A computer aided system for systematic production and revision of sequence patterns

Area of Science:

• Bioinformatics
• Computational Biology
• Machine Learning

Background:

• Protein sequence pattern analysis is crucial for understanding protein function.
• Interpreting novel protein sequences remains a challenge.

Purpose of the Study:

• To develop a novel method for producing and revising protein sequence patterns.
• To enhance the interpretation of biological functions for uncharacterized sequences.

Main Methods:

• Utilizing object-oriented databases and query languages for pattern generation.
• Employing machine learning for sequence classification based on pattern matches.
• Applying concept lattices for classification and sequence multiple alignment for pattern revision.

Main Results:

• Object-oriented query languages effectively produce and interpret protein sequence patterns.
• A classification system was built using pattern matches, allowing for iterative refinement.
• The integrated approach successfully revised both sequences and patterns.

Conclusions:

• The combination of object-oriented databases and machine learning offers a robust framework for protein sequence analysis.
• This methodology improves the accuracy and utility of protein sequence patterns.
• The approach facilitates a deeper understanding of protein function, particularly for novel sequences.