Identification of regulatory elements using a feature selection method

Area of Science:

• Computational Biology
• Genomics
• Bioinformatics

Background:

• Identifying regulatory motifs in gene transcription control regions is crucial for understanding gene regulation.
• Existing methods often focus on co-expressed genes across multiple conditions.
• This study addresses motif identification under a single experimental condition.

Purpose of the Study:

• To identify sequence motifs associated with activated genes under a specific experimental condition.
• To develop a method utilizing gene expression data for motif discovery.
• To apply the method to yeast Saccharomyces cerevisiae.

Main Methods:

• Utilized a linear model with two-way interactions to correlate gene expression with sequence features (words) in upstream control regions (UCRs).
• Employed stepwise selection with Monte Carlo cross-validation for feature selection.
• Focused on the 800 basepairs upstream of the translation start site in yeast.

Main Results:

• Successfully identified known active regulatory motifs under the analyzed experimental conditions.
• Discovered additional significant sequences potentially representing novel regulatory motifs.
• Demonstrated the utility of gene expression data for motif discovery in a single condition.

Conclusions:

• The developed method effectively identifies regulatory motifs associated with gene activation in a specific condition.
• The approach offers an alternative to traditional methods relying on co-expressed gene clusters.
• The findings contribute to a deeper understanding of gene regulation in yeast.