{"id":29625212,"url":"https://github.com/novartis/jaeger","last_synced_at":"2025-07-21T06:07:39.850Z","repository":{"id":146943385,"uuid":"426423154","full_name":"Novartis/JAEGER","owner":"Novartis","description":"JAEGER is a deep generative approach for small-molecule design","archived":false,"fork":false,"pushed_at":"2021-12-21T01:02:17.000Z","size":3925,"stargazers_count":27,"open_issues_count":0,"forks_count":6,"subscribers_count":5,"default_branch":"main","last_synced_at":"2025-01-18T06:33:09.781Z","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":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Novartis.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.txt","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":"2021-11-09T23:48:06.000Z","updated_at":"2024-11-05T04:26:49.000Z","dependencies_parsed_at":"2023-04-01T11:32:34.291Z","dependency_job_id":null,"html_url":"https://github.com/Novartis/JAEGER","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/Novartis/JAEGER","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Novartis%2FJAEGER","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Novartis%2FJAEGER/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Novartis%2FJAEGER/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Novartis%2FJAEGER/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Novartis","download_url":"https://codeload.github.com/Novartis/JAEGER/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Novartis%2FJAEGER/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":266248501,"owners_count":23899056,"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":"2025-07-21T06:07:39.337Z","updated_at":"2025-07-21T06:07:39.836Z","avatar_url":"https://github.com/Novartis.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# JAEGER\n\n\n**J**T-V**AE** **G**en**er**ative Modeling (**JAEGER**) is a deep\ngenerative approach for small-molecule design. JAEGER is based on the\nJunction-Tree Variational Auto-Encoder (JT-VAE) method [1], which\nensures chemical validity for the generated molecules.\n\nJAEGER is trained on existing molecules associated with activity\nvalues measured in a given assay. During training, JAEGER learns how\nto map each molecule onto a (high-dimensional) coordinate space, often\nreferred to as the **latent space**. JAEGER also learns how to\n**decode** a coordinate position in the latent space back to a\nmolecule.\n\nTo generate new molecules, JAEGER defines numerical search strategies\nto efficiently and effectively explore that latent space. JAEGER\ncouples the exploration of that latent space together with activity\npredictive models to discover and optimize novel active molecules.\n\n[1] Wengong Jin, Regina Barzilay, Tommi S. Jaakkola: Junction Tree\nVariational Autoencoder for Molecular Graph Generation. ICML 2018:\n2328-2337\n\n## Contents\n\n- `src`: Python source code for JAEGER\n- `models`: Data for building demo model\n\n## System requirements\n\n### Hardware\n#### GPUs\n\nWe have tested JAEGER on machines with the following GPUs:\n\n- NVIDIA Tesla K80 \n- NVIDIA Tesla V100\n\n### Software\n\n#### Operating systems\n\nWe have tested JAEGER on machines with the following systems:\n\n- Red Hat Enterprise Linux 6\n- CentOS Linux 7\n\n\n#### Sofware dependencies\n\n- python 3.8.6\n- pandas 1.1.5\n- numpy 1.19.5\n- pyjanitor 0.20.10\n- pytorch 1.7.0\n- rdkit 2020.09.3\n- scikit-learn 0.24.0\n- streamlit 0.74.1\n\n## Installation\n\n* Install the python libraries mentioned in **Software dependencies**\n  above into your python environment.\n\n* Get the `jaeger` branch from the JT-VAE JAEGER fork located\n  [here](https://github.com/PsiGamma/icml18-jtnn/tree/jaeger).\n  Include the `icml18-jtnn` **and** the `icml18-jtnn/jtnn` directories\n  in your python path.\n\n* Get a copy of the JAEGER repo (this repo). Include the `src`\n  directory in your python path\n  \nInstallation time of the software dependencies will vary depending on\nyour computational environment. The larger packages like `pytorch` can\ntake a couple of hours.\n\n  \n## Demo dataset\n\nWe include a demo dataset with all molecules with measured 3D7\ninhibition activity from the\n[Deposited Set 2: Novartis GNF Whole Cell Dataset](https://chembl.gitbook.io/chembl-ntd/downloads/deposited-set-2-novartis-gnf-whole-cell-dataset-20th-may-2010)\nhosted at the\n[ChEMBL - Neglected Tropical Disease archive](https://chembl.gitbook.io/chembl-ntd/). The\ndemo dataset is located at `./models/training_data/Novartis_GNF.csv`.\n\n## Training a model\n\nSee TRAINING.md\n\n## Generating molecules\n\nSee GENCHEM.md\n\n## License\n\nCopyright 2021 Novartis Institutes for BioMedical Research Inc.\n\nLicensed under the Apache License, Version 2.0 (the \"License\");\nyou may not use this file except in compliance with the License.\nYou may obtain a copy of the License at\n\n    http://www.apache.org/licenses/LICENSE-2.0\n\nUnless required by applicable law or agreed to in writing, software\ndistributed under the License is distributed on an \"AS IS\" BASIS,\nWITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\nSee the License for the specific language governing permissions and\nlimitations under the License.\n\n\nSee LICENSE.txt\n\n## Contact\n\nwilliam_jose.godinez_navarro@novartis.com\n\n## Citation\n\nPlease cite the code as follows:\n\nGodinez, W. J. \u0026 Ma, E. J. Novartis/JAEGER: Public. Zenodo, doi:https://doi.org/10.5281/zenodo.5794429 (2021).\n\n\n## DOI\n\n[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.5794429.svg)](https://doi.org/10.5281/zenodo.5794429)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnovartis%2Fjaeger","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnovartis%2Fjaeger","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnovartis%2Fjaeger/lists"}