Computational and Structural Biotechnology Journal, Journal Year: 2025, Volume and Issue: unknown

Published: March 1, 2025

The genomic diversification of viral pathogens during epidemics and pandemics represents a major adaptive route for infectious agents to circumvent therapeutic public health initiatives. Historically, strategies address evolution have relied on responding emerging variants after their detection, leading delays in effective responses. Because this, long-standing yet challenging objective has been forecast by predicting potentially harmful mutations prior emergence. promises artificial intelligence (AI) coupled with the exponential growth data collection infrastructures spurred COVID-19 pandemic, resulted research ecosystem highly conducive this objective. Due pandemic accelerating development mitigation preparedness strategies, many methods discussed here were designed context SARS-CoV-2 evolution. However, most these pipelines intentionally be adaptable across RNA viruses, several already applied multiple species. In review, we explore recent breakthroughs that facilitated forecasting an ongoing particular emphasis deep learning architectures, including promising potential language models (LM). approaches employ leverage genomic, epidemiologic, immunologic biological information.

Language: Английский