Mining of structural motifs in proteins using artificial bee colony optimization framework for druggability

Area of Science:

• Bioinformatics
• Computational Biology
• Structural Biology

Background:

• DNA binding proteins play crucial roles in cellular processes.
• Identifying common structural patterns in these proteins is essential for understanding their function.
• Existing methods may lack efficiency in pattern discovery and motif generation.

Purpose of the Study:

• To develop an optimization framework for discovering common structural patterns in DNA binding proteins.
• To propose a novel variant of the artificial bee colony algorithm for improved exploitation.
• To identify optimal features for the Helix Turn Helix structural pattern.

Main Methods:

• Developed a novel artificial bee colony optimization algorithm variant.
• Applied the algorithm to identify structural patterns in DNA binding proteins.
• Utilized objective functions based on secondary structure occurrence counts.
• Performed docking studies for druggability assessment.

Main Results:

• The algorithm demonstrated speedier convergence on benchmark functions.
• Generated optimal features for the Helix Turn Helix structural pattern.
• Outperformed compared methods in convergence speed and motif quality.
• Achieved 92% matching motif locations with other motif detection tools.
• Exhibited higher sensitivity, specificity, and area under the curve values.

Conclusions:

• The proposed optimization framework effectively identifies structural patterns in DNA binding proteins.
• The novel algorithm variant enhances motif discovery and feature generation.
• The approach shows significant potential for improving drug discovery through targeted docking studies.