Zhuoran Qiao Awarded Inaugural Chen Institute and Science Prize for AI Accelerated Research
Zhuoran Qiao has been honored with the first-ever Chen Institute and Science Prize for AI Accelerated Research. His work focuses on capturing and visualizing dynamic, small-scale behaviors of biomolecules that have previously remained hidden. These insights are crucial for applications such as drug development.
The Chen Institute and Science Prize highlights young researchers who apply artificial intelligence to tackle significant scientific challenges and speed up research progress.
Machine Learning and Molecular Structures
Biomolecular interactions, especially among proteins and smaller molecules, drive essential life processes. Understanding these interactions at a detailed level is key to developing new drugs. However, this requires decoding the three-dimensional structures of molecular complexes, which traditionally involves time-consuming lab techniques like X-ray crystallography and cryogenic electron microscopy.
AI-powered protein structure prediction has made impressive progress by predicting protein shapes from amino acid sequences. Yet, according to Qiao, this progress is just the beginning of building a comprehensive “computational microscope” for molecular biology. It’s important to visualize systems with thousands of atoms and multiple conformations to truly understand molecular behaviors.
In a recent Nature Machine Intelligence publication, Qiao and his team introduced novel machine learning methods to improve the view of two critical aspects: protein-ligand interactions and the environments where these interactions happen. Ligands, molecules that bind to proteins, influence their structure and thus affect many chemical and biological processes.
Qiao explained, “To develop new drugs, you need extremely accurate models of biomolecular interactions. Getting the structure right and understanding the interaction strength between proteins or molecules is essential for drug success.” He noted the complexity involved in representing molecular movements, comparing it to solving a maze with thousands of dimensions.
Introducing NeuralPLexer
The team developed a tool named NeuralPLexer to visualize these interactions more effectively. It accounts for the dynamic nature of biomolecules by generating multiple snapshots to capture their behavior. Starting from an initial rough model, NeuralPLexer progressively refines structural details, enabling researchers to quickly obtain detailed views of molecular interactions.
Using NeuralPLexer, the researchers predicted the formation of “cryptic pockets”—special binding sites that only appear upon ligand binding. The tool showed strong ability to identify new drug binding pockets and other key molecular features.
Qiao emphasized the efficiency gains: “Compared to traditional methods, our approach can deliver what normally takes six months of high-throughput experiments in just one day.”
His motivation stems from recognizing that many computational chemistry problems, while theoretically understood, are not practically computable with existing methods. Winning this prize is a significant acknowledgment of his research direction and a reminder to continue impactful work, including mentoring the next generation in computational chemistry.
Qiao is optimistic about the future of molecular modeling and looks forward to advancing this work at Chai Discovery.
About the Prize and Finalists
Chrissy Luo, cofounder of the Chen Institute, highlighted the quality and diversity of applications received for the prize, spanning multiple scientific fields. She stressed the importance of AI in accelerating scientific discovery and expressed enthusiasm about recognizing young researchers pushing the boundaries of knowledge.
Besides Qiao, the finalists for the prize include:
- Aditya Nair, incoming assistant professor at Nanyang Technological University, Singapore, for his machine learning research on neural codes linked to emotion states.
- Alizée Roobaert, fellow at the Flanders Marine Institute, for applying machine learning to monitor key ocean variables, especially along coasts, to better understand the ocean’s role in the carbon cycle.
This recognition underscores how AI is becoming integral to addressing complex challenges across disciplines, from molecular biology to environmental science.
Your membership also unlocks:
