Model Card for viral-sequence-prediction

This model was a part of the University of Virginia's NLP course final project: NLP for social good. Using publically available data from the NCBI virus data portal, we designed an LSTM-based classifier that can predict the virus of origin for any arbitrary protein sequence.

Specifically, we were interested in classifying COVID v Flu.

Using the model

Code for the model can be found at: https://github.com/nleroy917/CS6501-final

You can use the train.ipynb notebook, or use the code directly:

from dna_classification.models import DNASequenceClassifier

model = DNASequenceClassifier("nleroy917/viral-sequence-prediction")

virus = model.predict("MGYINVFAFPFTIYSLLLCRMNFRNYIAQVDVVNFNLT")
>>> COVID