About Quantum Forge
A unified web workstation for quantum chemistry, transition state optimization, machine learning interatomic potentials (MLIP), and comprehensive pharmacopeia curricula.
Platform Overview
Quantum Forge is an advanced, zero-install computational chemistry research dashboard built with Flutter Web, WebGL, and state-of-the-art machine learning surrogates. Designed and developed by Ali Asghar, the platform provides seamless transition state finding, dynamic nudged elastic band (NEB) reaction trajectory visualization, force-field relaxation, and comprehensive multi-format chemical structure I/O with bidirectional parity to desktop Avogadro 2.
Unlike traditional desktop-bound quantum chemistry software, Quantum Forge runs directly in the browser across Windows, macOS, Linux, and ChromeOS, accelerating research and education worldwide.
Integrated Curricula & Clinical Pharmacopeia
Quantum Forge serves as an educational benchmark compendium and computational laboratory for students and professionals across three rigorous disciplines:
🩺 MBBS Curriculum
Covers human medicine, clinical pharmacology, pharmacodynamics (PD), pharmacokinetics (ADME-Tox), and biochemical reaction energetics. Features high-yield benchmarks such as Aspirin Serine acetylation of COX-1/COX-2 and Paracetamol CYP2E1 metabolism into NAPQI.
💊 Pharm-D Curriculum
Focused on pharmaceutical sciences, organic drug synthesis, transition state activation barriers (ΔG‡, ΔH‡, ΔS‡), enantioselective synthesis, esterification kinetics, and beta-lactam antibacterial mechanisms.
🐾 DVM Curriculum
Addresses veterinary medicine, comparative pharmacokinetics across species (canine, feline, equine, bovine), feline glucuronidation deficiencies, species-specific drug toxicity, and antimicrobial resistance profiles.
Machine Learning Interatomic Potentials (MLIP)
Quantum Forge integrates an ensemble of cutting-edge neural network interatomic potentials and quantum chemical calculators:
- MACE: Higher-order equivariant message passing neural network potential for atomic systems with ab initio accuracy.
- CHGNet: Pretrained Crystal Hamiltonian Graph Neural Network for inorganic and organometallic crystal structures.
- ANI-2x: Deep neural network potential for universal organic and medicinal chemistry.
- GFN2-xTB: Robust semi-empirical quantum mechanical method with extended tight-binding Hamiltonian.
Users can compute activation enthalpies, transition states, reaction free energies, and vibrational partition functions in milliseconds directly from their web browser.