Automated protein motif generation in the structure-based protein function prediction tool ProMOL

Area of Science:

• Biochemistry
• Structural Biology
• Bioinformatics

Background:

• ProMOL is a PyMOL plugin for structure-based protein function prediction.
• Manual generation of motif templates for enzyme active site screening is time-consuming and requires expert optimization.
• Existing methods lack automated routines for motif building and testing.

Purpose of the Study:

• To develop and validate an automated algorithm for generating and testing protein active site motif templates within ProMOL.
• To reduce user intervention in the motif generation process.
• To expand the library of available motif templates for enhanced protein function prediction.

Main Methods:

• Developed an algorithm utilizing empirically derived parameters for automated motif template generation and optimization.
• Compared the performance of automatically generated motifs against manually created ones using 112 PDB entries.
• Generated an additional 296 motifs based on Catalytic Site Atlas entries.

Main Results:

• Automated and manual motifs demonstrated equal effectiveness in identifying homologous enzyme active sites and rejecting unrelated structures.
• Newly generated automated motifs showed performance comparable to existing manual templates in hit rates and RMSD values.
• The automated approach requires minimal user input for motif generation and testing.

Conclusions:

• The automated motif generation algorithm effectively creates reliable protein active site templates.
• This automation streamlines the use of ProMOL for structure-based protein function prediction.
• The expanded library of automated motifs enhances the capability to screen query structures for enzyme active sites.