Skip to content

A transformer network trained to predict end-to-end single sequence protein structure as a set of angles given amino acid sequences.

License

Notifications You must be signed in to change notification settings

danielathome19/ProteiNN-Structure-Predictor

Repository files navigation

About

ProteiNN is a Transformer network trained to predict end-to-end single-sequence protein structure by amino acid sequences. To find out more, check out the provided research paper:

  • "Deep Learning for Protein Structure Prediction: Advancements in Structural Bioinformatics" (DOI: 10.1101/2023.04.26.538026)
  • Also contained in the "PaperAndPresentation" folder is the research paper.

Usage

Run main.py to choose from either "train" or "predict" modes; train will retrain the model and predict will provide users the option to enter an amino acid sequence to predict the structure of (which will be output as a PDB file).

Bugs/Features

Bugs are tracked using the GitHub Issue Tracker.

Please use the issue tracker for the following purpose:

  • To raise a bug request; do include specific details and label it appropriately.
  • To suggest any improvements in existing features.
  • To suggest new features or structures or applications.

License

The code is licensed under Apache License 2.0.

Citation

If you use this code for your research, please cite this project:

@software{Szelogowski_ProteiNN-Structure-Predictor_2023,
 author = {Szelogowski, Daniel},
 doi = {10.1101/2023.04.26.538026},
 month = {April},
 title = {{ProteiNN-Structure-Predictor}},
 license = {Apache-2.0},
 url = {https://github.com/danielathome19/ProteiNN-Structure-Predictor},
 version = {1.0.0},
 year = {2023}
}

About

A transformer network trained to predict end-to-end single sequence protein structure as a set of angles given amino acid sequences.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages