Clustering Matrix Protein of Influenza Virus Using Deep Embedded Network

Abstract

Influenza A virus is a type of virus that can cause respiratory illness in humans and animals. They are classified into subtypes based on the combination of two proteins on the surface of the virus: hemagglutinin (HA) and neuraminidase (NA). There are 18 different HA subtypes and 11 different NA subtypes, and many different combinations of these subtypes are possible. One way to study these viruses is to use clustering techniques to group them based on certain features. A deep embedded network is used to learn a low-dimensional representation of the virus sequences, which is used as input to a clustering algorithm called K-means. To perform this analysis, we collected a dataset of influenza A viruses. Then The deep embedded network is used as a learning representation. K-means clustering is applied to the learned representation to cluster the virus sequences into clusters based on their similarity. The number of clusters can be determined using techniques such as the elbow method or the silhouette score. Using deep embedded networks and K-means clustering can provide insights into the relationships between different influenza A viruses and help researchers understand patterns and trends in the data. It can also be useful for tracking the evolution and spread of these viruses over time.