Discovering Drugs Through Big Data

Wuhan University researchers develop FingerDTA: an algorithm to predict drug-target binding affinities.

Asian Scientist Magazine (Dec. 20, 2022) — Researchers at Wuhan University School of Computer Science have created a novel computer framework that could help search for new drugs. The framework, which uses an artificial intelligence method called the convolutional neural network, provides global information about potential drug candidates. The study was published in IEEE Xplore.

Fingerprints of drugs are abstract representations of a molecule’s specific structural features, which help determine the most effective way to treat a disease. The Wuhan University team developed a fingerprint-embedding framework, FingerDTA, to predict the affinity of potential drugs to bind to their targets.

Generally, scientists use several methods for drug discovery. High-throughput screening involves testing many compounds at once. Another strategy uses molecular docking—analyzing how a protein interacts with small molecules —to predict which molecules might be useful as drugs. Another method involves predicting the drug’s and its target’s affinity using drug-target affinity prediction models.

More recently researchers have been using deep neural networks to predict the affinity of drugs for their targets. That’s the method Wuhan University researchers focused on for their study.

While using this method, scientists employ a programming model called MapReduce to processes large amounts of data across hundreds or thousands of servers. However, the memory dependency and high communication costs of MapReduce make it unsuitable for handling big data.

So, researchers at Wuhan University proposed a non-MapReduce computing framework that reduces data communication costs and makes computing less dependent on memory.

According to the research team, improving the prediction of a drug’s target binding affinity can lead to new drug discoveries. This may also save substantial human and material resources and speed up drug research. The team aims to implement FingerDTA in big data platforms, an integrated computing solution that combines numerous software systems, tools, and hardware for big data management.

Source: Wuhan University; Images: Shelly Liew/ Asian Scientist Magazine

The article can be found at : FingerDTA: A Fingerprint-Embedding Framework for Drug-Target Binding Affinity Prediction

Disclaimer: This article does not necessarily reflect the views of AsianScientist or its staff.


Asian Scientist Magazine is an award-winning science and technology magazine that highlights R&D news stories from Asia to a global audience. The magazine is published by Singapore-headquartered Wildtype Media Group.

Related Stories from Asian Scientist