🧪 OmicsAgent.ai — Interactive Pipeline Demo

Simulate all 15 omics skills on PBMC data. Click Run Demo to watch the pipeline execute live.

Live Pipeline Simulator

15 skills · PBMC Healthy vs Sepsis · SHA-256 reproducible

$ python3 omics_agent.py --demo
Press Run Demo to start...

🧪Genomics

📊Bulk RNA

🔬scRNA-seq

🧪Proteomics

⚗️Metabolomics

🏔Bulk ATAC

🗺Spatial v1

🦠Metagenomics

💎sc-Proteomics

🔀Integration

🎯scATAC-seq

⛰️Bulk Epigen.

🌐Spatial Full

🔗TCR/BCR

🔭sc-Prot Full

Ready to run0 / 15

0Skills done

0Figures

0Tables

0sElapsed

YOU

Cluster my PBMC scRNA-seq and identify exhausted CD8 T cells

OMICSAGENT.AI scrnascTCR_BCR

Running QC → normalization → UMAP → Leiden clustering...
10 cell types found. Exhausted CD8 T: TOX+, NR4A1+, PDCD1+, LAG3+ (8% of PBMC). Report → output/scrna_report/

YOU

Find TFs active in those exhausted cells using ATAC-seq

OMICSAGENT.AI scatac

TF-IDF + LSI → ChromVAR TF deviations (JASPAR 2022)...
Top TFs: TOX (z=4.8), NR4A1 (z=4.2), NFATC1 (z=3.9), PDCD1 (z=3.1). Report → output/scatac_report/

YOU

Integrate RNA, ATAC and proteomics with MOFA+

OMICSAGENT.AI integration

MOFA+ latent factor analysis (3 layers, 40 samples)...
Factor 1 (18% var): Exhaustion — TOX/NR4A1 in ATAC, PD-1/LAG3 in RNA, PD-1/TIM-3 in protein.

Run locally: python3 omics_agent.py --chat

Output file tree

output/
  scrna_report/
    figures/umap.png dotplot.png qc.png
    tables/markers.csv
    commands.sh environment.yml checksums.sha256
  scatac_report/ spatial_full_report/ scTCR_BCR_report/
  ... 11 more skill report folders ...

Cell type composition (scRNA-seq)

PBMC 3,000 cells — 10 cell types

CD4 T

28%

Mono Class.

18%

CD8 T

14%

NK cell

10%

B cell

Exhausted CD8

Other

14%

UMAP clusters (simulated)

Verify: sha256sum -c output/scrna_report/checksums.sha256