Juliacon 2021 BoF: Building a Chemistry and Materials Science Ecosystem in Julia

**EDIT:** This has now been archivied. For the up to date version see https://github.com/JuliaMolSim/Juliacon2021. # Juliacon 2021 BoF: Building a Chemistry and Materials Science Ecosystem in Julia ## BoF schedule 1. Who are we? 2. Panel discussion: Structures and interfaces 3. Small group discussions 4. Wrapup ## Who are we? * [Poll link](http://etc.ch/npeb) ## Structures and interfaces * A common interface for atoms, molecules, lattices, crystals ? * Common parameters (structures, basis sets, ...) ? * Interesting external packages to integrate (ASE, Chemfiles, MolSSI, Lammps, ...) ? * Inter-code communication ? * Your points here ... ### What is needed to specify a model in a "universal" way? * AbstractStructure.jl? (maybe even this is not a good name) * What is needed? * atomic/particles positions * atomic/particles kind (not only atomic number if you want to support coarse-grained simulations, or different ions, or different isotopes, …) * periodic or not? * support different cell shapes for periodic systems (orthorhombic, triclinic, truncated octahedra?) * support partially periodic systems (e.g. periodic along x and y only) * Array of structs versus struct of arrays * Two representations and a dual interface? * existing packages: Molly.jl, JuLIP.jl, ... others ? * atoms, flexible AtomType * boundary conditions / periodicity * connectivity? * multiple layers: bonds, residues, chains, groups, … * atomic properties (pseudos, basis set, velocities, extensible) * extensible for special cases (e.g. non-spherical particles/bodies, …) * associated visualization tools? yes please * Possible consumers * visualisation * IO * follow-up simulations (e.g. MD) * 'Periodic table data' - useful to have a canonical source of e.g. atom masses (there is: [PeriodicTable.jl](https://github.com/JuliaPhysics/PeriodicTable.jl)) * DataBase of SMILES and Identifiers [ChemicalIdentifiers.jl](https://github.com/longemen3000/ChemicalIdentifiers.jl ) * Possible implementations * ASE * chemfiles * RDKit ([RDKitMinimalLib.jl](https://github.com/eloyfelix/RDKitMinimalLib.jl) provides direct access to RKit without using PyCall) * Selling point / goals: * Composability between existing packages * Modularity rather than large and complicated packages * Automatic differentiation / sensitivity analysis through the full pipeline * Rapid development: Same code structure for simplified problems and full-fletched applications * Easy and unified introspection * AbstractTrajectory.jl? * Representation of simulation time; tracking of atoms * Possible consumers: * High performance molecular-dynamics trajectory analysis might be a good 'sell' for Julia's speed. See https://www.mdanalysis.org/ for the SOTA Python package. * AbstractBasisSet.jl (also maybe not a good name? AbstractOrbitals.jl ?) * specifying basis sets, molecular orbitals, etc. in a transferable way... * https://cclib.github.io/ exists for reading (vacuum) calculations in Python * Simultaneous periodic electronic-structure (plane-waves, muffin-tin orbitals) & molecular vacuum methods (Gaussian Type Orbitals) ? * Possible consumers: * Post-HartreeFock Quantum-Chemistry methods * Quantum Monte-Carlo (molecular Hamiltonian) * Quantum chemistry by quantum computing (variational quantum eigensolvers, unitary coupled cluster etc.) * AbstractAtomicPotentials.jl (also perhaps a terrible name) interface for interatomic potentials and force fields (AbstractForces.jl / AbstractPotential.jl / name?) * getting energies and forces from underlying force field * maybe taking inspiration/making sure it is compatible with existing force field sharing methods: [OpenKIM](https://openkim.org/), [OpenFF](https://openforcefield.org/), [MMSchema](https://molssi.github.io/mmschema/) * General problems * Handling units * Julia needs new units packages which don't overuse types * This is way more general than chem/mat. sci. (agreed) #### How are you doing this sort of thing in Julia right now (if you are)? What Julia packages already implement (parts of) this functionality? Could we "pull out" that functionality to be the beginning of another, more universal package? #### How do you do it in other languages? * ASE... * pymatgen * Atomsk * ## Small group discussions * Your ideas for topics here! ## Staying in touch * [JuliaMolSim](https://juliamolsim.github.io) github org * [Join our Slack workspace](https://join.slack.com/t/juliamolsim/shared_invite/zt-tc060co0-HgiKApazzsQzBHDlQ58A7g) ## Related packages to check out I'm sure this list is not exhaustive – feel free to add to it! :smiley: ### Simulations * a bunch of great stuff over at [JuliaMolSim](https://github.com/JuliaMolSim), including * [DFTK.jl](https://github.com/JuliaMolSim/DFTK.jl) plane-wave density functional theory, * [Molly.jl](https://github.com/JuliaMolSim/Molly.jl) molecular dynamics, * [JuLIP.jl](https://github.com/JuliaMolSim/JuLIP.jl) Julia inter-atomic potentials, etc. * [JuliaChem.jl](https://github.com/davpoolechem/JuliaChem.jl): restricted Hartree-Fock models * [SKTB.jl](https://github.com/cortner/SKTB.jl): Slater-Koster style tight-binding models * [Quante.jl](https://github.com/jarvist/Quante.jl/): restricted Hartree-Fock method, based in Rick Muller's PyQuante2 Julia rewrite * [Fermi.jl](https://github.com/FermiQC/Fermi.jl): several post-HF methods * [PorousMaterials.jl](https://github.com/SimonEnsemble/PorousMaterials.jl): GCMC for adsorption in nanoporous materials * [NBodySimulator.jl](https://github.com/SciML/NBodySimulator.jl): molecular dynamics engine * [ProtoSyn.jl](https://github.com/sergio-santos-group/ProtoSyn.jl): protein design simulation (Monte Carlo, Steepest Descent, etc.) * [Rimu.jl](https://github.com/joachimbrand/Rimu.jl): projector quantum Monte Carlo (FCIQMC) currently for Hubbard models, could be extended to molecular Hamiltonians * [JuliaAtoms](http://juliaatoms.org/dev/index.html) energy structure of atoms (spherical symmetry), so far only HF; WIP ### Numerics * [Lebedev.jl](https://github.com/stefabat/Lebedev.jl) Lebedev quadrature rules for surface integrals * [Lints.jl](https://github.com/stefabat/Lebedev.jl): Julia interface to Libint2 * [MolecularIntegrals.jl](https://github.com/rpmuller/MolecularIntegrals.jl): variety of one- and two-electron integrals * [SciML](https://github.com/SciML/): Fast solvers for differential equations, solving nonlinear equations, quadrature, etc. ### Others * [Chemellia](https://github.com/Chemellia): packages for machine learning with atoms * [MolecularGraph.jl](https://github.com/mojaie/MolecularGraph.jl) * [ReactionMechanismSimulator.jl](https://github.com/ReactionMechanismGenerator/ReactionMechanismSimulator.jl): Package for simulating chemical kinetic mechanisms and reactors. * [Clapeyron.jl](https://github.com/ypaul21/Clapeyron.jl): Package for solving equations of state for thermodynamic properties. * [ACEsuit](https://acesuit.github.io) : atomic cluster expansion, ML for interatomic potentials and force fields * [Chemfiles](https://github.com/chemfiles/Chemfiles.jl) Read and write files used in computational chemistry. This package contains binding to the corresponding C++ library * [Xtals.jl](https://github.com/SimonEnsemble/Xtals.jl) read/write structure files (crystal-focused), generate bond graphs * [RDKitMinimalLib.jl](https://github.com/eloyfelix/RDKitMinimalLib.jl) Unofficial wraps for RDKit. With RDKit you can do everything.