Related Experiment Videos
Predicting human immunodeficiency virus protease cleavage sites in proteins by a discriminant function method
K C Chou1, A G Tomasselli, I M Reardon
1Pharmacia & Upjohn Laboratories, Kalamazoo, Michigan 49001-4940, USA.
Proteins
|January 1, 1996
Summary
This study introduces a novel discriminant function to predict HIV-1 protease cleavage sites in proteins. The method accurately identifies substrates, aiding in understanding this crucial therapeutic target.
Area of Science:
- Biochemistry
- Molecular Biology
- Computational Biology
Background:
- HIV-1 protease is a critical target for antiviral therapy.
- Understanding HIV-1 protease specificity is essential for drug development.
- Predicting cleavage sites aids in studying viral polyprotein processing.
Purpose of the Study:
- To develop a predictive method for identifying HIV-1 protease cleavage sites.
- To enhance understanding of HIV-1 protease substrate specificity.
- To provide a novel computational tool for HIV research.
Main Methods:
- Utilized a sequence-coupled (Markov chain) model and vector-projection principle.
- Defined a discriminant function (delta = phi+ - phi-) based on cleavable and noncleavable peptide attributes.
- Trained and validated the method on known HIV-1 protease substrates and non-substrates.
Main Results:
- Achieved 100% prediction accuracy for known cleavable peptides and 96.7% for noncleavable peptides in the training set.
- Accurately predicted 100% of outside sequences known to be cleaved by HIV-1 protease.
- Successfully predicted cleavage sites in native HIV-1/HIV-2 reverse transcriptases and HIV-1 polyproteins.
Conclusions:
- The developed discriminant function method accurately predicts HIV-1 protease cleavage sites.
- This algorithm offers a new approach to understanding HIV-1 protease specificity.
- The findings contribute to the rational design of HIV therapeutics.