Related Experiment Videos
Bayesian protein family classifier
K Qu1, L A McCue, C E Lawrence
1Wadsworth Center for Laboratories and Research, Albany, NY 12201, USA. quk, mccue, lawrence@wadsworth.org
Proceedings. International Conference on Intelligent Systems for Molecular Biology
|October 23, 1998
Summary
This study introduces a Bayesian method for aligning and classifying protein sequences into subclasses. The approach identifies distinct subclasses within cyclic nucleotide-binding proteins, revealing their unique sequence characteristics.
Area of Science:
- Bioinformatics
- Computational Biology
- Molecular Evolution
Background:
- Accurate classification of protein sequences is crucial for understanding protein function and evolution.
- Identifying conserved motifs and subclass-specific characteristics aids in functional annotation.
- Cyclic nucleotide-binding proteins represent a diverse superfamily with varied biological roles.
Purpose of the Study:
- To develop a Bayesian procedure for simultaneous sequence alignment and classification.
- To identify conserved sequence features and subclass-specific characteristics within protein superfamilies.
- To apply the method to the cyclic nucleotide-binding protein superfamily and delineate its subclasses.
Main Methods:
- A Gibbs sampling algorithm iterating between alignment and classification steps.
- Bayesian inference for determining the number of conserved columns, motifs, and class sizes.
- Bayesian prediction to highlight inter-class differences for classification refinement.
Main Results:
- Simultaneous alignment and classification of sequences into distinct subclasses.
- Identification of five subclasses within the cyclic nucleotide-binding protein superfamily.
- Characterization of similarities and differences in sequence features across identified subclasses.
Conclusions:
- The Bayesian procedure effectively classifies protein sequences and reveals subclass-specific characteristics.
- The method successfully delineated five subclasses within the cyclic nucleotide-binding protein superfamily.
- Findings provide insights into the evolutionary relationships and functional diversity of these proteins.