This section collects research notes on computational chemistry. The bulk of the notes cover optical chemical structure recognition (OCSR), the problem of extracting structured chemical information from images of molecular diagrams. The remaining notes span molecular dynamics simulations, molecular representations, chemical language models, molecular modeling, curated datasets, and a small set of classic papers in the field.
Notes on 3D molecular representations, conformer generation, and neural network potentials. Covers geometric deep learning approaches for molecules.
An archive of notes and research covering the application of large language models to chemical problems. Topics range from standard 1D sequence translation (like SMILES to IUPAC mapping) to multimodal vision-language models that link 2D molecular graphs with text.
Standard test problems, validation platforms, and datasets for evaluating computational chemistry methods. Covers analytical energy surfaces, optimization challenges, and generative modeling benchmarks.
Explore the landmark papers that shaped computational chemistry, covering breakthroughs in molecular dynamics, quantum chemistry, and statistical mechanics.
A curated collection of molecular databases and datasets used in computational chemistry. These dataset cards provide overviews of molecular enumeration databases, conformer generation datasets, and reaction modeling resources.
A collection of notes on classical molecular dynamics literature from the 1980s and 1990s. The focus is on interatomic potential models (embedded atom method, Stillinger-Weber), adatom diffusion and self-diffusion on metal surfaces, and oscillatory reactions on platinum surfaces.
Notes on SMILES, InChI, SELFIES, and other molecular string formats. Covers encoding schemes, their properties, and applications in cheminformatics.
Notes on recognizing molecular structures from images, covering 35 years of methods: from rule-based vectorization to vision-language models.