF, a unified framework integrating imputation, batch correction, and principal component analysis.
• Applied omicsGMF to bulk and single-cell proteomics datasets.

Main Results:

• omicsGMF offers fast, scalable, and interpretable dimensionality reduction.
• Demonstrated superior performance in imputing missing values compared to existing tools.
• Showcased enhanced statistical power for detecting differentially abundant proteins.

Conclusions:

• omicsGMF significantly improves proteomics data processing and analysis.
• Provides a scalable solution for dimensionality reduction in large-scale proteomics studies.