Home / News / Technology / AI Antibiotic Discovery: Deep-Learning Algorithm Leads Researchers to Groundbreaking Compound
Technology
5 min read

AI Antibiotic Discovery: Deep-Learning Algorithm Leads Researchers to Groundbreaking Compound

Last Updated January 12, 2024 3:27 PM
Samantha Dunn
Last Updated January 12, 2024 3:27 PM

         Key Takeaways

Key Takeaways

  • Research into antibiotic resistance has led to the discovery of a new class of antibiotics.
  • AI development has allowed researchers to discover new compounds to treat MRSA in mice.
  • There hasn’t been a major antibiotic discovery in almost sixty years.
  • The rapid development of AI stands to significantly impact future drug discovery.

Antibiotic researchers have used AI to help them discover a new class of antibiotics that are effective against methicillin-resistant Staphylococcus aureus (MRSA) in mice. The research  initially conducted by a team at the Broad Institute of MIT and Harvard included deep learning artificial intelligence to identify these new classes of antibiotics.

In the update  to Broad Institute’s research, 283 compounds were tested in mice with findings of a number that were effective against MRSA. Explainable deep learning AI was used in this groundbreaking study, which facilitated the discovery.

The Use of AI to Discover New Drugs

The machine learning approach utilized by the Broad Institute in their antibiotic research study can be used by researchers to generalize beyond their training data set and look for patterns and arrangements within a chemical structure in a way that humans are not able to. This use of AI as a pharmaceutical research tool has the potential to allow researchers to discover new potential drugs.

The latest development in explainable deep learning for new drug discovery comes after researchers at MIT originally identified  promising antibiotic candidates in February 2020. At the time, James Collins, the Termeer Professor of Medical Engineering and Science in MIT’s Institute for Medical Engineering and Science (IMES) and Department of Biological Engineering noted:

“We wanted to develop a platform that would allow us to harness the power of artificial intelligence to usher in a new age of antibiotic drug discovery […] Our approach revealed this amazing molecule which is arguably one of the more powerful antibiotics that has been discovered.”

In a recent update, Felix Wong, a postdoctoral fellow at M.I.T. and co-founder of Integrated Biosciences, spoke about his learnings on The New York Times ‘Hard Fork  podcast, sharing how the artificial intelligence algorithm used to conduct this research was critical in reducing the labor process involved in such studies.

“This idea of scale is quite important, at least in our paper we looked at 12 million compounds in a candidate set […] that’s basically infinity for most practical purposes,” Wong noted.

The role of AI in being able to speed up the discovery of new drugs is particularly useful in the early stages of drug discovery. A 2020 paper  published by The National Center for Biotechnology Information highlighted AI’s use in reducing the human workload as well as achieving targets in a shorter time frame.

Another study  in 2023 highlighted the role of AI in drug discovery and its clinical relevance.

A typical learning pyramid with critical questions that must be kept in mind while developing AI applications for drug discovery.
Learning pyramid with critical questions that must be kept in mind while developing AI applications for drug discovery. Source: Qureshi R, Irfan M, Gondal TM, Khan S, Wu J, Hadi MU, Heymach J, Le X, Yan H, Alam T.

Black Box Versus White Box Models

Within AI models, the decision-making process of machine-learning models is referred to as black box – essentially AI systems whose internal workings are not visible to the user, and white box machine learning models which enable users to understand how the AI makes decisions and comes to conclusions, which in turn facilitates greater learning and development as demonstrated by Broad Institutes advanced drug research.

Black box models raise ethical concerns over unexplained biases or unethical decisions that are hidden from users. OpenAI’s ChatGPT is a black box learning model that has received some criticism  and unease around the transparency of the leading generative AI tool.

Use Cases For Emerging Technologies In Pharmaceutical Research

Figure of applications of artificial intelligence (AI) in different subfields of the pharmaceutical industry.
Applications of artificial intelligence (AI) in different subfields of the pharmaceutical industry, from drug discovery to pharmaceutical product management.

New and emerging technologies such as AI, blockchain technology, and quantum computing  are revolutionizing the pharmaceutical industry, all with the potential to speed up the lifecycle of developing pharmaceutical products. AI, in particular, stands to shape the face of the pharmaceutical landscape in the coming years with applications from drug discovery to pharmaceutical product management.

The Future of AI Will Deliver Novel Real-World Use Cases

The interest in developing AI technology within the pharmaceutical industry is leading researchers to discover even greater real-world use cases for AI, solving key problems through novel approaches. Of course, AI innovation is not limited to the pharmaceutical or medical industry and has applications across practically every industry.

Although AI is still in its infancy, there is huge potential for the development of new AI technologies to further innovation in a meaningful way. Looking forward, transparency and ethical decision-making will be something that AI developers will need to consider – even as they continue to create tools and products that are intended to better humanity.

Was this Article helpful? Yes No