Related Experiment Video
Updated: Jan 1, 2026
Enriching Subcellular Proteins in Leptospira Using a Triton X-114-Based Fractionation Approach
Published on: August 8, 2025
PSO-LocBact: A Consensus Method for Optimizing Multiple Classifier Results for Predicting the Subcellular
Supatcha Lertampaiporn1, Sirapop Nuannimnoi1, Tayvich Vorapreeda1
1Biochemical Engineering and Systems Biology Research Group, National Center for Genetic Engineering and Biotechnology (BIOTEC), King Mongkut's University of Technology Thonburi, Bangkhuntien, Bangkok 10150, Thailand.
This study introduces PSO-LocBact, a new method using particle swarm optimization (PSO) to combine bacterial subcellular localization predictions. It resolves conflicting results, improving accuracy for both Gram-negative and Gram-positive bacteria.
Area of Science:
- Bioinformatics
- Computational Biology
- Genomics
Background:
- Computational methods for predicting subcellular localization vary in performance due to differences in features, datasets, and algorithms.
- Conflicting predictions from multiple tools can hinder accurate protein annotation, especially for complex tasks like bacterial subcellular localization.
Purpose of the Study:
- To develop a consensus classifier that integrates predictions from multiple subcellular localization tools for bacteria.
- To enhance prediction accuracy and resolve inconsistencies using the particle swarm optimization (PSO) algorithm.
Main Methods:
- The particle swarm optimization (PSO) algorithm was employed to combine outputs from existing bacterial protein localization predictors.
- A consensus classifier, named PSO-LocBact, was developed based on PSO.
Main Results:
- PSO-LocBact effectively resolved inconsistencies in subcellular localization predictions for both Gram-negative and Gram-positive bacterial proteins.
- The proposed method achieved an average accuracy exceeding 98% on test datasets, outperforming individual predictors.
Conclusions:
- PSO-LocBact offers a robust approach to improve the reliability of bacterial subcellular localization predictions.
- Integrating diverse prediction models via PSO enhances overall prediction performance and aids in accurate protein annotation.
More Related Videos
08:40Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
Published on: April 8, 2016
07:35Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
Published on: October 11, 2018