Home / News / Technology / AI Revolutionizes Drug Discovery: Market to Surge by $4 Billion with a 40.2% CAGR by 2028
Technology
3 min read

AI Revolutionizes Drug Discovery: Market to Surge by $4 Billion with a 40.2% CAGR by 2028

Published
James Morales
Published

Key Takeaways

  • The market for AI drug discovery tools and services is projected to reach $4.9 billion by 2028.
  • Traditional machine learning techniques for drug discovery revolve around support vector regression (SVR) and various decision tree methods.
  • However, more sophisticated deep neural networks have emerged as increasingly potent tools for identifying and optimizing new compounds.

Around the turn of the millennium, drug discovery emerged as one of the first widespread commercial applications of machine learning (ML). However given the acceleration of ML development in recent years, the technology promises to further revolutionize how new medicine candidates are identified.

For the pharmaceutical industry AI drug discovery tools present a major growth opportunity, with the sector predicted to grow at an average rate of 40.2% for the next 4 years to reach $4.9 billion  by 2028.

How AI is Used to Discover New Drugs

Over the years, an increasing number of AI tools have been developed to help researchers create new compounds, screen libraries of existing compounds for potential medicinal properties and predict how they might interact with biological systems.

By the early 2000s, ML techniques such as support vector regression (SVR) and various decision tree methods had become integral to the dominant drug discovery model, helping researchers classify compounds and predict their properties.

However, the development of sophisticated deep neural networks that have driven AI innovation across industries in recent years initially had a limited impact on drug discovery.

Deep Learning in Drug Discovery

Unlike natural language programming of image analysis, early-phase drug discovery is not a data-rich field.

For applications such as compound classification or property prediction, the amount of well-defined molecular representations available for use as training data was traditionally highly restricted. As such, the process did not play to the strengths of deep learning (DL). 

However, with the development of more sophisticated DL models and larger libraries of drug data, new DL-based approaches have cropped up at every stage of the drug discovery process.

AI Identifies Identifies Entire New Class of Antibiotics

Thanks to advances in the field of deep learning for drug discovery, researchers recently identified a new class of antibiotics that could be used to fight viruses that have evolved resistance to existing treatments.

Commenting on the significance of the study, lead author Felix Wong described how “Explainable Deep Learning” was used to identify compounds in such a way that researchers could understand the underlying mechanisms that might make them suitable treatment candidates:

“This is a breakthrough result showing that explainable deep learning can uniquely catalyze drug discovery, and one of the first demonstrations that deep learning models can explain what they predict with immediate and far-reaching implications to how drug discovery is done and how efficiently we can find new drugs using AI.”

Was this Article helpful? Yes No