菜单
Load when computing per-cell QC metrics (n_genes, total counts, mt%, ribo%) on a single-cell AnnData before filtering. Skip when reads are still raw FASTQ (use sc-fastq-qc) or you want to filter cells now (use sc-filter).
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills- --skill sc-qc命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
| name | sc-qc |
| description | Load when computing per-cell QC metrics (n_genes, total counts, mt%, ribo%) on a single-cell AnnData before filtering. Skip when reads are still raw FASTQ (use sc-fastq-qc) or you want to filter cells now (use sc-filter). |
| version | 0.3.0 |
| author | OmicsClaw |
| license | MIT |
| tags | ["singlecell","scrna","qc","mitochondrial","ribosomal"] |
| requires | ["anndata","scanpy","matplotlib"] |
The user has a single-cell AnnData (post-counting / post-standardisation)
and wants to review cell quality — counts, detected genes, mitochondrial
percentage, ribosomal percentage — before any filtering. This skill
reports, it does not remove cells. Use sc-filter to actually drop
cells based on these metrics.
| Input | Format | Required |
|---|---|---|
| Single-cell AnnData | .h5ad | yes (unless --demo) |
| Output | Path | Notes |
|---|---|---|
| QC summary | tables/qc_metrics_summary.csv | one row per metric, with stats |
| Per-cell metrics | tables/qc_metrics_per_cell.csv | one row per cell |
| Top expressed genes | tables/highest_expr_genes.csv | by mean expression |
| Diagnostic figures | figures/qc_violin.png, figures/qc_scatter.png | always rendered |
| Provenance | result.json | summary includes n_cells, n_genes, expression_source |
| Report | report.md | always written |
var_names heuristic).scanpy.pp.calculate_qc_metrics.tables/highest_expr_genes.csv.report.md + result.json (no cell filtering performed).sc-qc does not remove cells or genes — it computes metrics and produces figures. Run sc-filter next with thresholds chosen from the QC violins. The result.json summary carries n_cells / n_genes as observed, not as filtered.sc_qc.py:635 raises FileNotFoundError when --input does not resolve. Pre-flight your path before the skill, especially in batch pipelines.expression_source records which matrix the run used. result.json["summary"]["expression_source"] reads layers.counts, adata.raw, or adata.X depending on what the loader picked; QC fractions (mt%, ribo%) are only meaningful on a count-like source. Verify after every run, especially if the input came from outside sc-standardize-input.# Demo (built-in PBMC3K)
python omicsclaw.py run sc-qc --demo --output /tmp/sc_qc_demo
# Real run
python omicsclaw.py run sc-qc --input processed.h5ad --output results/
# With R Enhanced figures
python omicsclaw.py run sc-qc --input processed.h5ad --output results/ --r-enhanced
references/parameters.md — every CLI flag and tuning hintreferences/methodology.md — mt/ribo gene-pattern detection, scanpy QC parametersreferences/output_contract.md — table column schemas + figure rolessc-standardize-input (upstream — required if input is external), sc-filter (next step — actually removes cells), sc-doublet-detection (parallel — finds doublets)Operate MedSAM2 for promptable segmentation of 3D medical images and medical videos, including CT lesion propagation, MRI volumes, RECIST-guided prompts, efficient CPU-oriented variants, training, and 3D Slicer integration. Use when generating or validating volumetric masks from sparse prompts or propagating masks through image slices or video frames.
Build reproducible healthcare imaging pipelines with Project MONAI for DICOM, NIfTI, pathology, and multidimensional imaging tasks including preprocessing, augmentation, training, sliding-window inference, evaluation, model bundles, labeling, and deployment. Use when implementing medical image classification, segmentation, registration, detection, generative, or foundation-model workflows in PyTorch.
Operate Google TxGemma prediction and chat models for therapeutic property prediction across small molecules, proteins, nucleic acids, diseases, targets, and cell lines. Use when formatting Therapeutics Data Commons tasks, choosing TxGemma model size or variant, running local or Model Garden inference, fine-tuning on private therapeutic data, or evaluating TxGemma in drug-discovery workflows.
基于 SOC 职业分类
