AI Case Study

Healx's scientist predicts effectiveness of combinations of antibiotics using machine learning

Healx's scientist Daniel Mason has developed a tool that uses machine learning to predict the effectiveness of combinations of antibiotics. Developed during his postdoctoral research associate position at the University of Cambridge under the supervision of Dr Andreas Bender, Chief Technology Officer at Healx, the tool aims to reduce the time and resources spent on experimental screening for new treatments. The Combination Synergy Estimation (CoSynE) is based on known structure of compounds and together with compound combination experimental screening data it can predict the activity of new compound combinations.

Industry

Healthcare

Pharmaceuticals And Biotech

Project Overview

"Daniel Mason, Molecular Informatics Data Scientist at Healx, has developed a machine learning approach to predict the effectiveness of combinations of antibiotics, potentially reducing the time and resources spent on experimental screening for new treatments. Such an approach also shows potential for repurposing approved drugs for novel combination treatments.

Daniel developed the tool, called Combination Synergy Estimation (CoSynE), during his postdoctoral research associate position at the University of Cambridge. CoSynE uses compound combination experimental screening data, together with knowledge of the structure of compounds, to predict whether or not a novel combination is likely to exert a synergistic effect, that is, activity greater than the sum of its parts. These predictions can then be used to prioritise the testing of new combinations in subsequent assays.

The approach has also been used with a number of other experimental combination datasets, and is being investigated for use in other areas such as drug repurposing.

Daniel worked under the supervision of Dr Andreas Bender, Chief Technology Officer at Healx. Dr Bender is a Reader for Molecular Informatics with the Centre for Molecular Science Informatics at the Department of Chemistry of the University of Cambridge.

The work described here is published in the the Journal of Medicinal Chemistry." (healx)

According to the authors of the paper:
"Here, we present a computational framework that uses substructure profiles derived from the molecular structures of drugs and predicts antibiotic interactions. Using a previously published data set of 153 drug pairs, we showed that substructure profiles are useful in predicting synergy. We experimentally measured the interaction of 123 new drug pairs, as a prospective validation set for our approach, and identified 37 new synergistic pairs. Of the 12 pairs predicted to be synergistic, 10 were experimentally validated, corresponding to a 2.8-fold enrichment. Having thus validated our methodology, we produced a compendium of interaction predictions for all pairwise combinations among 100 antibiotics."

Reported Results

Their "methodology can make reliable antibiotic interaction predictions for any antibiotic pair within the applicability domain of the model since it solely requires chemical structures as an input." (paper)

Technology

Function

R And D

Core Research And Development

Background

"Combination antibiotic therapies are clinically important in the fight against bacterial infections. However, the search space of drug combinations is large, making the identification of effective combinations a challenging task." (paper)

Benefits

Data

Compound combination experimental screening data