Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

TCGA bulk data preprocessing with omicverse · FreedomIntelligence | DecimalAI

Tcga Bulk Data Preprocessing With Omicverse

SkillScore68Maintained

tcga-bulk-data-preprocessing-with-omicverse

Guide Claude through ingesting TCGA sample sheets, expression archives, and clinical carts into omicverse, initialising survival metadata, and exporting annotated AnnData files.

scienceby FreedomIntelligence

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Data Analyst pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/tcga-bulk-data-preprocessing-with-omicverse/SKILL.md

decimalai skills pull tcga-bulk-data-preprocessing-with-omicverse --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/tcga-bulk-data-preprocessing-with-omicverse/SKILL.md https://app.decimal.ai/s/tcga-bulk-data-preprocessing-with-omicverse/SKILL.md

Raw file: /s/tcga-bulk-data-preprocessing-with-omicverse/SKILL.md ↗ · machine-readable summary at /s/tcga-bulk-data-preprocessing-with-omicverse.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 58d ago

Yes — +64% lift-16% tokens0% turns68% of whole cases pass with skill (15/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-17	✗→✓	▲ Improved	—	—
case-18	✗→✓	▲ Improved	—	—
case-13	✗→✓	▲ Improved	—	—
case-15	✗→✓	▲ Improved	—	—
case-01	✗→✓	▲ Improved	—	—

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/30/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

FreedomIntelligence / OpenClaw-Medical-Skills

skills/tcga-preprocessing/SKILL.md

View on GitHub ↗

Imported 8/1/2026Version v1

Skill contents

SKILL.md

TCGA bulk data preprocessing with omicverse

Overview

Follow this skill to recreate the preprocessing routine from t_tcga.ipynb. It automates loading TCGA downloads, generating raw/normalised matrices, initialising metadata, and running survival analyses through ov.bulk.pyTCGA.

Instructions

Gather required downloads

Confirm the user has:
gdc_sample_sheet.<date>.tsv from the TCGA Sample Sheet export.
The decompressed gdc_download_xxxxx directory containing expression archives.
The clinical.cart.<date> directory with clinical XML/JSON files.
Mention that sample data are available under omicverse_guide/docs/Tutorials-bulk/data/TCGA_OV/.

Initialise the TCGA helper

Import omicverse as ov (and scanpy as sc if plotting) then call ov.plot_set().
Instantiate aml_tcga = ov.bulk.pyTCGA(sample_sheet_path, download_dir, clinical_dir).
Run aml_tcga.adata_init() to build the AnnData object with raw counts, FPKM, and TPM layers.

Persist the dataset

Encourage saving the initial AnnData: aml_tcga.adata.write_h5ad('data/TCGA_OV/ov_tcga_raw.h5ad', compression='gzip').
When reloading, reconstruct the class with the same paths and call aml_tcga.adata_read(<path>).

Initialise metadata and clinical information

Populate sample metadata using aml_tcga.adata_meta_init() to convert gene IDs to symbols and attach patient info.
Add survival attributes via aml_tcga.survial_init() (note the intentional spelling in the API).

Perform survival analyses

Plot gene-level survival curves with aml_tcga.survival_analysis('GENE', layer='deseq_normalize', plot=True).
To process all genes, call aml_tcga.survial_analysis_all(); warn that it may take time.

Export results

Save enriched metadata to a new AnnData file (aml_tcga.adata.write_h5ad('.../ov_tcga_survial_all.h5ad', compression='gzip')).
Suggest exporting summary tables (e.g., survival statistics) if users need to share outputs outside Python.

Troubleshooting tips

Ensure TCGA archives are fully extracted; missing XML/TSV files trigger parsing errors.
The helper expects matching case IDs between the sample sheet and expression files—direct users to re-download if IDs do not

align.

Survival plots require clinical dates; if absent, instruct users to check the clinical_cart contents.

Examples

"Read my TCGA OV download, initialise metadata, and plot MYC survival curves using DESeq-normalised counts."
"Reload a saved AnnData file, attach survival annotations, and export the updated .h5ad."
"Run survival analysis for all genes and store the enriched dataset."

