Ni-os/MPRALegNet-container

Containers for Genomic API for Model Evaluation (GAME) using DREAM-RNN Model

Authors Nikita Penzin, Ivan V Kulakovskiy, Ilya Vorontsov, Dmitry Penzar

The Predictor container (predictor.sif) includes:

• Predictor script for sequence processing and error handling.
• Integrated MPRALegNet model with its dependencies and LegNet conda environment created using legnet_environment.yml.
• Pre-trained model weights (k562_best_model_test1_val2.ckpt, hepg2_best_model_test1_val2.ckpt, wtc11_best_model_test1_val2.ckpt) for predictions downloaded from https://zenodo.org/records/10558183.
• Support scripts like:
  • api_preprocessing_utils.py
  • error_checking_functions.py
  • predictor_help_message.json.
• Dependencies required by the Predictor.

The Evaluator container (evaluator.sif) includes:

• Evaluator API script for genomic sequence evaluation.
• Dependencies required by the Evaluator.

Evaluator data directory (evaluator_data.zip)

This ZIP file contains sample JSON data for testing the Evaluator container.

Contents:

• evaluator_message_gosai_5seqs.json (Default input)
• evaluator_message_more_complex.json
• evaluator_message_simple_test.json
• evaluator_input_sample_test.json

Important Notes:

1. Input JSON file restriction:
   1. Currently, the script in this container can only process evaluator_message_simple_test.json as input.
   2. If a different JSON needs to be tested, please rename the JSON file to 'evaluator_message_simple_test' to allow for the script to be able to find the mounted path in evaluator_data/
2. Predictions directory must be created:
   1. Before running the Evaluator container, please also create and mount predictions/ directory. Additional information can be found on GitHub: Genomic Model Evaluation API MPRALegNet Model-specific information can be found within the same repository: MPRALegNet(will be availabe later)