New software could aid the impact of antibiotic in people

Scientists have developed a new weapon in the battle to safeguard the effectiveness of antibiotics. 

A computer programme has been designed which can quickly analyse bacterial DNA from a patient’s infection and predict which antibiotics will work, and which will fail because of drug resistance.

The software is having trials in three British hospitals to see whether it could help speed up diagnosis of drug-resistant infections and enable doctors to better target the prescription of antibiotics.

The Mykrobe Predictor software, developed by Zamin Iqbal and colleagues at the Wellcome Trust Centre for Human Genetics, University of Oxford, runs on a standard laptop or tablet without the need for any specialist expertise.

It can analyse the entire genetic code of a bacterium in under three minutes, once a bacterial sample has been cultured and its DNA sequenced.

A study on more than 4,500 retrospective patient samples, published in Nature Communications, shows the software accurately detects antibiotic resistance in two life-threatening bacterial infections: Staphylococcus aureus (one form of which causes MRSA) and TB.

Drug resistance poses a major threat to global health and could mean serious and life-threatening infections become impossible to treat. As with other species, bacteria are constantly evolving through changes in their DNA. Some of these make them resistant to certain drugs, meaning they are more likely to survive and pass on their resistant traits to other bacteria.

One of the best ways to prevent the spread of resistant bacteria is to make sure patients with an infection are treated quickly with the right type of antibiotic.

However, to find out which particular strain of bacteria is causing a patient’s infection and which drugs it is resistant to, doctors must carry out drug susceptibility testing, where different antibiotics are applied to the bacteria in a petri dish to see whether they kill it. This process can take days, or even months for slow-growing infections like TB.

Scientists believe they could obtain the same information much faster by looking directly at the DNA sequence of the bacterium for mutations that are known to cause resistance.

However, the interpretation of genetic information often requires a large amount of computing power and the expertise of specialist bio-informaticians.

The software streamlines this process by automating genome analysis, cross-checking the bacterium’s DNA sequence with previous strains to look for resistance-causing mutations and presenting information about the bacteria in an easy-to-understand format.

It was able to detect resistance to the five first-line antibiotics in more than 99% of Staphylococcus aureus cases, matching the performance of traditional drug sensitivity testing.

For TB, where the genetic basis for drug resistance is less well understood, it still matched the performance of current DNA tests (which look at snippets of DNA, but not the whole sequence), detecting 82.6% of resistant infections around five to 16 weeks faster than traditional drug susceptibility testing.

Dr Iqbal, senior author of the paper, said: “Our software manages data quickly and presents the results to doctors and nurses in ways that are easy to understand, so they can instinctively use them to make better treatment decisions.”