Dimensionality reduction by UMAP to visualize physical and genetic interactions.

Area of Science:

• Genomics
• Bioinformatics
• Systems Biology

Background:

• Dimensionality reduction is crucial for visualizing high-dimensional biological data, such as gene expression profiles.
• Understanding gene function and interactions is fundamental in systems biology.

Purpose of the Study:

• To apply Uniform Manifold Approximation and Projection (UMAP) for visualizing and analyzing transcript profiles from Saccharomyces cerevisiae gene deletion mutants.
• To identify gene groupings, protein complexes, pathways, and novel protein interactions using UMAP.

Main Methods:

• Utilized published transcript profiles from 1484 single gene deletions of Saccharomyces cerevisiae.
• Applied the Uniform Manifold Approximation and Projection (UMAP) dimensionality reduction technique.
• Analyzed gene proximity in the low-dimensional UMAP space to infer functional relationships.

Main Results:

• UMAP effectively grouped genes corresponding to known protein complexes and pathways.
• The method identified novel protein interactions, including within well-characterized complexes.
• Proximity analysis in UMAP space proved more sensitive than previous methods for detecting gene relationships.

Conclusions:

• UMAP is a powerful tool for visualizing and analyzing complex transcriptomic data.
• This approach facilitates the discovery of gene function, protein complexes, and interactions.
• The UMAP method offers broad utility for future transcriptomic studies across various organisms.