Site navigation

Aberdeen University Uses Machine Learning to Fight Food Poisoning

Michael Behr

,

bacteria

Researchers used a new machine learning technique called Minimal Multilocus Distance to parse vast amounts of data and help map bacterial genomes.

By using a new machine learning method, University of Aberdeen researchers have developed a technique to make identifying the source of food poisoning or infection faster and more accurate.

According to an article published in Scientific Reports, the new technique could match campylobacter and potentially other common foodborne pathogens more accurately to their source of origin within a significantly reduced timeframe.

The new approach to diagnosis takes advantage of advances in Whole Genome Sequencing (WGS), where the complete DNA sequence of an organism’s genome can be obtained at a single time. It builds an accurate picture of a bacterial infection – the kind of bacteria and its origins.

WGS has been touted as a key factor in personalised medicine, where treatments are tailored to each individual based on their genetic code.

However, the methods are still in their infancy, meaning that the information locked in the DNA cannot be fully utilised to control infections.

Aberdeen’s scientists used a new machine learning method in their technique – the Minimal Multilocus Distance (MMD) method. This can be used to ‘train’ a computer to identify likely sources with a high degree of accuracy.

Campylobacter bacteria are responsible for more foodborne illness than any other organism – estimates suggest that around 500,000 people in the UK are affected by it every year, 100 fatally.

However, pinpointing the source of infection is often difficult since the bacteria can originate from several animal reservoirs including cattle, chicken, pigs, sheep and wild birds.

Recommended

The study was led by Dr Francisco Perez Reche and Professor Norval Strachan, both from the University of Aberdeen’s departments of Physics and Biological Sciences.

Dr Perez Reche said: “There are a number of existing methods to calculate the likely source of infection but in order to work effectively, they either use only part of the genome sequencing, meaning results are less targeted, or if they use the whole genome, the calculations can take up to two days to perform.

“When dealing with an outbreak of infection, speed and accuracy of identifying the likely source are key. Our MMD method trains the computer to identify likely sources of origin of a campylobacter infection within seconds.”

Professor Strachan added: “This has the potential to rapidly provide information on the potential source of infection and could be used to inform strategies to reduce food poisoning.”

While the team have demonstrated that the technique works at a theoretical level, further research is needed to build it into technology that could be at the fingertips of health protection professionals.

The researchers also examined applications for technology beyond food poisoning. According to them, the technique could be applied to human evolution by matching individuals to populations to determine the geographical distribution of different species and to determine the type of breast cancer tumours.

The study was supported by the Scottish Government (Rural Environment and Analytical Services Division) and Food Standards Scotland.

Michael Behr

Senior Staff Writer

Latest News

Infrastructure Trending Articles
Cybersecurity Editor's Picks Government
Events Featured Social Media
%d bloggers like this: