Annotation of proteins of unknown function: initial enzyme results

Area of Science:

• Structural bioinformatics
• Computational biology
• Protein science

Background:

• The Protein Data Bank contains approximately 3,500 structures with unassigned functions.
• Accurate functional annotation is crucial for understanding protein roles in biological systems.

Purpose of the Study:

• To assign functional annotations to proteins with unknown functions in the Protein Data Bank.
• To identify promising candidates for experimental validation.

Main Methods:

• Utilized a combination of ProMOL, BLAST, Pfam, and Dali for homology searches.
• Employed a library of 388 enzymatic motifs for local structural comparisons.
• Screened ~3,500 Protein Data Bank entries classified as 'unknown function'.

Main Results:

• Identified over 500 promising in silico matches using ProMOL.
• Discovered 65 exceptionally strong matches with high confidence.
• Characterized the features of these top-scoring matches.

Conclusions:

• Computational approaches can effectively identify potential functions for proteins with unknown roles.
• The identified high-confidence matches warrant further experimental investigation.
• This study provides a foundation for annotating a significant portion of the structural proteome.