Skip to main content
Loading navigation
Back to challenges

Build an Intrinsically Disordered Protein Demethylase Targeting Classifier with BioPython

Intrinsically disordered regions (IDRs) actively constrain genomic target selection by RNA and histone demethylases. Build an IDR sequence analysis model using BioPython and PyTorch to predict demethylase chromatin targeting specificity with ROC-AUC exceeding 0.87.

Data ScienceHosted by Vera
Sign in free to start

What you are building

Construct an IDR sequence classifier using BioPython sequence utilities and PyTorch to model chromatin targeting restrictions.

How work is evaluated

Evaluation

Evaluates IDR-driven targeting restriction classification ROC-AUC across demethylase variants.

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

disorder_range_test

Ensures disorder percentage is between 0 and 100.

Binary check

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

Dimension 2

roc_auc

ROC-AUC of IDR chromatin targeting restriction prediction. • target: 0.87 • range: 0.5-1

Binary check

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

Learning goals

  • Extract IUPred3 disorder scores and biophysical properties using BioPython.

  • Train PyTorch classifier neural networks on demethylase amino acid sequences.

  • Predict chromatin targeting restrictions caused by intrinsically disordered region configurations.

Resources

Dataset notes

Dataset of 500 mammalian and plant RNA/histone demethylase protein sequences annotated with experimental genomic targeting outcomes.

Challenge setup

Evaluates IDR-driven targeting restriction classification ROC-AUC across demethylase variants.

Challenge input

JSON containing protein_id and aa_sequence

Biopython

Provides tools for sequence manipulation and biophysical property estimation.

PyTorch

Deep learning library for sequence classification.

Evaluated output

JSON object with target_restricted boolean and disorder_percentage

Evaluation checks
  • Ensures disorder percentage is between 0 and 100.
  • ROC-AUC of IDR chromatin targeting restriction prediction. • target: 0.87 • range: 0.5-1
Configured evaluation evidence
  • Benchmark: ProteinDisorder-Bench
  • Roc Auc target: 0.87
  • 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
  • OATutor · Required
  • Transformers · Optional
  • Command R+ · Optional

Evaluation contract

  • disorder_range_test · Weight 1
  • roc_auc · Weight 1

Metrics

  • roc_auc · Target 0.87

Benchmarks

  • ProteinDisorder-Bench
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing protein_id and aa_sequence

    Output: JSON object with target_restricted boolean and disorder_percentage

    Execution harness: Python

Evaluation tools

  • Biopython · sequence_processing
  • PyTorch · model_training

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build an Intrinsically Disordered Protein Demethylase Targeting Classifier with BioPython