{"id":30019627,"url":"https://github.com/veupathdb/expression-shepherd","last_synced_at":"2026-02-09T05:03:12.823Z","repository":{"id":272764205,"uuid":"917402876","full_name":"VEuPathDB/expression-shepherd","owner":"VEuPathDB","description":"Wrangling our expression data with gpt-4o","archived":false,"fork":false,"pushed_at":"2025-11-19T12:59:05.000Z","size":2238,"stargazers_count":0,"open_issues_count":3,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-11-19T14:09:27.202Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"HTML","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/VEuPathDB.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-01-15T23:06:40.000Z","updated_at":"2025-10-18T12:28:13.000Z","dependencies_parsed_at":"2025-04-11T17:21:26.246Z","dependency_job_id":"dcef5de8-7340-44e9-905a-ab66f583955c","html_url":"https://github.com/VEuPathDB/expression-shepherd","commit_stats":null,"previous_names":["veupathdb/expression-wrangler","veupathdb/expression-shepherd"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/VEuPathDB/expression-shepherd","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VEuPathDB%2Fexpression-shepherd","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VEuPathDB%2Fexpression-shepherd/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VEuPathDB%2Fexpression-shepherd/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VEuPathDB%2Fexpression-shepherd/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/VEuPathDB","download_url":"https://codeload.github.com/VEuPathDB/expression-shepherd/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VEuPathDB%2Fexpression-shepherd/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29257514,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-09T04:11:57.159Z","status":"ssl_error","status_checked_at":"2026-02-09T04:11:56.117Z","response_time":56,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":[],"created_at":"2025-08-06T01:20:08.308Z","updated_at":"2026-02-09T05:03:12.818Z","avatar_url":"https://github.com/VEuPathDB.png","language":"HTML","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Expression Shepherd\n\nThis is a lightweight proof-of-concept tool for summarising the expression data on a gene page from VEuPathDB databases using OpenAI's GPT-4o model or Anthropic's Claude 4 Sonnet.\n\n## Non-Docker usage\n\n### 0. Requirements\n\n- API key for your chosen model:\n  - OpenAI: add `OPENAI_API_KEY=xxxxxxxxxxxxx` to a file called `.env` in this directory\n  - Anthropic: add `ANTHROPIC_API_KEY=xxxxxxxxxxxxx` to a file called `.env` in this directory\n- volta if possible: https://docs.volta.sh/guide/getting-started\n  - it takes care of your node and yarn versions\n- node - 18.20.5 tested (higher versions will likely work) \n- yarn - 1.22.19 tested\n\n### 1. Initialise the Node.js environment\n\n```bash\nyarn\n```\n\nIf you have `volta` installed, this will also make sure you have the right versions of `node` and `yarn`, otherwise you'll need to install those manually if you run into any issues.\n\n### 2. Compile the TypeScript\n\n```bash\nyarn build\n```\n\nThis compiles `src/main.ts` into `dist/main.js`\n\n### 3. Run the code\n\nYou can run the script with any gene ID from supported VEuPathDB databases:\n\n**With OpenAI GPT-4o (default):**\n```bash\nnode dist/main.js PlasmoDB PF3D7_1016300\n```\n\n**With Claude 4 Sonnet:**\n```bash\nnode dist/main.js PlasmoDB PF3D7_1016300 --claude\n```\n\n**Supported databases:** PlasmoDB, VectorBase, ToxoDB, CryptoDB, FungiDB, GiardiaDB, TrichDB, AmoebaDB, MicrosporidiaDB, PiroplasmaDB, TriTrypDB\n\n**Note:** Use `node dist/main.js` directly instead of `yarn start` when using the `--claude` flag, as npm/yarn scripts don't pass through additional arguments.\n\nIt will output three files in the `example-output` directory:\n\n1. `GENE_ID.01.MODEL.summaries.json` - the per experiment AI summaries (JSON)\n2. `GENE_ID.01.MODEL.summary.json` - the AI summary-of-summaries and grouping (JSON)  \n3. `GENE_ID.01.MODEL.summary.html` - a nice HTML version of the summary\n\nWhere `MODEL` is either `OpenAI` or `Claude` depending on which API you used.\n\nTo view the HTML open it as a local file in your web browser (Ctrl-O usually).\n\nYou can commit any generated files to the repo if you like (within reason)!\n\n## Docker usage\n### 0. Requirements\n\n- OpenAI API key\n  - add `OPENAI_API_KEY=xxxxxxxxxxxxx` to a file called `.env` in this directory\n- [Docker](https://www.docker.com/) installed on your system\n\n### 1. Build the Docker image\n\nTo build the Docker image, use the following command:\n\n```bash\ndocker build -t expression-shepherd .\n```\n\n### 2. Run the container\n\nTo start a container from the image and get a shell.\n\nThe command below \"mounts\" ./example-output inside the container so any outputs will be seen in the host filesystem too.\n\n```bash\ndocker run -d --rm --env-file .env -v $(pwd)/example-output:/app/example-output expression-shepherd sh\n```\nThe container will be removed when you exit the shell. (But not the image.)\n\n### 3. Run the code\n\nIf the container is already running but you need a new shell:\n\n```bash\ndocker ps\n# find the CONTAINER_ID\ndocker exec -it --env-file .env \u003cCONTAINER_ID\u003e sh\n```\n\nYou can then manually run the script (see step 3. in the non-Docker section above):\n\n```bash\nnode dist/main.js PlasmoDB PF3D7_0818900\n```\n\nOr you can just run the script at container launch time: \n\n```bash\ndocker run -d --rm --env-file .env -v $(pwd)/example-output:/app/example-output expression-shepherd node dist/main.js PlasmoDB PF3D7_0818900\n```\n\nOr like this in an already running container:\n\n```bash\ndocker exec -it --env-file .env \u003cCONTAINER_ID\u003e node dist/main.js PlasmoDB PF3D7_0818900\n```\n\nNote that volta is not available in the node container but it does have suitable versions of node and yarn installed anyway.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fveupathdb%2Fexpression-shepherd","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fveupathdb%2Fexpression-shepherd","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fveupathdb%2Fexpression-shepherd/lists"}