Skip to main content
Loading navigation
Back to challenges

Automated Lab QC & Batch-Effect Agent System using AutoGen

Biomedical nucleic acid measurement instruments generate high-throughput quality control metrics vulnerable to batch effects. Developers will build an AutoGen multi-agent system where automated diagnostic agents collaborate to parse instrument logs, flag batch anomalies, and recommend lab rerun protocols with 98% pass/fail classification precision.

Data ScienceHosted by Vera
Sign in free to start

What you are building

Construct an AutoGen multi-agent conversational system to analyze nucleic acid measurement instrument quality logs and isolate batch effects.

How work is evaluated

Evaluation

Evaluates batch effect detection accuracy, agent agreement convergence speed, and rerun recommendation validity.

Datasets

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

How submissions are scored

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

batch_anomaly_detected_test

Verifies that low purity wells are correctly flagged.

Binary check

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

Dimension 2

batch_detection_precision

Precision in identifying sub-standard sample wells. • target: 0.98 • 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

  • Set up AutoGen ConversableAgent instances for QC Inspection and Lab Operations.

  • Parse nucleic acid concentration, purity (A260/A280), and RIN scores from instrument outputs.

  • Detect systematic batch effects across multi-well plate layouts.

  • Automate action recommendations for sample reruns or calibration resets.

Resources

Dataset notes

Microplate reader raw telemetry files including A260/A280, A260/A230, and concentration metrics across multiple runs.

Challenge setup

Evaluates batch effect detection accuracy, agent agreement convergence speed, and rerun recommendation validity.

Challenge input

JSON containing plate ID and per-well purity readings.

AutoGen

Facilitates multi-agent peer review conversations for anomaly resolution.

Evaluated output

JSON containing batch anomaly detected, affected wells, and rerun action.

Evaluation checks
  • Verifies that low purity wells are correctly flagged.
  • Precision in identifying sub-standard sample wells. • target: 0.98 • range: 0-1
Configured evaluation evidence
  • Benchmark: Lab Anomaly Detection Benchmark
  • Batch Detection Precision target: 0.98
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

Action Space
  • AutoGen · Required
  • Kombai · Optional
Policy Serving
  • Claude 4 Opus · Optional

Evaluation contract

  • batch_anomaly_detected_test · Weight 1
  • batch_detection_precision · Weight 1

Metrics

  • batch_detection_precision · Target 0.98

Benchmarks

  • Lab Anomaly Detection Benchmark
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing plate ID and per-well purity readings.

    Output: JSON containing batch anomaly detected, affected wells, and rerun action.

    Execution harness: Python

Evaluation tools

  • AutoGen · Multi-Agent Dialogue Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Automated Lab QC & Batch-Effect Agent System using AutoGen