Skip to main content
Back to challenges
Challenge

Implement a Single-Cell Epigenomic Cell-State Triage with Scanpy and AnnData

Identifying drug-resistant tumor sub-populations requires multi-omic integration of single-cell RNA and DNA methylation assays. Construct a single-cell multi-omic clustering engine using Scanpy and AnnData to isolate resistant tumor cell sub-clusters with Adjusted Rand Index (ARI) > 0.82.

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.

Single-cell multi-omic cell state triage engine built with Scanpy and AnnData matrix manipulation toolkit.

Delivery guide

How work is evaluated

Evaluation

Evaluates single-cell cluster fidelity using Adjusted Rand Index (ARI) on multi-omic single-cell matrices.

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

resistant_fraction_validity

Ensures fraction is between 0 and 1.

Binary check

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

Dimension 2

adjusted_rand_index

ARI clustering accuracy metric against known cell state labels. • target: 0.82 • 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

  • Incorporate multi-modal single-cell data (scRNA-seq + scMethylation) into AnnData objects.

  • Perform dimensionality reduction (PCA, UMAP) and Leiden community clustering using Scanpy.

  • Isolate drug-resistant single-cell states based on biomarker expression thresholds.

Resources and assets

Reference links and supporting material

Dataset notes

Dataset of 1,000 synthetic single-cell multi-omic profiles representing responsive and resistant tumor sub-populations.

Challenge setup

Evaluates single-cell cluster fidelity using Adjusted Rand Index (ARI) on multi-omic single-cell matrices.

Challenge input

Path to AnnData h5ad file or matrix dict

Scanpy

Toolkit for single-cell gene expression analysis.

AnnData

Data structure format for annotated data matrices.

Evaluated output

JSON containing cluster_assignments array and detected_resistant_fraction

Evaluation checks
  • Ensures fraction is between 0 and 1.
  • ARI clustering accuracy metric against known cell state labels. • target: 0.82 • range: 0-1
Configured evaluation evidence
  • Benchmark: scRNA-Bench
  • Adjusted Rand Index target: 0.82
  • 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
  • Exa · Required
  • Anyscale · Optional
Policy Serving
  • GPT-4o · Optional

Evaluation contract

  • resistant_fraction_validity · Weight 1
  • adjusted_rand_index · Weight 1

Metrics

  • adjusted_rand_index · Target 0.82

Benchmarks

  • scRNA-Bench
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: Path to AnnData h5ad file or matrix dict

    Output: JSON containing cluster_assignments array and detected_resistant_fraction

    Execution harness: Python

Evaluation tools

  • Scanpy · data_analysis
  • AnnData · data_structure

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Implement a Single-Cell Epigenomic Cell-State Triage with Scanpy and AnnData