AutoSolvate

AutoSolvate automates high-throughput quantum chemistry (QC) calculations for explicitly solvated molecules to generate solution-phase molecular properties at the quantum mechanical level.

Key Features:

• Automated Workflow: Automates solvated-structure generation, configuration sampling, and extraction of microsolvated cluster structures for downstream QC calculations.
• Machine Learning Integration: Incorporates machine learning models that predict solute–solvent closeness based on solute and solvent identities to guide initial structure generation.
• Force Field Fitting and Configuration Sampling: Performs force field fitting and configuration sampling to produce representative molecular configurations for quantum calculations.
• Microsolvated Cluster Extraction: Extracts microsolvated cluster structures prepared for use in quantum chemistry packages to predict molecular properties.
• High-throughput Explicit-Solvent QC: Supports large-scale explicit-solvent QC calculations for datasets intended for computational and data-driven studies.

Scientific Applications:

• Dataset Generation for AI-driven Design: Generates large, high-quality explicit-solvent quantum chemistry datasets for artificial intelligence–driven chemical design and discovery.
• Reorganization Energy Calculations: Has been applied to calculate outer-sphere reorganization energy for a dataset of 166 redox couples.

Methodology:

Solvated-structure generation aided by machine learning prediction of solute–solvent closeness, followed by force field fitting and configuration sampling, and extraction of microsolvated cluster structures for quantum chemistry analysis.