skills.rest
cpanse
cpanseCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

scmultiomics-analysis

Submit, debug, and deploy the FGCZ ScMultiOmics SUSHI app for multi-assay single-cell analysis.

Updated May 28, 2026
Search Tags:#deployment#single-cell#seurat#vdj#multiomics#cite-seq#sushi
One-click install
npx skills add https://github.com/cpanse/skills --skill scmultiomics-analysis-cpanse
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: scmultiomics-analysis
Source: https://github.com/cpanse/skills/tree/main/fgcz-infrastructure/skills/scmultiomics-analysis
Command: npx skills add https://github.com/cpanse/skills --skill scmultiomics-analysis-cpanse

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Running single-cell multi-omics analysis at FGCZ requires correctly chaining ScSeurat and ScMultiOmics SUSHI jobs, auto-discovering ADT, VDJ, and ATAC modalities, and deploying Ruby app files across four SUSHI instances — a process with many silent failure modes like empty exploreSC links and blank interactive tables. ## Core Features & Use Cases - Job Submission & Modality Detection: Submit ScMultiOmics jobs downstream of ScSeurat with automatic detection of ADT (CITE-seq), VDJ-T/B (scRepertoire), and ATAC (CellRanger ARC + Signac) modalities, plus optional WNN integration. - Debugging Failed Runs: Diagnose known failure modes such as empty exploreSC links, missing ADT DotPlot circles, blank interactive tables, and ScSeuratCombine not finding scData.qs2, using the documented gotchas index. - Multi-Instance Deployment: Push ScMultiOmicsApp.rb and ScSeuratApp.rb changes to production, course, demo, and test SUSHI instances with Passenger restarts and a post-deployment smoke checklist. - Use Case: A bioinformatician receives a CellRanger Multi dataset (RNA + ADT + VDJ-T), runs ScSeurat, submits ScMultiOmics on the output dataset, and verifies the rendered report against the canonical smoke reference before delivery. ## Quick Start Ask the agent to submit an ScMultiOmics job on the ScSeurat output dataset for your CellRanger Multi sample and verify the ADT, VDJ, and WNN tabs in the rendered report.

Frequently Asked Questions about scmultiomics-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I submit an ScMultiOmics job in SUSHI?▼

Run ScSeurat first on the CellRanger output, then select the ScSeurat output dataset as input to ScMultiOmics. Required columns are Name, Species, refBuild, and SC Seurat; ScSeurat propagates CountMatrix and ResultDir so sibling modalities are auto-discovered.

How does ScMultiOmics detect ADT, VDJ, and ATAC modalities?▼

Detection runs against the original CountMatrix directory: ADT from Antibody Capture features in the H5 file, VDJ from sibling vdj_t/vdj_b filtered_contig_annotations.csv files, and ATAC from atac_fragments.tsv.gz plus atac_peaks.bed. VDJTPath and VDJBPath columns override auto-discovery.

Can ScMultiOmics process BD Rhapsody data?▼

Yes, set SCDataOrigin to BDRhapsody and BDRhapsodyPath to the directory containing the prebuilt Seurat rds file. BD bypasses ScSeurat, but the SUSHI gate still requires an SC Seurat column, so use a hand-built dataset.tsv with a stub value.

Why is the exploreSC link empty or stalling after an ScMultiOmics run?▼

A missing ?data= file triggers a shiny.silent.error with an empty message, showing only a stalled spinner. Check /srv/shinyproxy/logs/exploreSC_*_stderr.log and verify the gstore-relative path; newer runs set param$exploreSCUrl from the Report column to avoid the old cwd heuristic.

Why does ScSeuratCombine fail on ScMultiOmics output?▼

ScSeuratCombine reads the SC Seurat column and expects a scData.qs2 path. The app creates a relative symlink scData.qs2 pointing to scMultiData.qs2 and adds an SC Seurat [Link] row in next_dataset; both must be present for the combine job to work.

Which SUSHI instances need the ScMultiOmics app files deployed?▼

Both ScSeuratApp.rb and ScMultiOmicsApp.rb must land on all four instances: production (fgcz-h-082), course and demo (fgcz-h-081), and test (fgcz-h-083). Restart Passenger via tmp/restart.txt after every copy, and reinstall ezRun when R runtime or templates change.

Related Skills

Explore More Data & Analytics Skills→
tony-zhelonkintony-zhelonkin
1

cellranger-multi-to-anndata

Build one pooled AnnData from CellRanger Multi per-sample outputs with gene-union concatenation.

Community
Advanced
pwwangpwwang
22

celltypeannotation

Annotate scRNA-seq clustering results with cell type labels using multiple tools.

Community
Advanced
CHENyiru3CHENyiru3

seurat-v5

Integrate multi-modal single-cell RNA and protein data with Seurat v5 WNN.

Community
Advanced
whatagraphwhatagraph

whatagraph-sources-and-data

Discover, query, and manage marketing data sources in Whatagraph.

Official
Advanced