New method accurately detects virulent infections in advance

When viruses infect cells, changes occur in the cell nucleus, and these can be observed by fluorescence microscopy. Using fluorescence images of living cells, researchers at the University of Zurich have trained an artificial neural network to reliably recognize cells infected with adenoviruses or herpes viruses. The procedure also identifies severe acute infections at an early stage.

In humans, adenoviruses can infect cells in the respiratory tract, while herpes viruses can infect cells in the skin and nervous system. In most cases, this does not lead to the production of new viral particles, as the viruses are suppressed by the immune system. However, adenoviruses and herpes viruses can cause persistent infections that the immune system is unable to completely suppress and which produce viral particles for years. These same viruses can also cause sudden, violent infections where affected cells release large amounts of the virus, so the infection spreads quickly. This can lead to severe acute illnesses of the lungs or nervous system.

Automatic detection of virus infected cells

The research group of Urs Greber, professor in the Department of Molecular Life Sciences at the University of Zurich (UZH), has just shown for the first time that a machine learning algorithm can recognize cells infected with the herpes or adenoviruses based solely on the fluorescence of the cell nucleus.

Our method not only reliably identifies cells infected with the virus, but also accurately detects virulent infections in advance. “

Urs Greber, Professorm Department of Molecular Life Sciences, University of Zurich (UZH)

The study authors believe their development has many applications, including predicting the reaction of human cells to other viruses or microorganisms. “The method opens up new avenues for better understanding infections and discovering new agents active against pathogens such as viruses or bacteria,” adds Greber.

The analytical method is based on the combination of fluorescence microscopy in living cells with deep learning processes. Herpes and adenoviruses formed inside an infected cell change the organization of the nucleus, and these changes can be seen under a microscope. The group has developed a deep learning algorithm – an artificial neural network – to automatically detect these changes. The network is trained with a large number of microscopy images through which it learns to identify patterns that are characteristic of infected or uninfected cells. “Once training and validation is complete, the neural network automatically detects cells infected with the virus,” says Greber.

Reliably predict severe acute infections

The research team also demonstrated that the algorithm is able to identify acute and severe infections with 95% accuracy and up to 24 hours in advance. Images of living cells of lytic infections, in which viral particles multiply rapidly and cells dissolve, as well as images of persistent infections, in which viruses are produced continuously but only in small quantities, served as training material. Despite the great precision of the method, it is not yet clear what characteristics of infected cell nuclei are recognized by the artificial neural network to distinguish the two phases of infection. However, even without this knowledge, researchers are now able to study the biology of infected cells in more detail.

The group has already discovered some differences: The internal pressure of the nucleus is greater during virulent infections than during persistent phases. In addition, in a cell with lytic infection, viral proteins accumulate more quickly in the nucleus. “We suspect that distinct cellular processes determine whether or not a cell decays after being infected. We can now investigate these and other questions,” says Greber.