> ## Documentation Index
> Fetch the complete documentation index at: https://docs.biohub.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Analyze TCR data with TCR-Workbench

> Use the tcr-workbench Skill to choose, run, and interpret local TCR-Workbench analyses of processed TCR and peptide–MHC data.

Use `tcr-workbench` to run TCR-Workbench analyses of processed human or mouse alpha-beta TCR data on your own computer.
The Skill picks the workflow that matches the quantity you ask for, runs the `python3 tcr.py` launcher, and explains what the outputs do and do not show.

> [!NOTE]
>
> This Skill calls no Biohub tools.
> Your AI client runs the TCR-Workbench launcher in a local checkout of [TCR-Workbench](https://github.com/akdslab/tcr-workbench), so it needs a shell on your own computer.
> Model workflows also need a completed setup, which the [`tcr-workbench-setup`](https://docs.biohub.ai/biohub-mcp/skills/tcr-workbench-setup) Skill covers.

## Before you begin

* Connect Biohub MCP to a client that can run commands on your computer; see [Connect your client](https://docs.biohub.ai/biohub-mcp/connect-clients.md).
* Review the commands and downloads before approving them; see [Skills that run commands on your computer](https://docs.biohub.ai/biohub-mcp/security.md#skills-that-run-commands-on-your-computer).

## Run the Skill

Score peptides for an HLA allele

Using the Biohub MCP tcr-workbench Skill, score the peptides in peptides.csv against HLA-A\*02:01 with DecoderTCR and tell me which ones rank highest.

## What the Skill guides

1. It checks the local setup with `python3 tcr.py doctor` and hands off to `tcr-workbench-setup` when the environment is missing or lacks what the analysis needs.
2. It chooses the workflow from the requested quantity, such as `pmhc-score` for peptide scores in an MHC context, `tcr-score` for a paired receptor, `repertoire-score` for panel screening, or `screen` for curated reference evidence without model weights.
3. It ingests Cell Ranger V(D)J, AIRR, or paired receptor tables, using the supplied V/J genes and anchored junctions without repairing sequences or inventing cell pairing.
4. It keeps species, MHC class II alpha/beta pairs, model size, device, and precision explicit, and does not switch them silently.
5. It writes each analysis to a new output folder and inspects the result tables, unresolved reasons, QC, and manifests, not only the HTML report.

## Expected result

The result includes:

* An output folder with an offline `report.html`, the complete result tables, and manifests that record the model, device, precision, and file hashes.
* Scored and unresolved counts, with the reason each unresolved row could not be scored.
* For score reports, the top-ranked peptides per receptor, MHC, and peptide length, compared with actually scored random peptides.

> [!WARNING]
>
> DecoderTCR scores are uncalibrated model outputs, not binding probabilities.
> `Unresolved` means insufficient support, not non-binding, and random reference peptides are not known nonbinders.

## Skill files

| Resource | URI |
| - | - |
| Instructions | `skill://tcr-workbench/SKILL.md` |
| File list | `skill://tcr-workbench/_manifest` |
| License (MIT) | `skill://tcr-workbench/LICENSE.md` |

## Related documentation

* [Biohub MCP Skills](https://docs.biohub.ai/biohub-mcp/skills/overview.md)
* [Set up TCR-Workbench](https://docs.biohub.ai/biohub-mcp/skills/tcr-workbench-setup.md)
* [TCR-Workbench on GitHub](https://github.com/akdslab/tcr-workbench)


## Related topics

- [Set up TCR-Workbench](https://docs.biohub.ai/biohub-mcp/skills/tcr-workbench-setup.md)
- [Biohub MCP Skills](https://docs.biohub.ai/biohub-mcp/skills/overview.md)
- [ESM Atlas](https://docs.biohub.ai/learn/guides/esm-atlas.md)
- [Reconstruct microscopy data with WaveOrder](https://docs.biohub.ai/biohub-mcp/skills/waveorder-tutorial.md)
- [Biohub MCP security, privacy, and data handling](https://docs.biohub.ai/biohub-mcp/security.md)