/viral-dna-classification

Viral DNA Genome Classification with Deep Learning

Primary LanguageJupyter NotebookMIT LicenseMIT

Human and Virus DNA Classification with TensorFlow 2

In this notebook, we will classify human and viral DNA with Deep Learning using TensorFlow 2. At the end of this tutorial, our model will reach approximately 90% accuracy.

We will follow ViraMiner paper and code which is published by Vicente R. et.al. You can access the original article and source code. I have made some updates to the original code for educational purposes.

  • Updated code base from Python 2 to Python 3.
  • Added modules like pandas and removed plain Python code for I/O.
  • Used single model.

Dataset

Train, validation and test files are here.

Tutorial

Jupyter notebook is here.

TODO

  • Detailed explanation about genetics will be added.
  • More model alternatives will be added.

Licence

See LICENCE file.