JPOP
JPOP predicts operon structures within sequenced genomes by integrating COG (Cluster of Orthologous Groups) assignments, phylogenetic profiles, and intergenic distances to distinguish within-operon (WO) from transcription-unit-border (TUB) gene pairs.
Key Features:
- COG assignments: Uses COG (Cluster of Orthologous Groups) functional category assignments to assess functional similarity between adjacent genes.
- Phylogenetic profiles: Incorporates gene phylogenetic profile similarity to identify co-evolving and potentially co-expressed gene pairs.
- Intergenic distances: Employs intergenic distance measurements as a genomic indicator of co-transcription for adjacent genes.
- Log-likelihood estimation: Estimates log-likelihood distributions for each data type (COG, phylogenetic profiles, intergenic distances).
- Perceptron-based data fusion: Fuses the per-data-type likelihoods using a perceptron model to classify WO versus TUB gene pairs.
- Implementation: Implemented using Java programs and MATLAB scripts.
Scientific Applications:
- Escherichia coli K12 operon prediction: Evaluated on 236 known operons from Escherichia coli K12, achieving an overall accuracy of 83.8% when integrating multiple data sources.
- Individual data-source performance: Reported accuracies using single information types were 80.4% for intergenic distances, 74.4% for COG assignments, and 70.6% for phylogenetic profiles.
- Synechococcus sp. WH8102 application: Applied to Synechococcus sp. WH8102 to predict 556 putative operons, including results combined with a previous comparative genome analysis-based method.
Methodology:
Estimate log-likelihood distributions for COG assignments, phylogenetic profiles, and intergenic distances, then fuse these distributions with a perceptron model to classify WO and TUB gene pairs; implemented with Java programs and MATLAB scripts, and leveraging the observation that WO pairs tend to have shorter intergenic distances, higher COG sharing probability, and more similar phylogenetic profiles than TUB pairs.
Topics
Details
- Tool Type:
- command-line tool
- Operating Systems:
- Linux, Windows, Mac
- Programming Languages:
- Java, MATLAB
- Added:
- 8/3/2017
- Last Updated:
- 12/10/2018
Operations
Publications
Chen X, et al. Computational prediction of operons in Synechococcus sp. WH8102. Genome Inform. 2004; 15:211-22.