Skip to main content
Back to challenges
Challenge

Build a Structural Variant Triage Pipeline with Claude Agents SDK and Google Jules

Oxford Nanopore long-read sequencing requires real-time quality control and structural variant categorization for clinical biopharma applications. Build an autonomous agent using Claude Agents SDK and Google Jules to parse raw BAM/VCF files and achieve 95% accuracy on SV pathogenicity scoring.

Data ScienceHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Implement an automated long-read genomic variant triage system using Claude Agents SDK with extended thinking and Google Jules automation.

Delivery guide

How work is evaluated

Evaluation

Evaluates SV pathogenicity scoring accuracy and workflow execution time.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

schema_validation

Ensures output JSON conforms to expected payload structure.

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

classification_accuracy

Percentage of correctly classified SV pathogenicity levels. • target: 0.95 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Master long-read genomic sequence parsing and quality filtering.

  • Implement autonomous variant classification using Claude Agents SDK.

  • Integrate Google Jules for automated workflow execution and CI/CD validation.

  • Deploy an auditable genomic QC pipeline adhering to clinical standards.

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic long-read sequencing VCF dataset with annotating Oxford Nanopore structural variants.

Challenge setup

Evaluates SV pathogenicity scoring accuracy and workflow execution time.

Challenge input

JSON containing list of structural variants with chromosome, position, and read depth.

Claude Agents SDK

Required framework for building multi-step reasoning agents with extended thinking.

Google Jules

Provides automated task workflow automation and continuous evaluation runners.

Evaluated output

JSON mapping variant IDs to classification and confidence score.

Evaluation checks
  • Ensures output JSON conforms to expected payload structure.
  • Percentage of correctly classified SV pathogenicity levels. • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: Genomic Variant Benchmarks (GIAB)
  • Classification Accuracy target: 0.95
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Google Jules · Required
  • Google · Optional
  • Hugging Face · Optional
Orchestration
  • Google Jules · Required

Evaluation contract

  • schema_validation · Weight 1
  • classification_accuracy · Weight 1

Metrics

  • classification_accuracy · Target 0.95

Benchmarks

  • Genomic Variant Benchmarks (GIAB)
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing list of structural variants with chromosome, position, and read depth.

    Output: JSON mapping variant IDs to classification and confidence score.

    Execution harness: Python

Evaluation tools

  • Claude Agents SDK · Orchestration Framework
  • Google Jules · AI Engineering Tool

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Structural Variant Triage Pipeline with Claude Agents SDK and Google Jules