Claude Code Plugins

Community-maintained marketplace

Feedback

dorado-bench-v2

@Single-Molecule-Sequencing/ont-ecosystem
0
0

Oxford Nanopore basecalling with Dorado on University of Michigan HPC clusters (ARMIS2 and Great Lakes). Use when running dorado basecalling, generating SLURM jobs for basecalling, benchmarking models, optimizing GPU resources, or processing POD5 data. Captures model paths, GPU allocations, and job metadata. Integrates with ont-experiments for provenance tracking. Supports fast/hac/sup models, methylation calling, and automatic resource calculation.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name dorado-bench-v2
description Oxford Nanopore basecalling with Dorado on University of Michigan HPC clusters (ARMIS2 and Great Lakes). Use when running dorado basecalling, generating SLURM jobs for basecalling, benchmarking models, optimizing GPU resources, or processing POD5 data.

Dorado-Bench v2 - ONT Basecalling

Basecalling toolkit for UM HPC clusters with provenance tracking.

Integration

Run through ont-experiments for provenance tracking:

ont_experiments.py run basecalling exp-abc123 --model sup --output calls.bam --json stats.json

Or standalone:

python3 dorado_basecall.py /path/to/pod5 --model sup --cluster armis2 --output calls.bam

Cluster Configurations

ARMIS2 (sigbio-a40)

partition: sigbio-a40
account: bleu1
gres: gpu:a40:1
dorado: /nfs/turbo/umms-bleu-secure/programs/dorado-1.1.1-linux-x64/bin/dorado
models: /nfs/turbo/umms-bleu-secure/programs/dorado_models

Great Lakes (gpu_mig40)

partition: gpu_mig40
account: bleu99
gres: gpu:nvidia_a100_80gb_pcie_3g.40gb:1
dorado: /nfs/turbo/umms-atheylab/dorado-bench/dorado-1.1.1-linux-x64/bin/dorado
models: /nfs/turbo/umms-atheylab/dorado-bench/Models

Athey Lab Turbo Drive Paths:

  • Base: /nfs/turbo/umms-atheylab/
  • Dorado: /nfs/turbo/umms-atheylab/dorado-bench/dorado-1.1.1-linux-x64/bin/dorado
  • Models: /nfs/turbo/umms-atheylab/dorado-bench/Models/Simplex/
  • Input: /nfs/turbo/umms-atheylab/dorado-bench/Input/
  • Output: /nfs/turbo/umms-atheylab/dorado-bench/Output/
  • Sequencing data: /nfs/turbo/umms-atheylab/sequencing_data/

Model Tiers

Tier Accuracy ARMIS2 Resources Great Lakes Resources
fast ~95% batch=4096, mem=50G, 24h batch=2048, mem=32G, 24h
hac ~98% batch=2048, mem=75G, 72h batch=1024, mem=64G, 72h
sup ~99% batch=1024, mem=100G, 144h batch=512, mem=64G, 144h

Options

Option Description
--model TIER fast, hac, sup (default: hac)
--version VER Model version (default: v5.0.0)
--cluster armis2 or greatlakes
--output FILE Output BAM file
--json FILE Output JSON statistics
--slurm FILE Generate SLURM script
--emit-moves Include move table
--modifications MOD Enable 5mCG_5hmCG methylation

SLURM Generation

python3 dorado_basecall.py /path/to/pod5 \
  --model sup@v5.0.0 \
  --cluster armis2 \
  --slurm job.sbatch

sbatch job.sbatch

Event Tracking

When run through ont-experiments, captures:

  • Model name and full path
  • Model tier/version/chemistry
  • Batch size and device
  • BAM statistics (reads, qscore, N50)
  • SLURM job ID, nodes, GPUs

Methylation Calling

ont_experiments.py run basecalling exp-abc123 \
  --model sup \
  --modifications 5mCG_5hmCG \
  --output calls_5mc.bam

Resources adjusted: memory +50%, batch size -30%

Great Lakes Quick Start

Using greatlakes_submit.py (Simplified)

# Generate and submit a SUP basecalling job
python3 greatlakes_submit.py \
  /nfs/turbo/umms-atheylab/dorado-bench/Input/sample1 \
  --model sup@v5.0.0 \
  --output /nfs/turbo/umms-atheylab/dorado-bench/Output/sample1_sup.bam \
  --email user@umich.edu

# Dry run (generate script only)
python3 greatlakes_submit.py \
  /nfs/turbo/umms-atheylab/sequencing_data/my_sample/pod5 \
  --model hac \
  --output /nfs/turbo/umms-atheylab/dorado-bench/Output/my_sample_hac.bam \
  --dry-run

Using dorado_basecall.py

# Generate SLURM script for Great Lakes
python3 dorado_basecall.py \
  /nfs/turbo/umms-atheylab/dorado-bench/Input/sample1 \
  --model sup@v5.0.0 \
  --cluster greatlakes \
  --output /nfs/turbo/umms-atheylab/dorado-bench/Output/sample1_sup.bam \
  --slurm job.sbatch

sbatch job.sbatch

Batch Job Submission

# Generate sbatch files from command list
python3 make_sbatch_from_cmdtxt.py \
  -i dorado_basecaller_sup_cmd.txt \
  -o Sbatch/sup \
  --account bleu99 \
  --partition gpu_mig40 \
  --gres "gpu:nvidia_a100_80gb_pcie_3g.40gb:1" \
  --time "16:00:00" \
  --email user@umich.edu

# Submit all jobs
bash sbatch_all.sh

Monitoring Jobs

squeue -u $USER                    # Check your jobs
squeue -p gpu_mig40                # Check partition queue
seff <job_id>                      # Job efficiency (after completion)
sacct -j <job_id> --format=JobID,JobName,State,ExitCode,Elapsed