SAMTI: Sampling Adaptive Thermodynamic Integration for Alchemical Free Energy Calculations

Accurate and efficient calculation of alchemical free energies is a critical challenge in computational chemistry, frequently hindered by the inherent limitations of conventional thermodynamic integration (TI) methods. These limitations include poor phase-space overlap between discrete alchemical states, inefficient allocation of computational resources, and a fundamental time scale separation between alchemical transformations and molecular conformational sampling, which collectively lead to slow convergence and high statistical uncertainty. This work presents sampling adaptive thermodynamic integration (SAMTI), a unified computational framework designed to systematically overcome these challenges. SAMTI synergistically integrates four components: (1) serial tempering (ST) with a fine-grained alchemical grid to ensure phase-space continuity; (2) variance adaptive resampling (VAR) to dynamically allocate computational effort to high-uncertainty regions; (3) replica exchange (RE) to enhance conformational sampling; and (4) alchemical enhanced sampling (ACES) to resolve kinetic bottlenecks by selectively scaling torsional energy barriers. We evaluated SAMTI's performance against conventional TI across a benchmark suite of eight molecular systems of increasing complexity, including ion solvation, small molecule annihilation, and challenging protein-ligand transformations. The results demonstrate that SAMTI variants reduce statistical error by 40-75% and, for the most complex systems, the complete ST+VAR+RE (mACES) configuration consistently achieves chemical accuracy (σ<sub>Δ<i>G</i></sub> < 0.1 kcal/mol) within 10 ns of the total simulation time, a challenging task for conventional methods. Despite using a finer alchemical discretization, SAMTI achieves superior computational efficiency through adaptive resource allocation and faster convergence while automating the optimization of the alchemical pathway. By providing a robust, automated, and reliable solution to both alchemical and conformational sampling challenges, SAMTI establishes a new benchmark for free energy calculations, positioning it as a powerful tool for accelerating molecular design in drug discovery and materials science.

Authors

Institutions

Publication Details

Journal
The Journal of Physical Chemistry B
Published
2025-12-09
DOI
https://doi.org/10.1021/acs.jpcb.5c05358
Primary Topic
Protein Structure and Dynamics
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

SAMTI: Sampling Adaptive Thermodynamic Integration for Alchemical Free Energy Calculations

Tai‐Sung Lee, Saikat Pal, Darrin M. York, Omid Jahanmahin
The Journal of Physical Chemistry B
Protein Structure and Dynamics
article

SAMTI: Sampling Adaptive Thermodynamic Integration for Alchemical Free Energy Calculations

Tai‐Sung Lee, Saikat Pal, Darrin M. York, Omid Jahanmahin
article en

Abstract

Accurate and efficient calculation of alchemical free energies is a critical challenge in computational chemistry, frequently hindered by the inherent limitations of conventional thermodynamic integration (TI) methods. These limitations include poor phase-space overlap between discrete alchemical states, inefficient allocation of computational resources, and a fundamental time scale separation between alchemical transformations and molecular conformational sampling, which collectively lead to slow convergence and high statistical uncertainty. This work presents sampling adaptive thermodynamic integration (SAMTI), a unified computational framework designed to systematically overcome these challenges. SAMTI synergistically integrates four components: (1) serial tempering (ST) with a fine-grained alchemical grid to ensure phase-space continuity; (2) variance adaptive resampling (VAR) to dynamically allocate computational effort to high-uncertainty regions; (3) replica exchange (RE) to enhance conformational sampling; and (4) alchemical enhanced sampling (ACES) to resolve kinetic bottlenecks by selectively scaling torsional energy barriers. We evaluated SAMTI's performance against conventional TI across a benchmark suite of eight molecular systems of increasing complexity, including ion solvation, small molecule annihilation, and challenging protein-ligand transformations. The results demonstrate that SAMTI variants reduce statistical error by 40-75% and, for the most complex systems, the complete ST+VAR+RE (mACES) configuration consistently achieves chemical accuracy (σ<sub>Δ<i>G</i></sub> < 0.1 kcal/mol) within 10 ns of the total simulation time, a challenging task for conventional methods. Despite using a finer alchemical discretization, SAMTI achieves superior computational efficiency through adaptive resource allocation and faster convergence while automating the optimization of the alchemical pathway. By providing a robust, automated, and reliable solution to both alchemical and conformational sampling challenges, SAMTI establishes a new benchmark for free energy calculations, positioning it as a powerful tool for accelerating molecular design in drug discovery and materials science.

The Journal of Physical Chemistry BVol. 129(51)
QB3 (US), Rutgers Sexual and Reproductive Health and Rights (NL), Quantitative BioSciences (US)
Openalex Percentile: Top 64%
Protein Structure and Dynamics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.