Novel sequence-based method for identifying transcription factor binding sites in prokaryotic genomes

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Microbial genomic sequence analysis is crucial due to rapid sequencing advancements.
• Current experimental methods cannot keep pace with the volume of genomic data generated.
• Existing computational work focuses on sequence assembly, gene annotation, and metabolic reconstruction.

Purpose of the Study:

• To develop a computational method for determining prokaryotic transcription factor (TF) binding specificities.
• To leverage existing sequence data for predicting TF-DNA interactions.

Main Methods:

• Identified specificity-determining residues from DNA-protein complex crystal structures.
• Grouped TFs with shared critical residues into specificity classes.
• Defined putative TF binding regions including promoters and operons.
• Utilized MEME algorithm to identify motifs within TF specificity classes.

Main Results:

• Achieved 86% prediction sensitivity for the LacI TF family.
• Achieved 80% prediction sensitivity for the TetR TF family.
• Demonstrated the method's effectiveness using RegulonDB annotated sites.

Conclusions:

• The developed computational method accurately predicts prokaryotic TF binding specificities.
• This approach offers a scalable solution for analyzing microbial genomic data.
• The method can aid in understanding gene regulation in prokaryotes.