Related Experiment Videos
Gleaning non-trivial structural, functional and evolutionary information about proteins by iterative database
1National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health, Bethesda, MD 20894, USA.
Journal of Molecular Biology
|May 1, 1999
Summary
PSI-BLAST effectively identifies subtle protein relationships, but success hinges on selecting diverse query sequences for iterative database searches. Careful analysis of protein superfamilies enhances structural and functional predictions.
Area of Science:
- * Bioinformatics
- * Structural Biology
- * Computational Biology
Background:
- * Identifying subtle evolutionary relationships between proteins is crucial for understanding function.
- * Traditional methods often require structure-structure comparisons, limiting large-scale analysis.
- * Iterative sequence database search methods offer a promising alternative for detecting distant homology.
Purpose of the Study:
- * To evaluate the efficacy of PSI-BLAST in detecting subtle protein relationships.
- * To determine the optimal strategy for using PSI-BLAST to maximize detection of non-trivial inferences.
- * To illustrate the method's utility through unexpected structural predictions.
Main Methods:
- * Utilized diverse protein families as test cases for PSI-BLAST analysis.
- * Investigated the impact of query sequence selection on PSI-BLAST performance.
- * Correlated the diversity of initial search results with the ability to detect subtle relationships in subsequent iterations.
Main Results:
- * PSI-BLAST successfully identified many subtle protein relationships, though not all.
- * Optimal query sequence selection and diverse initial screening significantly improved detection rates.
- * Revealed novel structural and functional insights, including new members of the HSP70-actin fold and a shared fold for ATP/NAD-dependent DNA ligases.
Conclusions:
- * PSI-BLAST is a powerful tool for uncovering distant protein homology.
- * Strategic query selection and thorough superfamily analysis are essential for maximizing PSI-BLAST's predictive power.
- * The study provides novel structural and functional predictions for several protein families, expanding our understanding of protein evolution and function.