Skip to main content
Back to challenges
Challenge

Newborn Genetic Cancer Risk Triage Engine with Scanpy and Biopython

Newborn genomic screening can identify pediatric cancer predispositions early. You will build a bioinformatics pipeline in Python using Scanpy and Biopython to process single-cell sequencing and genomic variant files, predicting high-risk pathogenic germline variants with 95% sensitivity.

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.

Construct a bioinformatics pipeline using Scanpy and Biopython to triage germline cancer risk variants in newborn genomic screening cohorts.

Delivery guide

How work is evaluated

Evaluation

Evaluates germline variant pathogenic scoring accuracy against expert genetic counselor classifications.

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

pathogenic_class_match

Ensure high-risk pathogenic variants are flagged correctly

Binary check

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

Dimension 2

sensitivity

Sensitivity in identifying true pathogenic variants • 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

  • Process genomic sequence alignments using Biopython and PySAM libraries

  • Analyze single-cell expression trajectories of embryonic tissues using Scanpy

  • Evaluate pathogenic classification of germline variants according to ACMG guidelines

Resources and assets

Reference links and supporting material

Dataset notes

Newborn genomic screening dataset with 500 annotated germline VCF files and paired single-cell RNA profiles.

Challenge setup

Evaluates germline variant pathogenic scoring accuracy against expert genetic counselor classifications.

Challenge input

JSON object with gene_symbol, cdn_change, and pop_af

Scanpy

Single-cell transcriptome processing framework.

Biopython

Genomic file and sequence processing engine.

Evaluated output

JSON object with acmg_class string and pathogenic_probability float

Evaluation checks
  • Ensure high-risk pathogenic variants are flagged correctly
  • Sensitivity in identifying true pathogenic variants • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: PedGeno-Bench
  • Sensitivity 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
  • Reducto · Required
Reward / Eval
  • Distributional · Optional
Policy Serving
  • Claude 4 Sonnet · Optional
Safety / Guardrails
  • Distributional · Optional

Evaluation contract

  • pathogenic_class_match · Weight 1
  • sensitivity · Weight 1

Metrics

  • sensitivity · Target 0.95

Benchmarks

  • PedGeno-Bench
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with gene_symbol, cdn_change, and pop_af

    Output: JSON object with acmg_class string and pathogenic_probability float

    Execution harness: Python

Evaluation tools

  • Scanpy · Single-Cell Analysis Toolkit
  • Biopython · Biomedical Computational Library

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Newborn Genetic Cancer Risk Triage Engine with Scanpy and Biopython