Final - Modeling Breast Cancer Subtypes + TABPFN

Run Workflow in Galaxy

• Galaxy France
• Galaxy Europe
• Galaxy Australia
• Galaxy Main
• Import to another server (≥23.0+ only!)

flowchart TD
  0["ℹ️ Input Dataset\nclin_train"];
  style 0 stroke:#2c3143,stroke-width:4px;
  1["ℹ️ Input Dataset\nclin_test"];
  style 1 stroke:#2c3143,stroke-width:4px;
  2["ℹ️ Input Dataset\ngex_train"];
  style 2 stroke:#2c3143,stroke-width:4px;
  3["ℹ️ Input Dataset\ngex_test"];
  style 3 stroke:#2c3143,stroke-width:4px;
  4["ℹ️ Input Dataset\ncna_train"];
  style 4 stroke:#2c3143,stroke-width:4px;
  5["ℹ️ Input Dataset\ncna_test"];
  style 5 stroke:#2c3143,stroke-width:4px;
  6["ℹ️ Input Parameter\nnumber of features to keep"];
  style 6 fill:#ded,stroke:#393,stroke-width:4px;
  317be0d5-bbba-4f29-95b3-d7c58c2736b3["Output\n"];
  6 --> 317be0d5-bbba-4f29-95b3-d7c58c2736b3;
  style 317be0d5-bbba-4f29-95b3-d7c58c2736b3 stroke:#2c3143,stroke-width:4px;
  7["ℹ️ Input Parameter\nPerforme TABPFN?"];
  style 7 fill:#ded,stroke:#393,stroke-width:4px;
  865c8339-1baa-42cd-a096-a5f159ae5221["Output\n"];
  7 --> 865c8339-1baa-42cd-a096-a5f159ae5221;
  style 865c8339-1baa-42cd-a096-a5f159ae5221 stroke:#2c3143,stroke-width:4px;
  8[" Flexynesis niter1"];
  5 -->|output| 8;
  4 -->|output| 8;
  1 -->|output| 8;
  3 -->|output| 8;
  0 -->|output| 8;
  2 -->|output| 8;
  9[" Flexynesis niter5"];
  5 -->|output| 9;
  4 -->|output| 9;
  1 -->|output| 9;
  3 -->|output| 9;
  0 -->|output| 9;
  2 -->|output| 9;
  b0c3401f-ab62-49f2-a314-30a4ec36155c["Output\nflexynesis output"];
  9 --> b0c3401f-ab62-49f2-a314-30a4ec36155c;
  style b0c3401f-ab62-49f2-a314-30a4ec36155c stroke:#2c3143,stroke-width:4px;
  6d26fdb7-6fb0-403d-a484-7b0abdb5bd67["Output\nembedding plots pca"];
  9 --> 6d26fdb7-6fb0-403d-a484-7b0abdb5bd67;
  style 6d26fdb7-6fb0-403d-a484-7b0abdb5bd67 stroke:#2c3143,stroke-width:4px;
  10["🛠️ Subworkflow\nPrepare data for TABPFN"];
  style 10 fill:#edd,stroke:#900,stroke-width:4px;
  6 -->|output| 10;
  1 -->|output| 10;
  3 -->|output| 10;
  0 -->|output| 10;
  2 -->|output| 10;
  7 -->|output| 10;
  b93c8cbe-538d-4875-96b6-2073943f3f65["Output\n"];
  10 --> b93c8cbe-538d-4875-96b6-2073943f3f65;
  style b93c8cbe-538d-4875-96b6-2073943f3f65 stroke:#2c3143,stroke-width:4px;
  11["🛠️ Subworkflow\nPrepare data for TABPFN"];
  style 11 fill:#edd,stroke:#900,stroke-width:4px;
  6 -->|output| 11;
  1 -->|output| 11;
  5 -->|output| 11;
  0 -->|output| 11;
  4 -->|output| 11;
  7 -->|output| 11;
  8b726af8-7712-453e-b987-2e549248089c["Output\n"];
  11 --> 8b726af8-7712-453e-b987-2e549248089c;
  style 8b726af8-7712-453e-b987-2e549248089c stroke:#2c3143,stroke-width:4px;
  12["Extract prediction table - niter1"];
  8 -->|results| 12;
  13["Extract test embeddings - niter1"];
  8 -->|results| 13;
  14["Extract prediction table - niter5"];
  9 -->|results| 14;
  15["Extract test embeddings - niter5"];
  9 -->|results| 15;
  16["Pick parameter value"];
  10 -->|ready_train| 16;
  17["Pick parameter value"];
  10 -->|ready_test| 17;
  18["Pick parameter value"];
  11 -->|ready_train| 18;
  19["Pick parameter value"];
