Artificial intelligence and machine learning

There is a newer version of the record available.

Published October 5, 2021 | Version 1.0.0
Dataset Open

Replication Data for: Geometric Transformers for Protein Interface Contact Prediction

Creators

  • 1. University of Missouri

Description

This dataset contains replication data for the paper titled "Geometric Transformers for Protein Interface Contact Prediction". The dataset consists of pickled Python dictionaries containing pairs of DGLGraphs that can be used to train and validate protein interface contact prediction models. It also contains our best model checkpoint saved as a PyTorch LightningModule. Our GitHub repository, DeepInteract, linked in the "Additional notes" metadata section below provides more details on how we use these files as examples for cross-validation.

Notes

This dataset can be used for training and cross-validation of protein interface contact prediction models via our GitHub repository for DeepInteract (https://github.com/BioinfoMachineLearning/DeepInteract).

Files

Files (6.8 GB)

Name Size Download all
md5:c3dbfaecec447ce5f0fdbc179baad8c9
596 Bytes Download
md5:a57152d8580645c788d5fa80b923f6e8
4.3 GB Download
md5:c59dbab87669fee3f669dc32f00b5d70
2.4 GB Download
md5:331cf17b557694863a19aa23ba1477a9
3.0 MB Download
md5:1571ef687b62e4c60976f9827700dcb2
139.8 kB Download
md5:c2c90f736df529eb30f3b5d20b783493
60.5 MB Download

Additional details

Related works

Cites
10.7910/DVN/H93ZKK (DOI)

Funding

U.S. National Science Foundation
III: Medium: Collaborative Research: Guiding Exploration of Protein Structure Spaces with Deep Learning 1763246
U.S. National Science Foundation
ABI Innovation: Deep learning methods for protein bioinformatics 1759934