Superior performance in protein homology detection with the Blocks Database servers

Area of Science:

• Bioinformatics
• Computational Biology
• Structural Biology

Background:

• Protein homology detection is crucial for understanding protein function and evolution.
• Alignment blocks represent conserved protein regions, aiding in identifying evolutionary relationships.
• Existing tools require enhancement for more sensitive and accurate protein analysis.

Purpose of the Study:

• To introduce updated tools and resources for protein homology detection and analysis.
• To enhance the Blocks Database with additional conserved region blocks.
• To evaluate the performance of blocks-based searching against other methods.

Main Methods:

• Supplementation of the Blocks Database with conserved blocks from the Prints Database.
• Utilizing LAMA for blocks-versus-blocks searching and visualization of logos and trees.
• Integration with MAST, BLAST, and PSI-BLAST servers for sensitive sequence databank searches.

Main Results:

• The Blocks Database now contains 4754 blocks from 1163 protein families.
• Blocks-based tools demonstrated superior performance compared to expert pairwise searching and Hidden Markov Models.
• The integrated system provides sensitive and up-to-date protein sequence analysis.

Conclusions:

• Blocks-based tools offer a superior approach for protein homology detection and analysis.
• The enhanced Blocks Database and associated search tools improve the accuracy and sensitivity of bioinformatics analyses.
• These advancements facilitate a deeper understanding of protein function and evolution.