  11 -->|ready_test| 19;
  20["Sort"];
  12 -->|output| 20;
  21["Sort"];
  14 -->|output| 21;
  22["PR cruve plot - flexynesis"];
  14 -->|output| 22;
  c64e7cb2-2e54-4176-a914-b5b291c3f8da["Output\nPR curve - Flexynesis"];
  22 --> c64e7cb2-2e54-4176-a914-b5b291c3f8da;
  style c64e7cb2-2e54-4176-a914-b5b291c3f8da stroke:#2c3143,stroke-width:4px;
  23["TABPFN on gex data 2"];
  17 -->|data_param| 23;
  16 -->|data_param| 23;
  cc9c63af-9aca-4030-9e5a-f196d30c849d["Output\nTABPFN predictions - gex"];
  23 --> cc9c63af-9aca-4030-9e5a-f196d30c849d;
  style cc9c63af-9aca-4030-9e5a-f196d30c849d stroke:#2c3143,stroke-width:4px;
  5f9a68f4-fa2c-400e-b78b-8b2bba3c7b91["Output\nPR curve - TABPFN - gex"];
  23 --> 5f9a68f4-fa2c-400e-b78b-8b2bba3c7b91;
  style 5f9a68f4-fa2c-400e-b78b-8b2bba3c7b91 stroke:#2c3143,stroke-width:4px;
  24["TABPFN on cna data 2"];
  19 -->|data_param| 24;
  18 -->|data_param| 24;
  229fd7b2-f8b8-4257-919c-550c2154c689["Output\nPR curve - TABPFN - cna"];
  24 --> 229fd7b2-f8b8-4257-919c-550c2154c689;
  style 229fd7b2-f8b8-4257-919c-550c2154c689 stroke:#2c3143,stroke-width:4px;
  e99d56b0-8486-4344-9fdc-78ac8f92402e["Output\nTABPFN predictions - cna"];
  24 --> e99d56b0-8486-4344-9fdc-78ac8f92402e;
  style e99d56b0-8486-4344-9fdc-78ac8f92402e stroke:#2c3143,stroke-width:4px;
  25["Flexynesis plot"];
  13 -->|output| 25;
  20 -->|outfile| 25;
  1bd99aab-3844-4496-9597-bb7e49258f66["Output\nembedding plot umap - niter1"];
  25 --> 1bd99aab-3844-4496-9597-bb7e49258f66;
  style 1bd99aab-3844-4496-9597-bb7e49258f66 stroke:#2c3143,stroke-width:4px;
  26["UMAP plot on test embeddings"];
  15 -->|output| 26;
  21 -->|outfile| 26;
  69108048-9068-4c4f-b8bf-b422b0d31625["Output\nembedding plot umap - niter5"];
  26 --> 69108048-9068-4c4f-b8bf-b422b0d31625;
  style 69108048-9068-4c4f-b8bf-b422b0d31625 stroke:#2c3143,stroke-width:4px;

Inputs

Outputs

Tools

To use these workflows in Galaxy you can either click the links to download the workflows, or you can right-click and copy the link to the workflow which can be used in the Galaxy form to import workflows.

Importing into Galaxy

Hands On: Importing a workflow

1. Click on galaxy-workflows-activity Workflows in the Galaxy activity bar (on the left side of the screen, or in the top menu bar of older Galaxy instances). You will see a list of all your workflows
2. Click on galaxy-upload Import at the top-right of the screen
3. Provide your workflow
   • Option 1: Paste the URL of the workflow into the box labelled “Archived Workflow URL”
   • Option 2: Upload the workflow file in the box labelled “Archived Workflow File”
4. Click the Import workflow button

Below is a short video demonstrating how to import a workflow from GitHub using this procedure:

Version History

For Admins

Installing the workflow tools

wget https://training.galaxyproject.org/training-material/topics/statistics/tutorials/flexynesis_classification/workflows/main_workflow.ga -O workflow.ga
workflow-to-tools -w workflow.ga -o tools.yaml
shed-tools install -g GALAXY -a API_KEY -t tools.yaml
workflow-install -g GALAXY -a API_KEY -w workflow.ga --publish-workflows