Agent skill · Testing & QA

Gene Panel Selection Workflow

End-to-end workflow for gene panel design in scRNA-seq and spatial transcriptomics, that should be **STRICTLY** followed: dataset understanding + smart downsampling + train/test splits, algorithmic selection (HVG/DE/RF/scGeneFit/SpaPROS), optimal sub-panel discovery (ARI vs size), biological completion with a stability gate (Completion Rule), consensus scoring and completion (only if there is still room), and benchmarking on test splits (ARI/NMI/Silhouette + UMAP similarity).

BioTender-maxgithub.com/BioTender-maxGitHub ↗
claude-codeNOASSERTION
Install
npx skills add BioTender-max/awesome-bio-agent-skills --skill gene_panel_selection --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 1
SKILL.md size: 28 KB
Bundled scripts: none
Path: skills/pantheon/gene_panel_selection/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 135
Language: Python

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

Review
written from the skill's own SKILL.md · Aug 5, 2026

What it does

Aimed at constructing biologically meaningful and robust gene panels for scRNA-seq and spatial transcriptomics by guiding the agent through a mandatory, stage-based workflow. It emphasizes dataset understanding, smart downsampling, train/test splits, algorithmic selection (HVG/DE/RF/scGeneFit/SpaPROS), optimal sub-panel discovery (ARI vs size), biological completion with a stability gate, consensus scoring with potential completion, and benchmarking on test splits (ARI/NMI/Silhouette + UMAP similarity). The workflow is enforced to follow clearly defined steps with mandatory sub-steps and requires reporting at the end.

How it works

  • Determine the current workflow stage (Steps 1–5) required for the task and strictly execute the corresponding steps with all mandatory sub-steps.
  • Operate in the workdir provided by the leader agent.
  • Use call_agent(agent_name, instruction) to interact with other agents (browser_use for data collection, system_manager for software installation, biologist for results interpretation, reporter for final PDF).
  • Use observe_images to inspect figures and replot if necessary.
  • Produce a final markdown report named report_analysis.md containing Summary, Data, Results, Key findings, and Next steps.
  • If the dataset is large, perform smart downsampling while preserving all cell types.
  • Include hyperparameters from the provided constants for gene panel methods and ensure downsampling and splitting follow the specified limits (e.g., DOWNSAMPLE_MAX_CELLS, GENE_COUNT_THRESHOLD, N_TRAINING_SPLITS, N_TEST_SPLITS, SPLIT_CELL_LIMIT).
  • Implement the listed algorithmic methods using either Scanpy (HVG/DE) or the helper script scripts/gene_panel_helpers.py for RF, scGeneFit, and SpaPROS, loading the helper from the skill directory.

When to use it

  • Triggered when constructing a gene panel for scRNA-seq or spatial transcriptomics that requires end-to-end workflow enforcement including dataset understanding, downsampling, train/test split, algorithmic selection, and benchmarking as described in the workflow.

What it can touch

  • The workflow relies on tools and scripts specified in the hyperparameters and helper scripts: scripts/gene_panel_helpers.py (functions: select_spapros, select_random_forest, select_scgenefit, estimate_spapros_runtime) and the helper script loaded from the skill directory. It uses Clinic-style file I/O via file_manager to save datasets and intermediate artifacts and may invoke Observations/UMAP plots through the agent interactions.

Caveats

  • License is NOASSERTION for the skill folder; ensure compliance when running in environments with license checks.
  • The workflow requires strict adherence to the mandatory steps; partial execution is not allowed.
  • The description warns against re-hardcoding defaults and requires using constants from the module gene_panel_helpers.py for parameters."
From the SKILL.md

# Gene Panel Selection Workflow This skill is used when you need to construct **biologically meaningful** and **algorithmically robust** gene panels. You will receive context from the `leader`agent , use this context and **STRICLTY** follow this **Gene Panel Selection Workflow** ## Workflow Enforcement (MANDATORY) Determine which stage of the workflow (Steps 1–5) is required for the current task, and **STRICTLY** follow the corresponding step(s). Once a step is entered, all its mandatory sub-steps must be executed. No partial execution or silent degradation is allowed. ## Workdir Always work in the workdir provided by the leader agent. ## Calling other agents You can call other agents by using the `call_agent(agent_name, instruction)` function. - **Call the `browser_use` agent** for information collection: When you encounter software or biological knowledge you are not familiar with, call `browser_use` to search the web and collect the necessary information. - **Call the `system_manager` agent** for software environment installation: When you need to install software packages, call `system_manager` to install them. - **Call the `biologist` agent** for results interpretation: When y

What's inside
Steps it walks through
  1. Workflow Enforcement (MANDATORY)
  2. Workdir
  3. Calling other agents
  4. Visual understanding
  5. Reporting
  6. Large datasets
  7. Gene Panel Selection Hyperparameters
  8. 0. Dataset
  9. 0.1 Parse the user query
  10. 0.2 Search CELLxGENE Census (PRIMARY source)
  11. 0.3 Alternative sources (if Census is insufficient)
  12. 0.4 Validate the retrieved dataset
  13. 1) Dataset Understanding and Splitting
  14. 1.1 Basic structure
More from awesome-bio-agent-skills
All skills →
report-templatePublication-quality PDF report generation using Typst templates. Produces professional scientific…Code Review & Qualitysec-reportSEC (size-exclusion chromatography) analysis with peak detection, oligomer classification, and…Code Review & Qualitybio-atac-seq-consensus-peaksetBuild a differential-ready consensus peakset from per-replicate ATAC-seq peaks using iterative…Otherbio-toolsBiology research tools reference. Always available inside agent containers.AI & Agentsbio-remote-homologyDetect distant homologs using profile and structure-aware methods that go beyond standard BLAST.…Otherbio-variant-callingCall SNPs and indels from aligned reads using bcftools mpileup and call. Use when detecting…Other
About this skill
What does the Gene Panel Selection Workflow skill do?

End-to-end workflow for gene panel design in scRNA-seq and spatial transcriptomics, that should be **STRICTLY** followed: dataset understanding + smart downsampling + train/test splits, algorithmic selection (HVG/DE/RF/scGeneFit/SpaPROS), optimal sub-panel discovery (ARI vs size), biological completion with a stability gate (Completion Rule), consensus scoring and completion (only if there is still room), and benchmarking on test splits (ARI/NMI/Silhouette + UMAP similarity).

How do I install it?

Run `npx skills add BioTender-max/awesome-bio-agent-skills --skill gene_panel_selection --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From BioTender-max/awesome-bio-agent-skills, a repository with 135 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going
All skills →
The full catalog, by category and by tool.
Open Source Radar →
The repositories these skills come from, watched daily.
The people behind them →
Makers and builders across the ecosystem.