Towards Interpreting Zoonotic Potential of Betacoronavirus Sequences With Attention

TL;DR

This paper introduces an attention-enhanced LSTM neural network to predict the zoonotic potential of betacoronaviruses, achieving high accuracy and providing insights into protein features linked to transmission.

Contribution

It presents a novel deep learning approach using attention mechanisms to assess zoonotic risk from viral protein sequences, advancing viral discovery methods.

Findings

Achieved 94% accuracy in predicting zoonotic potential.

Identified key protein features associated with zoonotic transmission.

Provided sequence and structure-level insights into viral replication mechanisms.

Abstract

Current methods for viral discovery target evolutionarily conserved proteins that accurately identify virus families but remain unable to distinguish the zoonotic potential of newly discovered viruses. Here, we apply an attention-enhanced long-short-term memory (LSTM) deep neural net classifier to a highly conserved viral protein target to predict zoonotic potential across betacoronaviruses. The classifier performs with a 94% accuracy. Analysis and visualization of attention at the sequence and structure-level features indicate possible association between important protein-protein interactions governing viral replication in zoonotic betacoronaviruses and zoonotic transmission.