{"id":22096206,"url":"https://github.com/baaihealth/OpenComplex","last_synced_at":"2025-07-24T22:31:50.757Z","repository":{"id":65576515,"uuid":"584707126","full_name":"baaihealth/OpenComplex","owner":"baaihealth","description":"Trainable PyTorch framework for developing protein, RNA and complex models.","archived":false,"fork":false,"pushed_at":"2024-11-13T04:16:27.000Z","size":15304,"stargazers_count":264,"open_issues_count":2,"forks_count":33,"subscribers_count":17,"default_branch":"master","last_synced_at":"2024-12-01T04:09:56.868Z","etag":null,"topics":["alphafold2","protein-structure","pytorch","rna-structure"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/baaihealth.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":"2023-01-03T10:01:36.000Z","updated_at":"2024-11-20T10:12:53.000Z","dependencies_parsed_at":"2024-11-13T05:30:17.019Z","dependency_job_id":null,"html_url":"https://github.com/baaihealth/OpenComplex","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/baaihealth/OpenComplex","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/baaihealth%2FOpenComplex","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/baaihealth%2FOpenComplex/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/baaihealth%2FOpenComplex/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/baaihealth%2FOpenComplex/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/baaihealth","download_url":"https://codeload.github.com/baaihealth/OpenComplex/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/baaihealth%2FOpenComplex/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":266913677,"owners_count":24005578,"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","status":"online","status_checked_at":"2025-07-24T02:00:09.469Z","response_time":99,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["alphafold2","protein-structure","pytorch","rna-structure"],"created_at":"2024-12-01T04:09:52.244Z","updated_at":"2025-07-24T22:31:49.054Z","avatar_url":"https://github.com/baaihealth.png","language":"Python","funding_links":[],"categories":["Based on the classification of biomolecules"],"sub_categories":["RNA-Protein ComplexFold"],"readme":"![header](img/logo.png)\n# OpenComplex\nOpenComplex is an open-source platform for developing protein and RNA complex models.  \nBased on DeepMind's [Alphafold 2](https://github.com/deepmind/alphafold) and AQ Laboratory's [OpenFold](https://github.com/aqlaboratory/openfold), OpenComplex support almost all features from Alphafold 2 and OpenFold, and introduces the following new features:\n* Reimplemented Alphafold-Multimer models.\n* RNA and protein-RNA complex models with high precision.\n* Kernel fusion and optimization on \u003e=Ampere GPUs, brings 16% \n\n![Figure 1. OpenComplex inference result of RNA and protein-RNA complex.](img/cases.png)\n\nWe will release training results and pretrained parameters soon.\n\n## Installation (Linux)\n\nAll Python dependencies are specified in `environment.yml`. For producing sequence \nalignments, you'll also need `kalign`, the [HH-suite](https://github.com/soedinglab/hh-suite), \nand one of {`jackhmmer`, [MMseqs2](https://github.com/soedinglab/mmseqs2) (nightly build)} \ninstalled on on your system.\nFinally, some download scripts require `aria2c` and `aws`.\n\nFor convenience, we provide a script that installs Miniconda locally, creates a \n`conda` virtual environment, installs all Python dependencies, and downloads\nuseful resources, including both sets of model parameters. Run:\n\n```bash\nscripts/install_third_party_dependencies.sh\n```\n\nTo activate the environment, run:\n\n```bash\nsource scripts/activate_conda_env.sh\n```\n\nWith the environment active, compile CUDA kernels with\n\n```bash\npython3 setup.py install\n```\n\nTo install the HH-suite to `/usr/bin`, run\n\n```bash\nscripts/install_hh_suite.sh\n```\n\n## Usage\n\n### Data preparation\n\nTo run feature generation pipeline from `.fasta` to `feature.pkl` on DeepMind's MSA and template database, run e.g.:\n```bash\npython ./scripts/extract_pkl_from_fas.py ./example_data/fasta/ ./example_data/features/\n```\nwhere `example_data` is the directory containing example fasta . If `jackhmmer`, \n`hhblits`, `hhsearch` and `kalign` are available at the default path of \n`/usr/bin`, their `binary_path` command-line arguments can be dropped.\nIf you've already computed alignments for the query, you have the option to \nskip the expensive alignment computation here with \n`--use_precomputed_alignments`.\n\n### Train and Inference\n\nSee example bash scripts in example_data/scripts\n\n## Testing\n\nTo run unit tests, use\n\n```bash\nscripts/run_unit_tests.sh\n```\n\nThe script is a thin wrapper around Python's `unittest` suite, and recognizes\n`unittest` arguments. E.g., to run a specific test verbosely:\n\n```bash\nscripts/run_unit_tests.sh -v tests.test_model\n```\n\nCertain tests require that AlphaFold (v2.0.1) be installed in the same Python\nenvironment. These run components of AlphaFold and OpenFold side by side and\nensure that output activations are adequately similar. For most modules, we\ntarget a maximum pointwise difference of `1e-4`.\n\n## Citation\n\nIf you find our open-sourced code \u0026 models helpful to your research, please also consider star🌟 and cite📑 this repo. Thank you for your support!\n```\n@misc{OpenComplex_code,\n  author={Jingcheng, Yu and Zhaoming, Chen and Zhaoqun, Li and Mingliang, Zeng and Wenjun, Lin and He, Huang and Qiwei, Ye},\n  title={Code of OpenComplex},\n  year={2022},\n  howpublished = {\\url{https://github.com/baaihealth/OpenComplex}}\n}\n```\nIt is recommended to also cite OpenFold and AlphaFold.\n\n\n## License and Disclaimer\n\nCopyright 2022 BAAI.\n\nExtended from AlphaFold and OpenFold, OpenComplex is licensed under\nthe permissive Apache Licence, Version 2.0.\n\n## Contributing\n\nIf you encounter problems using OpenComplex, feel free to create an issue! We also\nwelcome pull requests from the community.\n\n## Contact Information\nFor help or issues using the repos, please submit a GitHub issue.\n\nFor other communications, please contact Qiwei Ye (qwye@baai.ac.cn).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbaaihealth%2FOpenComplex","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbaaihealth%2FOpenComplex","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbaaihealth%2FOpenComplex/lists"}