Posted in | News | Medical Robotics

Researchers Use AI for Faster, Simpler Identification of Small Active Substance Molecules

Currently, over one-third of the available medicines are based on active substances from nature. A research team led by the University of Jena has developed a method for faster and simpler identification of small active substance molecules.

Researchers Use AI for Faster, Simpler Identification of Small Active Substance Molecules.
Martin Hoffmann presents the COSMIC workflow. Image Credit: Jens Meyer (University of Jena)

The secondary natural substances that are derived from several plants, bacteria and fungi can be anti-inflammatory, which can eliminate pathogens or even restrict cancer cell growth. However, utilizing the medicines provided by nature through the identification of new natural substances is time-consuming, costly and labor-intensive.

A team of bioinformaticians at Friedrich Schiller University Jena have now developed an approach that facilitates quicker and simpler identification of small active substance molecules.

The research regarding the development of the method called COSMIC (Confidence Of Small Molecule IdentifiCations) was published in the journal Nature Biotechnology.

Millions of Structural Data Items not yet Deciphered

To trace the kind of substances held in a biological sample like plant extract, the study used mass spectroscopy to analyze the molecules. In this process, the mass is determined by breaking down the molecules into fragments.

The CSI:FingerID molecule search engine we developed allows us to search specifically for molecular structures that match these fragments. Whether this search is successful — i.e., whether the search result represents the correct structure — is not something we can distinguish in this way.

Sebastian Böcker, Professor, University of Jena

A huge collection of data with billions of mass spectrometry data items is available from millions of analyses of biological samples, a wide majority of which are yet to be identified as to their structure. The COSMIC is playing its role in this situation, where it enables the structure to be automatically understood for a large proportion of these still unidentified molecules.

To this end, we use machine-learning methods. First, the mass spectrum of the sample under examination is compared with the available structural data.

Martin Hoffmann, Study Lead Author, University of Jena

According to the results, an extensive list of possible hits is shown in the Google search.

Our method now indicates how confident one can be that the hit found in the first place is actually the structure we are looking for.

Martin Hoffmann, Study Lead Author, University of Jena

To perform this, the COSMIC quantifies a score that tests the quality of the recommended hit and concludes whether it is correct or incorrect.

New Bile Acids Discovered

Böcker and his colleagues successfully explained the effectiveness of their approach, in association with colleagues from the University of California, San Diego. They analyzed mass spectroscopy data from the digestive system of mice, searching for some unknown bile acids.

As part of this study, over 28,000 theoretically possible bile acids structures were built and put into comparison with the measurement data from the microbiome of the mice. The corresponding analyses using COSMIC produced a total of 11 new, previously fully unknown bile acid structures. Specifically synthesized reference samples were used to confirm two of those samples.

This shows, firstly, that our method works reliably,” added Sebastian Böcker. Also, the COSMIC enables significant acceleration for searching new and interesting substances, as the screening can be executed automatically and in a shorter amount of time, with no manual effort. Böcker anticipates that in the future, it would be possible to clarify thousands of new molecular structures through this method.

Journal Reference:

Hoffmann, M. A., et al. (2021) High-confidence structural annotation of metabolites absent from spectral libraries. Nature Biotechnology. doi.org/10.1038/s41587-021-01045-9.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.