Enforce reproducible preprocessing standards for single-cell multi-omics datasets.
npx skills add https://github.com/Mang30/myskills --skill -mang30
Please help me install this Agent Skill. Skill: 单细胞多组学数据处理规范 Source: https://github.com/Mang30/myskills/tree/main/skills/bioinformatics-sc Command: npx skills add https://github.com/Mang30/myskills --skill -mang30
It helps you avoid inconsistent, error-prone single-cell multi-omics preprocessing by enforcing clear, reproducible rules for loading formats, QC thresholds, randomness control, and memory-safe handling.
Ask the AI to convert your dataset loading plan and QC checklist into an implementation checklist that follows the exact rules in this skill for MTX, Hi-C pairsam streaming, scRNA QC, deterministic seeding, and h5ad-backed memory safety.
High-intent search queries and answers about installing and using this skill.
To load large 10x Genomics MTX files safely, enable cache-enabled MTX reads and use backed h5ad reads for intermediate outputs. This approach standardizes scRNA ingestion and prevents memory blowups when processing datasets larger than 1GB.
The best way to process compressed Hi-C pairsam files is line-by-line streaming iteration over gzip pairsam files. This memory-safe preprocessing standard prevents memory exhaustion by avoiding loading entire large files into memory at once.
To ensure reproducible scRNA preprocessing, fix random seeds for deterministic randomness, apply predefined scRNA QC thresholds, and record software versions. Saving intermediate results as h5ad files guarantees auditability across multiple samples.
Yes, this standard requires h5ad-based intermediate outputs and version recording for traceability. Enforcing h5ad formats ensures consistent memory-safe data handling and reproducible auditability across scRNA, scATAC, and scHiC pipelines.
You should use h5ad-backed reads to safeguard against memory blowups when handling large single-cell multi-omics files. Backed reads allow streaming access to intermediate outputs, ensuring memory-safe processing for datasets exceeding 1GB.
Standardizing single-cell multi-omics preprocessing requires applying fixed scRNA QC thresholds, using gene symbols as feature names, enabling MTX caching, and fixing random seeds. These rules ensure deterministic and traceable pipeline outputs.
ya-wayProfile raw 10x Genomics single-cell matrices to output QC distributions as JSON.
K-Dense-AIAutomates end-to-end RNA-seq workflow to extract QC metrics clusters and cell-type annotations using Scanpy.
PharMolixApply scvi-tools probabilistic models to single-cell multi-omics data.
KetomihineProvide offline ArchR documentation for scATAC-seq analysis workflows.
