English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The influenza virus hemagglutinin is an important part of the virusattachment to the host cells. The hemagglutinin proteins are one of the geneticregions of the virus with a high potential for mutations. Due to the importanceof predicting mutations in producing effective and low-cost vaccines, solutionsthat attempt to approach this problem have recently gained a significantattention. A historical record of mutations have been used to train predictivemodels in such solutions. However, the imbalance between mutations and thepreserved proteins is a big challenge for the development of such models thatneeds to be addressed. Here, we propose to tackle this challenge throughanomaly detection (AD). AD is a well-established field in Machine Learning (ML)that tries to distinguish unseen anomalies from the normal patterns using onlynormal training samples. By considering mutations as the anomalous behavior, wecould benefit existing rich solutions in this field that have emerged recently.Such methods also fit the problem setup of extreme imbalance between the numberof unmutated vs. mutated training samples. Motivated by this formulation, ourmethod tries to find a compact representation for unmutated samples whileforcing anomalies to be separated from the normal ones. This helps the model tolearn a shared unique representation between normal training samples as much aspossible, which improves the discernibility and detectability of mutatedsamples from the unmutated ones at the test time. We conduct a large number ofexperiments on four publicly available datasets, consisting of 3 differenthemagglutinin protein datasets, and one SARS-CoV-2 dataset, and show theeffectiveness of our method through different standard criteria.

InForecaster: Forecasting Influenza Hemagglutinin Mutations Through the Lens of Anomaly Detection