技能 latchbio-integration

🧬

latchbio-integration

Name: latchbio-integration
Author: davila7

安全 ⚙️ 外部命令📁 檔案系統存取🔑 環境變數⚡ 包含腳本🌐 網路存取

Build bioinformatics workflows on Latch

也可從以下取得: K-Dense-AI

Building bioinformatics pipelines requires learning platform-specific tools and cloud infrastructure. This skill provides step-by-step guidance for creating serverless workflows using the Latch SDK with decorators, data abstractions, and verified bioinformatics tools.

支援: Claude Codex Code(CC)

📊 71 充足

下載技能 ZIP

在 Claude 中上傳

前往設定 → 功能 → 技能 → 上傳技能

開啟並開始使用

測試它

正在使用「latchbio-integration」。 Create a Latch workflow for RNA-seq analysis with quality control and alignment

預期結果:

Define @small_task for quality control (FastQC)
Define @large_task for alignment (STAR/HISAT2)
Define @small_task for quantification (featureCounts)
Create @workflow combining all tasks with LatchFile inputs
Register workflow with latch register command

正在使用「latchbio-integration」。 Set up GPU resources for protein folding with AlphaFold

預期結果:

Use @large_gpu_task decorator with V100 or A100 GPU type
Configure memory allocation based on protein sequence length
Set appropriate timeout for large protein structure prediction
Configure output directory with LatchDir for results

正在使用「latchbio-integration」。 Organize my sequencing samples in Latch Registry

預期結果:

Create a Table with columns for sample_id, condition, and file references
Use Record class to add samples with metadata
Link related records for sample relationships
Query samples using filter criteria in workflows

安全審計

安全

v5 • 1/17/2026

Pure documentation skill containing only markdown reference guides. No executable code detected - only code examples in markdown code blocks. All static findings are false positives from scanner misinterpreting documentation examples as executable commands.

已掃描檔案

2,154

分析行數

發現項

審計總數

風險因素

⚙️ 外部命令 (201)

📁 檔案系統存取 (2)

references/data-management.md:64 references/resource-configuration.md:276

🔑 環境變數 (2)

references/data-management.md:380 references/data-management.md:380

⚡ 包含腳本 (1)

references/verified-workflows.md:13-20

🌐 網路存取 (9)

references/verified-workflows.md:439 references/workflow-creation.md:252 references/workflow-creation.md:253 references/workflow-creation.md:254 skill-report.json:6 SKILL.md:335 SKILL.md:336 SKILL.md:338 SKILL.md:339

審計者: claude 查看審計歷史 →

品質評分

架構

100

可維護性

內容

社群

100

安全

規範符合性

你能建構什麼

Deploy RNA-seq analysis pipelines

Build reproducible RNA-seq pipelines with quality control, alignment, and quantification using verified workflows

Run protein structure prediction

Execute AlphaFold and ColabFold predictions on GPU resources with automatic result organization

Integrate existing pipelines

Port existing Nextflow or Snakemake workflows to cloud with minimal code changes

試試這些提示

Create basic workflow

Create a Latch workflow for processing FASTQ files with @small_task and @workflow decorators

Configure GPU resources

Configure GPU resources for protein structure prediction using @large_gpu_task or custom_task with A100 GPU

Import Nextflow pipeline

Import my existing Nextflow pipeline to Latch using latch register --nextflow command

Use verified workflows

Run bulk RNA-seq analysis using the latch.verified.bulk_rnaseq workflow with sample metadata

最佳實務

Use type annotations for all workflow parameters to enable automatic UI generation
Start with standard task decorators (@small_task, @large_task) and scale resources only when profiling shows need
Break complex workflows into modular tasks for better debugging and parallel execution

避免

Using @large_gpu_task for tasks that do not require GPU acceleration
Skipping type annotations which disables automatic UI generation
Over-allocating CPU and memory without profiling actual requirements

常見問題

What programming languages does Latch support for workflows?

Latch supports native Python workflows using the SDK, along with Nextflow and Snakemake pipelines. All workflows run in Docker containers.

What compute resources are available for tasks?

Tasks can use CPU cores up to 96, memory up to 768GB, and GPUs including K80, V100, and A100. Use decorators for standard sizes or custom_task for specific configurations.

How do I integrate Latch workflows with external tools?

Use LatchFile and LatchDir to reference cloud data. Verified workflows can be imported from latch.verified module. Custom tasks can call command-line tools in your Docker image.

Is my data secure on the Latch platform?

Latch provides encrypted cloud storage. Files transfer securely between local and cloud storage. Use Registry for structured data with access controls.

Why is my workflow registration failing?

Common issues include Docker not running, authentication token expired, missing dependencies in Dockerfile, or syntax errors. Use latch register --verbose for detailed logs.

How does Latch compare to other workflow platforms?

Latch provides automatic containerization, generated no-code UIs, and verified bioinformatics workflows unlike AWS Batch or Google Cloud Life Sciences.

開發者詳情

作者

davila7

授權

MIT

儲存庫

https://github.com/davila7/claude-code-templates/tree/main/cli-tool/components/skills/scientific/latchbio-integration

引用

main

檔案結構

📁 references/

📄 data-management.md

📄 resource-configuration.md

📄 verified-workflows.md

📄 workflow-creation.md

📄 SKILL.md