Skip to main content
Loading navigation
Back to challenges

Automate Gestational Phenotype GWAS Analysis with Hail and LlamaIndex

Genome-wide association studies across gestational phenotypes reveal context-specific maternal-fetal genetic risk factors. Build a queryable GWAS synthesis agent using Hail and LlamaIndex to process summary statistics across 50 maternal traits and index context-specific variant locus associations.

Data ScienceHosted by Vera
Sign in free to start

What you are building

Distributed GWAS summary statistic analysis engine using Hail framework and queryable LlamaIndex search nodes.

How work is evaluated

Evaluation

Evaluates lead SNP locus extraction accuracy and trimester context alignment.

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

significance_filter_test

Ensures outputted p-value satisfies the significance threshold.

Binary check

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

Dimension 2

locus_identification_accuracy

Accuracy in mapping correct lead SNP per phenotype. • 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

  • Process large-scale GWAS summary statistics using PySpark/Hail matrix tables.

  • Index locus-level p-value and effect size signals using LlamaIndex.

  • Identify context-specific genetic effects distinct to specific gestational trimesters.

Resources

Dataset notes

Dataset of summary statistics across 50 simulated gestational traits covering 1,000,000 variants.

Challenge setup

Evaluates lead SNP locus extraction accuracy and trimester context alignment.

Challenge input

JSON containing phenotype_name and significance_threshold

Hail

Scalable framework for handling genomic summary statistics.

LlamaIndex

Indexes structured metadata for search query answering.

Evaluated output

JSON object with lead_rsid, p_value, and context_specificity

Evaluation checks
  • Ensures outputted p-value satisfies the significance threshold.
  • Accuracy in mapping correct lead SNP per phenotype. • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: GWAS-Bench
  • Locus Identification Accuracy target: 0.95
  • 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
  • Llama Index · Required
  • Framer · Optional
  • OpenAI o3 · Optional

Evaluation contract

  • significance_filter_test · Weight 1
  • locus_identification_accuracy · Weight 1

Metrics

  • locus_identification_accuracy · Target 0.95

Benchmarks

  • GWAS-Bench
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing phenotype_name and significance_threshold

    Output: JSON object with lead_rsid, p_value, and context_specificity

    Execution harness: Python

Evaluation tools

  • Hail · genomics_framework
  • LlamaIndex · agent_framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Automate Gestational Phenotype GWAS Analysis with Hail and LlamaIndex