{"id":29023318,"url":"https://github.com/rish-16/amber14-ff-openmm-gpu","last_synced_at":"2025-06-26T03:06:04.897Z","repository":{"id":261838745,"uuid":"885464261","full_name":"rish-16/amber14-ff-openmm-gpu","owner":"rish-16","description":null,"archived":false,"fork":false,"pushed_at":"2025-01-28T17:42:24.000Z","size":87325,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-01-28T18:32:55.112Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/rish-16.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-11-08T16:26:38.000Z","updated_at":"2025-01-28T17:42:28.000Z","dependencies_parsed_at":"2024-11-08T18:50:52.529Z","dependency_job_id":null,"html_url":"https://github.com/rish-16/amber14-ff-openmm-gpu","commit_stats":null,"previous_names":["rish-16/amber14-ff-openmm-gpu"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/rish-16/amber14-ff-openmm-gpu","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rish-16%2Famber14-ff-openmm-gpu","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rish-16%2Famber14-ff-openmm-gpu/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rish-16%2Famber14-ff-openmm-gpu/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rish-16%2Famber14-ff-openmm-gpu/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/rish-16","download_url":"https://codeload.github.com/rish-16/amber14-ff-openmm-gpu/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rish-16%2Famber14-ff-openmm-gpu/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":261990350,"owners_count":23241190,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-06-26T03:06:04.170Z","updated_at":"2025-06-26T03:06:04.879Z","avatar_url":"https://github.com/rish-16.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# OpenMM AMBER14 on GPU 🧬🏎️\n\nThis repo allows you to run the AMBER14 forcefield on a pre-run MD simulation via OpenMM without having to run _more_ MD. It extracts vector information from the trajectory, stores them in an OpenMM-friendly data format, and the OpenMM simulator parses these coordinates to give frame-wise energies and forces. \n\n## Installation and Setup\n\nRun the following to download the CUDA-accelerated version of OpenMM. Take note of your CUDA version (you can find out using `nvidia-smi`) and make appropriate changes when downloading `cudatoolkit`.\n\n```bash\nconda create -n amber14 python=3.6 -y\nconda activate amber14\nconda install -c conda-forge cudatoolkit=12.5 -y\nconda install -c conda-forge openmm mdanalysis -y\n```\n\nCheck whether OpenMM can access CUDA:\n```bash\npython -m openmm.testInstallation\n```\n\nYou should get benchmarking results that look like this:\n\n```bash\nOpenMM Version: 7.6\nGit Revision: ad113a0cb37991a2de67a08026cf3b91616bafbe\n\nThere are 4 Platforms available:\n\n1 Reference - Successfully computed forces\n2 CPU - Successfully computed forces\n3 CUDA - Successfully computed forces\n1 warning generated.\n1 warning generated.\n1 warning generated.\n1 warning generated.\n4 OpenCL - Successfully computed forces\n\nMedian difference in forces between platforms:\n\nReference vs. CPU: 6.29276e-06\nReference vs. CUDA: 6.73166e-06\nCPU vs. CUDA: 7.39089e-07\nReference vs. OpenCL: 6.74399e-06\nCPU vs. OpenCL: 7.80542e-07\nCUDA vs. OpenCL: 2.19129e-07\n\nAll differences are within tolerance.\n```\n\n## Running OpenMM AMBER14\n\nThe complete source code is in `openmm_amber14.py`. They take in a crystal structure (`.pdb`) and a GROMACS trajectory (`.xtc`) for the protein obtained from MD, and compute frame-wise energies and forces. You can save them as `.npy` tensors downstream. If you have a CUDA GPU, OpenMM will automatically detect it and run the forcefield on it.\n\nRelevant demo files for protein `16pk` are included to run the code. It has 400+ residues and the code is scalable over sequence lengths.\n\n\u003e Take note that this code can only extract energies and forces from conformations that are within a trajectory. Pointwise evaluations of new conformations are not possible and may require some form of relaxation (WIP). \n\n## License\n[MIT](https://github.com/rish-16/amber14-ff-openmm-gpu/blob/main/LICENSE)","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frish-16%2Famber14-ff-openmm-gpu","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Frish-16%2Famber14-ff-openmm-gpu","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frish-16%2Famber14-ff-openmm-gpu/lists"}