References

Tutorial notebook: t_tcga.ipynb
Sample dataset: data/TCGA_OV/
Quick copy/paste commands: reference.md

Illustrative Want to see what tcga-bulk-data-preprocessing-with-omicverse would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

7/27/2026, 06:44 UTC · 22 cases

Lift

+64%

68% vs 5%

Agent turns

1.0 vs 1.0

Tokens

-16%

1,978.18 vs 2,343.95

fail → pass · 14pass → pass · 1fail → fail · 7

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-22 How can I plot a oncoplot waterfall diagram from TCGA MAF mutation files using maftools in R?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-21 How do I perform cell type annotation and clustering on a single-cell RNA-seq 10x Genomics dataset using Scanpy?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-16 I get parsing errors when initializing TCGA data in omicverse. What condition regarding expression archives should be verified?	pass→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-17 I need to share my TCGA survival results with non-Python team members. What output format strategy is recommended for statistical outputs?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-18 List the three essential file and directory components needed from GDC before starting the omicverse TCGA bulk pipeline.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-13 Before starting bulk TCGA processing and visualization in omicverse, what setup function should be called after importing omicverse?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-15 What action should be taken if case IDs in the TCGA sample sheet do not match the downloaded expression file headers during omicverse parsing?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-20 Write a curl script or python script using requests to query the GDC API REST endpoints directly to download BAM files for BRCA patients.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-01 I've got a TCGA export containing a gdc_sample_sheet TSV file, an uncompressed directory with expression files, and a clinical cart folder. Please write a Python script using omicverse to process these files, initialize an AnnData object containing expression layers and mapped gene symbols, and save the raw matrix as a compressed h5ad file.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-02 I have my TCGA RNA-seq download folder along with sample sheets and clinical XML files. I need you to load the dataset using omicverse, populate the patient survival metadata, and generate a Kaplan-Meier survival plot for the EGFR gene using normalized expression matrix layers.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-03 Can you help me reload a previously saved TCGA AnnData file, attach patient clinical survival outcomes, run a genome-wide survival analysis across all genes in the matrix, and export the final enriched AnnData file to disk?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-04 Write a Python snippet to initialize the omicverse TCGA pipeline object passing the sample sheet path 'sheets/gdc.tsv', download directory 'raw/exp', and clinical directory 'raw/clin'.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-05 I have instantiated my TCGA helper object in Python. How do I construct the initial AnnData matrix with counts, TPM, and FPKM layers?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-06 After running initial AnnData creation on a TCGA dataset with omicverse, show how to write the AnnData object to 'data/raw.h5ad' using gzip compression.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-07 I want to load an existing 'data/raw.h5ad' into an omicverse TCGA helper instance. What is the correct pattern to restore the helper and read the file?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-08 How do I convert Ensembl gene IDs to official gene symbols and attach clinical patient attributes in the omicverse TCGA pipeline?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-09 Which method on the omicverse TCGA class adds survival attributes like overall survival status and time to the dataset metadata?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-10 Show how to generate a survival plot for TP53 from normalized expression in omicverse TCGA bulk object using the DESeq normalized layer.	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-11 What method processes survival stats for every single gene present in the matrix using the omicverse TCGA pipeline?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-12 Write the command to save the AnnData object containing all generated survival statistics to 'output/tcga_survival.h5ad' with gzip compression.	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—
case-14 My survival analysis plot in omicverse fails or lacks timeline data. What folder contents should be checked for missing patient clinical records?	fail→pass	—	—	—	—	—	—	—	—	—	—	—	—
case-19 How do I set up a design formula like '~condition' and run DESeq2 differential gene expression analysis using PyDESeq2 directly in Python?	fail→fail	—	—	—	—	—	—	—	—	—	—	—	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +64 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 4.6%
Pass rate with the skill: 68.2%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 1
Skill version measured: v1
Run completed: 7/27/2026, 06:44 UTC

The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run68

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns-19% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 57 days ago

Other measured skills in the registry, with their headline benchmark lift.