Machine Learning to Identify Viral Pathogens through Host Cell Gene Expression Patterns

Pandemics like COVID-19 and AIDS were caused by new viruses that went undetected for crucial periods, allowing them to spread before the underlying pathogen was identified. This delay in detection led to widespread transmission and made containment efforts more difficult. Would these viruses have caused global pandemics if they had been identified earlier? My research leverages the power of Machine Learning (ML) to detect viruses before they have the chance to spread undetected. By analyzing subtle changes in host cell gene expression that occur immediately after infection, my ML model can learn to distinguish infected cells based on the unique molecular responses to viral invasion. Each virus leaves a distinct molecular signature in the host cell, and by recognizing these patterns, the model can not only detect known viruses but also flag infections caused by new pathogens. This approach provides a proactive surveillance tool to identify emerging viruses early, enabling healthcare professionals and scientists to act swiftly and prevent pandemics before they escalate. This work will significantly improve our ability to respond to emerging infectious diseases and contribute to global efforts in pandemic preparedness, with the potential to save lives by identifying threats before they become unmanageable public health crises.