An AI system for predicting protein conformation. It is possible to predict the three-dimensional structure (folding structure) of a protein from its primary sequence (amino acid sequence). It learns hundreds of thousands of protein structure databases and uses DeepMind-based deep learning techniques to predict the conformation of new proteins from their amino acid sequences.
A library related to the symmetry of crystal structures. By providing a crystal structure, Spglib can detect information related to the symmetry of the structure, such as symmetry operations, a space group and a primitive cell. It can also generate irreducible wave numbers. Spglib is written in C, but various interfaces are available, including Python, Fortran, and Rust.
A tool for generating wavevector paths in band calculations of solids. It identifies high-symmetry points in reciprocal space based on the symmetry of the crystal and provides a standardized “path” connecting them. It supports various crystal structure formats (such as POSCAR and CIF) and is compatible with many electronic structure calculation software (e.g., VASP, Quantum ESPRESSO, ABINIT). A web-based interface is also available.
Script generation tools to manage large-scale computations on supercomputers and clusters. Moller is provided as part of the HTP-Tools package, designed to support high-throughput computations. It is a tool for generating batch job scripts for supercomputers and clusters, allowing parallel execution of programs under a series of computational conditions, such as parameter parallelism.
EPW (Electron-Phonon Wannier) is software for first-principles calculations of electron-phonon interactions. It utilizes Wannier functions to efficiently interpolate electronic and phononic states, enabling high-precision analysis of electron-phonon coupling and superconducting properties. EPW operates as an extension module of Quantum ESPRESSO (QE), using the results of electronic structure and phonon calculations as input.
A Boltzmann transport equation solver for calculating lattice thermal conductivity based on phonon information obtained from first-principles calculations. It takes into account three-phonon interactions and enables first-principles analysis of thermal transport properties in solids, including anisotropic crystals, complex structures, and those containing defects. Tutorials and input-support tools are also provided. A tool for calculating third-order force constants (thirdorder.py) is also available on the same website.
A Python package for extracting structural features from point cloud and image data using the mathematical framework of persistent homology. In the field of materials science, it is used to characterize structural differences between liquids and glasses, as well as for dimensionality reduction of microscope images. It is also useful for obtaining structural descriptors for machine learning.
A software package for calculating lattice thermal conductivity based on phonon information obtained from first-principles calculations, including four-phonon scattering processes. It extends ShengBTE to account for four-phonon interactions that become dominant at high temperatures. The program enables quantitative analysis of the competition between three- and four-phonon interactions as well as temperature dependence, allowing for more accurate evaluation of thermal transport properties.
A benchmark framework for evaluating general-purpose, i.e., universal, machine learning potentials, along with a leaderboard based on those evaluations. Rankings are determined by a comprehensive assessment that considers the accuracy of predicted formation energy of materials, structural relaxation, and thermal conductivity. Recently, in addition to public research institutions such as universities, major companies like Meta, Microsoft, and Google have also joined the development of universal potentials, taking top positions on the leaderboard.