Computational Modeling Streamlines Antiviral Discovery

Viruses, such as those responsible for COVID-19 and HIV, present significant challenges once they enter the human body. Antiviral therapies aim to obstruct a virus or prevent its replication. Nonetheless, the ever-changing nature of viruses constantly adapting and altering their forms complicates the development of effective antiviral treatments.

Recent research employs a novel computational modeling technique to represent the intricate and varied shapes that viral proteins can assume. This research will be showcased at the 69th Biophysical Society Annual Meeting, scheduled for February in Los Angeles.

This innovative method, integrated into the open-source Integrative Modeling Platform (IMP) software, merges multiple experimental strategies, including various visualization and measurement techniques for actual viruses (such as cryo-electron microscopy and mass spectrometry), along with molecular dynamics simulations, to provide a thorough understanding of a virus's dynamic behavior.

The project was led by Kenneth Huang, PhD, a Postdoctoral Computational Structural Biologist who worked in the University of California, San Francisco labs of Andrej Sali and Ignacia Echeverria. He likened viruses to nightmare homes, the interior of which can vary greatly depending on the circumstances.

To create antivirals, they are attempting to determine “the fastest way to demolish this house with the least number of whacks with an ax,” Kenneth said.

To date, they have focused their approach on a crucial protein involved in the replication of the COVID-19 virus, known as Nsp2. Through their modeling, they have developed a comprehensive representation of Nsp2 not merely as a static structure, but as a series of flexible forms it can take. Kenneth expressed surprise at the extent to which Nsp2 alters “in response to whatever is around it.”

Kenneth and associates can use this new tool to forecast where to target medications that would best block its replication and how to design those medications by comprehending this flexibility and the various shapes Nsp2 can take. Even though humans currently have antivirals for COVID-19, tools that aid in the most effective design of antiviral medications could save countless lives in a scenario like a pandemic.

Drug screens, in which businesses test thousands of molecules to determine whether they have the desired effect, are frequently used to find antivirals and other medications.

So, they just basically use brute force and keep screening compounds until they eventually find something that works.”

Kenneth Huang, Postdoctoral Computational Structural Biologist, University of California, San Francisco

He clarified that this “brute force” approach can be costly and time-consuming.

The time, personnel, and financial resources needed to screen thousands of compounds can be greatly reduced by instead creating compounds that are specifically targeted for a virus. This strategy may open the door to more effective and focused treatments for a variety of viral infections.

“We want to be able to kill these viruses so that they do not make people sick,” Kenneth said.