Predicting apoptosis protein subcellular location with PseAAC by incorporating tripeptide composition
Bo Liao1, Jun-Bao Jiang, Qing-Guang Zeng
1School of computer and communication, Hunan University, Changsha Hunan, 410082, China. dragonbw @163.com
Protein and Peptide Letters
|May 25, 2011
Summary
This study introduces a novel pseudo amino acid composition (PseAAC) method for encoding protein sequences. The approach accurately predicts protein subcellular localization, aiding in understanding protein function.
Area of Science:
- Bioinformatics
- Proteomics
- Computational Biology
Background:
- Protein function is intrinsically linked to its subcellular location.
- Understanding protein sorting mechanisms and predicting subcellular localization are crucial for elucidating protein functions.
Purpose of the Study:
- To develop a new method for encoding protein sequences using physicochemical properties.
- To predict protein subcellular localization more effectively.
Main Methods:
- Introduced a novel pseudo amino acid composition (PseAAC) approach.
- Encoded protein sequences into 146-dimensional vectors based on amino acid composition and adjacent triune residue content.
- Utilized support vector machine (SVM) algorithm for prediction and conducted jackknife tests on three datasets.
Main Results:
- Achieved prediction accuracies of 84.9%, 91.2%, and 92.6% on three independent datasets.
- Demonstrated the effectiveness of the proposed PseAAC encoding scheme.
Conclusions:
- The developed PseAAC method is a valuable tool for bioinformatics and proteomics.
- Accurate prediction of subcellular localization enhances the understanding of protein functions.
