菜单
Load when summarising structural variants from an SV VCF (DEL / DUP / INV / TRA) — BND-notation parsing, size classification, per-type counts. Skip when working with small SNVs / indels (use `genomics-variant-calling`) or calling SVs from BAM (run Manta / Delly / Sniffles first).
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。
基于 SOC 职业分类
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.
Evaluate and operate released Profluent OpenCRISPR gene-editing systems, especially OpenCRISPR-1, for controlled research workflows using its published Cas9-like protein, compatible guide RNA designs, protocols, licensing, specificity testing, and experimental validation. Use when comparing OpenCRISPR-1 with SpCas9, planning nonclinical editing studies, or assessing use in nuclease, nickase, deactivated, base, prime, or epigenome-editing contexts.
Operate CZI TranscriptFormer cross-species generative single-cell models to produce cell embeddings, contextual gene embeddings, likelihoods, zero-shot classifiers, disease-state representations, and regulatory analyses from raw-count AnnData files. Use when selecting TF-Sapiens, TF-Exemplar, or TF-Metazoa, processing in- or out-of-distribution species, or scaling embedding extraction across GPUs.
Operate ByteDance Protenix-v2 for open biomolecular structure prediction of proteins, antibodies, nucleic acids, ligands, and complexes using JSON inputs, MSA and template features, constraints, and inference-time sampling. Use when running Protenix locally or through its server, comparing AlphaFold3-style open models, or building reproducible co-folding evaluations.
| name | genomics-sv-detection |
| description | Load when summarising structural variants from an SV VCF (DEL / DUP / INV / TRA) — BND-notation parsing, size classification, per-type counts. Skip when working with small SNVs / indels (use `genomics-variant-calling`) or calling SVs from BAM (run Manta / Delly / Sniffles first). |
| version | 0.5.0 |
| author | OmicsClaw |
| license | MIT |
| tags | ["genomics","structural-variants","sv","manta","delly","sniffles","bnd"] |
| requires | ["pandas","numpy"] |
The user has an SV VCF (from Manta, Delly, Lumpy, Sniffles, etc.) and wants per-type counts (DEL / DUP / INV / TRA / INS), size classification (small 50 bp–1 kb / medium 1 kb–100 kb / large 100 kb–10 Mb / very-large > 10 Mb), and BND breakend resolution.
The script does NOT call SVs from a BAM. Run an external SV caller first; this skill summarises its VCF output.
| Input | Format | Required |
|---|---|---|
| Structural variants | .vcf (SV-flavoured: SVTYPE in INFO and/or BND ALT rows) | yes (unless --demo) |
| Output | Path | Notes |
|---|---|---|
| SV table | tables/structural_variants.csv | per-SV CHROM/POS/SVTYPE/SVLEN/size_class |
| Report | report.md + result.json | always |
--input <sv.vcf>) or generate a demo SV VCF at output_dir/demo_structural_variants.vcf with --n-svs records (sv_detection.py:170).INFO/SVTYPE (sv_detection.py:103). Records without INFO/SVTYPE (e.g. pure BND ALT notation from Manta) classify as UNKNOWN — there is NO BND-to-TRA resolution.abs(SVLEN) for size classification (sv_detection.py:105); bin into size classes; aggregate per-type counts.tables/structural_variants.csv (sv_detection.py:343) + report.md + result.json (:346).--input REQUIRED unless --demo. sv_detection.py:330 raises ValueError("--input required when not using --demo"); non-existent paths raise FileNotFoundError at :333.--n-svs only affects --demo (sv_detection.py:319, default 100). Silently ignored when --input is set.INFO/SVTYPE classify as UNKNOWN. sv_detection.py:103 reads only INFO/SVTYPE; there is no BND ALT-notation parser and no MATEID pairing logic. Manta callsets that emit translocations as paired BND records (without an SVTYPE=TRA INFO field) will appear as UNKNOWN, not TRA. Pre-process with bcftools view -i 'INFO/SVTYPE!=""' or with a Manta-specific BND→TRA resolver upstream.SVLEN is stored as absolute value in the CSV. sv_detection.py:105 writes abs(int(info.get("SVLEN", end - pos))) — a 1234-bp deletion becomes 1234 in the CSV regardless of the input sign. The original signed SVLEN is NOT preserved.# Demo (100 synthetic SVs)
python omicsclaw.py run genomics-sv-detection --demo --output /tmp/sv_demo
# Custom demo size
python omicsclaw.py run genomics-sv-detection --demo --n-svs 500 \
--output /tmp/sv_demo_large
# Real SV VCF
python omicsclaw.py run genomics-sv-detection \
--input manta_diploid.vcf --output results/
references/parameters.md — every CLI flagreferences/methodology.md — SVTYPE / BND semantics, size-class boundariesreferences/output_contract.md — tables/structural_variants.csv schemagenomics-alignment (upstream — provides BAMs for SV callers), genomics-variant-calling (parallel — small SNVs / indels), genomics-cnv-calling (parallel — copy-number from depth, complementary to SV callers), genomics-variant-annotation (downstream — functional impact of breakpoints)