PL-PatchSurfer2: Improved Local Surface Matching-Based Virtual Screening Method That Is Tolerant to Target and Ligand...

Area of Science:

• Computational chemistry
• Drug discovery
• Bioinformatics

Background:

• Virtual screening is crucial in drug discovery, with ligand-based and structure-based methods having distinct advantages and limitations.
• Structure-based methods offer higher potential for novel compound discovery but are computationally intensive.
• Existing methods face challenges in speed and accuracy, particularly with limited protein structural data.

Purpose of the Study:

• To introduce PL-PatchSurfer2, a novel and efficient structure-based virtual screening method.
• To enhance the accuracy and speed of identifying potential drug candidates.
• To improve virtual screening performance using apo-form or template-based protein models.

Main Methods:

• PL-PatchSurfer2 represents protein and ligand surfaces using overlapping local patches described by three-dimensional Zernike descriptors (3DZDs).
• Incorporates atom-based hydrophobicity and hydrogen-bond interactions to enhance complementarity calculations.
• Utilizes 3DZDs for concise and effective computation of local surface shapes and physicochemical properties.

Main Results:

• PL-PatchSurfer2 demonstrates improved performance compared to previous versions and existing popular methods.
• The method shows superior performance when employing apo-form or template-based protein models.
• Achieves a computational time approximately 20 times shorter than conventional structure-based methods.

Conclusions:

• PL-PatchSurfer2 offers a significant advancement in structure-based virtual screening, balancing speed and accuracy.
• The method holds promise for accelerating the identification of novel drug candidates.
• PL-PatchSurfer2 is available for broader use in the scientific community.