Fast Local Alignment of Protein Pockets (FLAPP): A System-Compiled Program for Large-Scale Binding Site Alignment.

Area of Science:

• Computational Biology
• Structural Bioinformatics
• Drug Discovery

Background:

• Protein function is determined by sequence, structure, and binding site arrangement.
• Sequence analysis offers initial insights, while structural data provides higher resolution for function inference.
• Large-scale structural comparison of binding sites is computationally expensive and underutilized.

Purpose of the Study:

• To develop a rapid algorithm for atomic-level alignment of protein binding sites.
• To enable efficient large-scale structural comparisons for inferring protein function and aiding drug discovery.
• To validate the accuracy and speed of the developed algorithm.

Main Methods:

• Developed FLAPP algorithm combining graph clique matching and modern CPU architectures.
• Achieved rapid atomic-level binding site alignments.
• Validated FLAPP through rigorous multi-level complexity assessments and a case study.

Main Results:

• FLAPP aligns binding sites in milliseconds on standard desktop machines.
• PDB-wide scans are completed within 1-2 minutes.
• Demonstrated accurate alignments and validated through a vitamin B12 binding site case study.

Conclusions:

• FLAPP offers a computationally efficient solution for large-scale binding site analysis.
• The tool facilitates insights into protein function and accelerates drug discovery, including target identification and polypharmacology.
• FLAPP is expected to be invaluable for the scientific community in analyzing millions of site pairs.