Skip to main content
Back to challenges
Challenge

Build a Nanopore Clinical Workflow Manager with Google ADK & Novita AI

Nanopore sequencing offers rapid clinical biopharma turnaround, but raw FASTQ quality control requires immediate automated intervention. Develop an automated QC pipeline using Google ADK and Novita AI to evaluate read length distribution, isolate quality drop-offs, and trigger re-sequencing flags under 5 minutes per run.

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.

Automate Oxford Nanopore sequence quality inspection using Google ADK and Novita AI inference endpoints.

Delivery guide

How work is evaluated

Evaluation

Evaluates automated sequencing quality flag precision and latency on FASTQ metric files.

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

qc_flag_correctness

Verifies correct pass/fail categorization based on clinical thresholds.

Binary check

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

Dimension 2

qc_classification_accuracy

Accuracy of QC status assignment across benchmark runs. • 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

  • Build workflow management agents using Google Agent Development Kit (ADK).

  • Deploy Novita AI inference acceleration for high-throughput genomic data validation.

  • Parse FASTQ telemetry metrics to compute N50 read lengths and Q-score distributions.

Resources and assets

Reference links and supporting material

Dataset notes

Dataset of 60 Oxford Nanopore run summary statistics covering standard, degraded, and high-yield biopharma sequencing runs.

Challenge setup

Evaluates automated sequencing quality flag precision and latency on FASTQ metric files.

Challenge input

JSON containing mean_q_score, n50_read_length, and total_yield_gb

Google ADK

Google Agent Development Kit for building structured agent workflows.

Novita AI

Provides fast LLM inference endpoints for analytical decision tools.

Evaluated output

JSON object with qc_passed boolean and recommendation

Evaluation checks
  • Verifies correct pass/fail categorization based on clinical thresholds.
  • Accuracy of QC status assignment across benchmark runs. • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: GenoQC-Bench
  • Qc 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 · Required
  • Novita AI · Optional
  • Google ADK · Optional

Evaluation contract

  • qc_flag_correctness · Weight 1
  • qc_classification_accuracy · Weight 1

Metrics

  • qc_classification_accuracy · Target 0.95

Benchmarks

  • GenoQC-Bench
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing mean_q_score, n50_read_length, and total_yield_gb

    Output: JSON object with qc_passed boolean and recommendation

    Execution harness: Python

Evaluation tools

  • Google ADK · agent_framework
  • Novita AI · inference_runtime

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Nanopore Clinical Workflow Manager with Google ADK & Novita AI