AI Algorithm Solves Structural Biology Challenges

Determining the 3D shapes of biological molecules is one of the hardest problems in modern biology and medical discovery. Companies and research institutions often spend millions of dollars to determine a molecular structure - and even such massive efforts are frequently unsuccessful.

Using clever, new machine learning techniques, Stanford University PhD students Stephan Eismann and Raphael Townshend, under the guidance of Ron Dror, associate professor of computer science, have developed an approach that overcomes this problem by predicting accurate structures computationally.

Most notably, their approach succeeds even when learning from only a few known structures, making it applicable to the types of molecules whose structures are most difficult to determine experimentally.

Their work is demonstrated in two papers detailing applications for RNA molecules and multi-protein complexes, published in Science on Aug. 27, 2021, and in Proteins in December 2020, respectively. The paper in Science is a collaboration with the Stanford laboratory of Rhiju Das, associate professor of biochemistry.

"Structural biology, which is the study of the shapes of molecules, has this mantra that structure determines function," said Townshend.

The algorithm designed by the researchers predicts accurate molecular structures and, in doing so, can allow scientists to explain how different molecules work, with applications ranging from fundamental biological research to informed drug design practices.

"Proteins are molecular machines that perform all sorts of functions. To execute their functions, proteins often bind to other proteins," said Eismann. "If you know that a pair of proteins is implicated in a disease and you know how they interact in 3D, you can try to target this interaction very specifically with a drug."

Eismann and Townshend are co-lead authors of the Science paper with Stanford postdoctoral scholar Andrew Watkins of the Das lab, and also co-lead authors of the Proteins paper with former Stanford PhD student Nathaniel Thomas.

Designing the algorithm

Instead of specifying what makes a structural prediction more or less accurate, the researchers let the algorithm discover these molecular features for itself. They did this because they found that the conventional technique of providing such knowledge can sway an algorithm in favor of certain features, thus preventing it from finding other informative features.

"The problem with these hand-crafted features in an algorithm is that the algorithm becomes biased towards what the person who picks these features thinks is important, and you might miss some information that you would need to do better," said Eismann.

"The network learned to find fundamental concepts that are key to molecular structure formation, but without explicitly being told to," said Townshend. "The exciting aspect is that the algorithm has clearly recovered things that we knew were important, but it has also recovered characteristics that we didn’t know about before."

Having shown success with proteins, the researchers next applied their algorithm to another class of important biological molecules, RNAs. They tested their algorithm in a series of “RNA Puzzles” from a long-standing competition in their field, and in every case, the tool outperformed all the other puzzle participants and did so without being designed specifically for RNA structures.

Broader applications

The researchers are excited to see where else their approach can be applied, having already had success with protein complexes and RNA molecules.

"Most of the dramatic recent advances in machine learning have required a tremendous amount of data for training. The fact that this method succeeds given very little training data suggests that related methods could address unsolved problems in many fields where data is scarce," said Dror, who is senior author of the Proteins paper and, with Das, co-senior author of the Science paper.

Specifically for structural biology, the team says that they’re only just scratching the surface in terms of scientific progress to be made.

"Once you have this fundamental technology, then you’re increasing your level of understanding another step and can start asking the next set of questions," said Townshend. "For example, you can start designing new molecules and medicines with this kind of information, which is an area that people are very excited about."

Raphael J L Townshend, Stephan Eismann, Andrew M Watkins, Ramya Rangan, Maria Karelina, Rhiju Das, Ron O Dror.
Geometric deep learning of RNA structure.
Science, 2021. doi: 10.1126/science.abe5650

Most Popular Now

Artificial Intelligence in Healthcare Re…

This study presents an overview of the development, adoption and use of Artificial Intelligence (AI) technologies and applications in the healthcare sector across all Member States. The main aim of...

New App Helps Parents Identify Treatable…

A ground-breaking new, mobile phone app, 'GrowthMonitor' places the accurate measurement of children's height in the hands of parents and carers. Preliminary data to be presented at the Society for...

Dedalus Acquires Swiftqueue to Support P…

Dedalus Group ("Dedalus"), a leading international healthcare software solutions provider, has announced to have completed the acquisition of 100% of Swiftqueue Technologies Ltd a fast-growing cloud-native appointment and scheduling solution...

Bittium Exhibits its High-Tech Medical T…

Bittium exhibits its innovative products and solutions for cardiology and neurophysiology as well as R&D services for the development of medical and healthcare technology at the MEDICA 2021 event. It...

Development of AI Technology for Produci…

Transcranial focused ultrasound can be used to treat degenerative movement disorders, intractable pain, and mental disorders by delivering ultrasound energy to a specific area of the brain without opening the...

FDA Authorizes Marketing of Virtual Real…

The U.S. Food and Drug Administration today authorized marketing of EaseVRx, a prescription-use immersive virtual reality (VR) system that uses cognitive behavioral therapy and other behavioral methods to help with...

Tulane University Study Uses AI to Detec…

A Tulane University researcher found that artificial intelligence (AI) can accurately detect and diagnose colorectal cancer from tissue scans as well or better than pathologists, according to a new study...

Siemens Healthineers and UCSF Create Fir…

Siemens Healthineers and UC San Francisco have formed a research and innovation-driven collaboration to make radiological imaging greener, while improving access to and quality of radiological imaging in Northern California...

Open Call DIGITAL-2021-DEPLOY-01-HEALTH:…

The consolidation of a European framework and ecosystem of digital health solutions and services, covering technological and organisational innovation and addressing the needs of the involved stakeholders, including those of...

MEDICA 2021 + COMPAMED 2021: Holding the…

15 - 18 November 2021, Düsseldorf, Germany. This week, all of the big decision makers and professional experts from the international healthcare industry will finally be making their way to Düsseldorf...

MEDICA 2021 and COMPAMED 2021 have Far E…

15 - 18 November 2021, Düsseldorf, Germany. After their four-day run as an in-person event, MEDICA and COMPAMED have achieved extremely successful results in Düsseldorf. From 15 to 18 November 2021...

Philips Integrates MedChat's AI Capabili…

Royal Philips (NYSE: PHG, AEX: PHIA), a global leader in health technology, announced a collaboration with USA-based MedChat to integrate MedChat's live chat and AI-driven chatbot services into Philips Patient...