Dr Amin Alibakhshi
Lecturer
School of Physical and Chemical Sciences
Queen Mary University of London
Research
Machine learning, Interatomic potentials, Uncertainty evaluation, Drug discovery, Molecular modeling, Molecular surfaces
Interests
My research combines artificial intelligence, machine learning, and computational chemistry to understand and predict how molecules behave and interact. A major focus is developing machine-learning methods that can reproduce complex molecular interactions with high accuracy while being much faster than traditional quantum-mechanical calculations. This can make computer simulations of large biological systems, particularly proteins, more realistic and practical.
My current work focuses especially on developing machine-learned interatomic potentials for protein simulations. These models aim to learn from both high-quality quantum-mechanical calculations and experimental measurements, improving their accuracy, reliability, and ability to work across different proteins. Such simulations can help us better understand protein structure, motion, and interactions—processes that are central to areas such as drug discovery, molecular medicine, and biotechnology.
More broadly, I develop and critically assess AI-based molecular models, with particular emphasis on their accuracy, robustness, transferability, and scientific reliability.
Publications
Publications of specific relevance to the Centre for Chemical Research
2025
Bridging the Early Science Gap with Artificial IntelligenceBush A Alibakhshi A
Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems., 1-6.
25-04-2025Bridging the Early Science Gap with Artificial Intelligence: Evaluating Large Language Models as Tools for Early Childhood Science EducationBush A Alibakhshi A
In
Arxiv 24-03-2025 2024
Hydrogen diffusion on Ni(100): A combined machine-learning, ring polymer molecular dynamics, and kinetic Monte Carlo studySteffen J Alibakhshi A
The Journal of Chemical Physics,
Aip Publishing vol. 161 (18)
13-11-2024 On the Theoretical Quantification of Radii of Atoms in MoleculesAlibakhshi A Schäfer LV
The Journal of Physical Chemistry A,
American Chemical Society (Acs) vol. 128 (32), 6819-6823.
02-08-2024 Electron iso-density surfaces provide a thermodynamically consistent representation of atomic and molecular surfacesAlibakhshi A Schäfer LV
Nature Communications,
Springer Nature vol. 15 (1)
19-07-2024 2023
Comparison of Implicit and Explicit Solvent Approaches in Ab Initio Evaluation of Thermochemistry in Solution: Application in Studying Boron Isotope Fractionation in WaterAlibakhshi A Steffen J Pinilla C Hartke B
The Journal of Physical Chemistry A,
American Chemical Society (Acs) vol. 127 (11), 2503-2510.
14-03-2023 2022
Dependence of Vaporization Enthalpy on Molecular Surfaces and Temperature: Thermodynamically Effective Molecular SurfacesAlibakhshi A Hartke B
Physical Review Letters,
American Physical Society (Aps) vol. 129 (20)
09-11-2022 Accurate evaluation of combustion enthalpy by ab-initio computationsAlibakhshi A Schäfer LV
Scientific Reports,
Springer Nature vol. 12 (1)
06-04-2022 Implicitly perturbed Hamiltonian as a class of versatile and general-purpose molecular representations for machine learningAlibakhshi A Hartke B
Nature Communications,
Springer Nature vol. 13 (1)
10-03-2022 2021
Improved prediction of solvation free energies by machine-learning polarizable continuum solvation modelAlibakhshi A Hartke B
Nature Communications,
Springer Nature vol. 12 (1)
18-06-2021 Evaluation of the Temperature Dependence of Vaporization Enthalpy and its Correlation with Surface Tension by Machine Learning and Predictive CorrelationsAlibakhshi A Hartke B
In
Chemrxiv 01-01-2021 Improved Prediction of Solvation Free Energies by Machine-Learning Polarizable Continuum Solvation ModelAlibakhshi A Hartke B
In
Chemrxiv 01-01-2021 2018
Strategies to develop robust neural network models: Prediction of flash point as a case studyAlibakshi A
Analytica Chimica Acta,
Elsevier vol. 1026, 69-76.
09-05-2018 A comparative study of molecular dynamics simulation methods for evaluation of the thermal conductivity and phonon transport in Si nanowiresSoleimani A Araghi H Zabihi Z Alibakhshi A
Computational Materials Science,
Elsevier vol. 142, 346-354.
01-02-2018 2017
Enthalpy of vaporization, its temperature dependence and correlation with surface tension: A theoretical approachAlibakhshi A
Fluid Phase Equilibria,
Elsevier vol. 432, 62-69.
01-01-2017 Prediction of flash points of pure organic compounds: Evaluation of the DIPPR databaseAlibakhshi A Mirshahvalad H Alibakhshi S
Process Safety and Environmental Protection,
Elsevier vol. 105, 127-133.
01-01-2017 2016
Correction to “A Modified Group Contribution Method for Accurate Prediction of Flash Points of Pure Organic Compounds”Alibakhshi A Mirshahvalad H Alibakhshi S
Industrial & Engineering Chemistry Research,
American Chemical Society (Acs) vol. 55 (14), 4099-4099.
29-03-2016 Correction to “A Modified Group Contribution Method for Accurate Prediction of Flash Points of Pure Organic Compounds”Alibakhshi A Mirshahvalad H Alibakhshi S
Industrial & Engineering Chemistry Research,
American Chemical Society (Acs) vol. 55 (8), 2504-2504.
22-02-2016 2015
A Modified Group Contribution Method for Accurate Prediction of Flash Points of Pure Organic CompoundsAlibakhshi A Mirshahvalad H Alibakhshi S
Industrial & Engineering Chemistry Research,
American Chemical Society (Acs) vol. 54 (44), 11230-11235.
03-11-2015 2014
Investigating the Mechanism of Effect of Carbon Nanotubes on Flame Spread Over Liquid FuelsAlibakhshi A Mirshahvalad H Alibakhshi S
Fire Technology,
Springer Nature vol. 51 (4), 759-770.
06-04-2014 Implicitly perturbed Hamiltonian: a class of versatile and general-purpose molecular representations for machine learningAlibakhshi A Hartke B
In
Research Square Theoretical evaluation of equilibrium constant for boron isotopes exchange between boric acid and borate in pure and saline waterAlibakhshi A Hartke B Steffen J Pinilla C
In
Chemrxiv High precision evaluation of the combustion enthalpy by ab-intio computationsAlibakhshi A
In
Chemrxiv Implicitly perturbed Hamiltonian: a class of versatile and general-purpose molecular representations for machine leaningAlibakhshi A Hartke B
In
Chemrxiv Thermodynamically effective molecular surfaces for more efficient study of condensed-phase thermodynamicsAlibakhshi A hartke B
In
Research Square Improved Prediction of Solvation Free Energies by Machine-Learning Polarizable Continuum Solvation ModelAlibakhshi A Hartke B
In
Chemrxiv Evaluation of the Temperature Dependence of Vaporization Enthalpy and its Correlation with Surface Tension by Machine Learning and Predictive CorrelationsAlibakhshi A Hartke B
In
Chemrxiv On the Relativistic Nature of Light: From the Relativistic Interpretation of Planck’s Equation to Theoretical Estimation of Photon’s Intrinsic PropertiesAlibakhshi A
In
Chemrxiv Hydrogen diffusion on Ni(100): A Combined Machine-Learning, Ring Polymer Molecular Dynamics, and Kinetic Monte Carlo StudyAlibakhshi A Steffen J
In
Chemrxiv Thermodynamically Consistent Atomic Radii of Main Group ElementsAlibakhshi A Schäfer LV
In
Chemrxiv Reconciling Experiment with Quantum Chemical Calculations: Electron Iso-Density Surfaces Represent Atomic and Molecular SurfacesAlibakhshi A Schäfer LV
In
Chemrxiv Theoretical Evaluation of Radii of Atoms in Molecules and their Dependence on Atomic Partial ChargeAlibakhshi A Schäfer LV
In
Chemrxiv
Research Group
