Macrocycle Conformational Sampling by DFT-D3/COSMO-RS Methodology.

Area of Science:

• Computational Chemistry
• Molecular Modeling
• Drug Discovery

Background:

• Accurate conformational sampling is crucial for understanding molecular behavior.
• Existing methods for macrocyclic compounds require optimization for efficiency and reliability.

Purpose of the Study:

• To establish a robust computational protocol for mapping the conformational space of medium-sized molecules.
• To compare the performance of different conformational sampling algorithms.

Main Methods:

• Exhaustive conformational sampling of seven macrocyclic compounds and one acyclic analogue.
• Free energy calculations using a composite DFT-D3/COSMO-RS protocol.
• Evaluation of MD/LLMOD and semi-empirical quantum mechanics (SQM) MD methods.

Main Results:

• The MD/LLMOD protocol identified a greater number of low-energy conformers compared to SQM MD.
• Computational cost for evaluating a medium-sized compound ranges from 1,000-100,000 CPU hours.
• A composite protocol with checkpoints and data truncation was formulated.

Conclusions:

• The MD/LLMOD method offers a more complete conformational sampling for macrocycles.
• The developed protocol provides superior insights into molecular conformations at an affordable computational cost.
• This work aids in the design of macrocyclic inhibitors and other molecules.