Perspectives on Ligand/Protein Binding Kinetics Simulations: Force Fields, Machine Learning, Sampling, and...

Area of Science:

• Computational chemistry
• Pharmacology
• Biophysics

Background:

• Computational drug discovery aids in identifying active compounds, reducing costs and time.
• Historically, focus has been on high-affinity ligands for stable drug-target complexes.
• Emerging research links in vivo drug efficacy to binding kinetics, necessitating kinetic simulations.

Purpose of the Study:

• To review the current state of ligand/protein binding kinetic simulation techniques in drug discovery.
• To evaluate the limitations of existing methodologies.
• To propose solutions for developing more accurate kinetic models.

Main Methods:

• Review of popular and advanced ligand/protein binding kinetic simulation techniques.
• Analysis of current challenges in computational drug discovery.
• Discussion of potential solutions and future research directions.

Main Results:

• Identified key areas for improvement in kinetic modeling: parametrization, force fields, transition states, sampling, and algorithm performance.
• Highlighted the need for a paradigm shift in current methodologies.
• Emphasized user-friendliness and data openness as crucial for progress.

Conclusions:

• Ligand/protein binding kinetics are crucial for predicting in vivo drug efficacy.
• Significant advancements are needed in computational methodologies for accurate kinetic modeling.
• Future efforts should focus on improving parametrization, sampling, and algorithm development for enhanced drug discovery.