OpDetect: A convolutional and recurrent neural network classifier for precise and sensitive operon detection from...

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Operons are key genetic structures in prokaryotes, regulating gene function and expression.
• Current operon detection methods often lack generalizability across species.
• Accurate operon identification is crucial for understanding prokaryotic and some eukaryotic gene regulation.

Purpose of the Study:

• To develop a general and accurate computational method for operon detection.
• To leverage RNA-sequencing data for improved operon identification.
• To create a species-agnostic tool applicable to diverse organisms.

Main Methods:

• Developed OpDetect, a novel computational approach utilizing RNA-sequencing reads.
• Employed a deep neural network architecture combining convolutional and recurrent layers.
• Applied the method to analyze genomic data for operon identification.

Main Results:

• OpDetect demonstrated superior performance in recall, F1-score, and AUROC compared to existing methods.
• The method achieved high accuracy in operon detection across various bacterial species.
• OpDetect successfully identified operons in the eukaryotic organism Caenorhabditis elegans.

Conclusions:

• OpDetect provides a robust and versatile tool for operon detection.
• The species-agnostic nature of OpDetect enhances its applicability in comparative genomics.
• This method advances the study of gene organization and regulation in prokaryotes and beyond.