[ad_1]
Researchers at the University of California at San Diego have developed an approach that uses machine learning to identify and predict which genes make infectious bacteria resistant to antibiotics. The approach has been tested on Mycobacterium tuberculosis – the bacteria responsible for tuberculosis in humans. He identified 33 known genes and 24 new genes for antibiotic resistance in these bacteria.
The researchers explain that this approach can be used to fight against other pathogens responsible for infections, including staphylococcus and bacteria responsible for urinary tract infections, pneumonia and meningitis. The work was recently published in Nature Communications.
"Knowing which genes conferring antibiotic resistance could change the way infectious diseases are treated in the future," said coauthor author Jonathan Monk, a research scientist in the Department of Bioengineering at UC San Diego. "For example, if there is a persistent infection of TB in the clinic, doctors can sequence this strain, look at its genes and determine which antibiotics it resists and which ones it is susceptible to, then prescribe the right antibiotic for this strain. "
"This could pave the way for a personalized treatment of your pathogen.Each strain is different and should potentially be treated differently," said coauthor author Bernhard Palsson, professor of bioengineering at Galletti's UC Engineering School. San Diego Jacobs. "Through this pan-genome automatic learning badysis – the complete set of all genes from all strains of a bacterial species – we can better understand the properties that make these strains different."
The team formed a machine learning algorithm using genome sequences and phenotypes – observable traits or physical characteristics, such as antibiotic resistance – of more than 1,500 strains of M. tuberculosis. From these inputs, the algorithm predicted a set of genes and variant forms of these genes, called alleles, that cause resistance to antibiotics. 33 were validated with known antibiotic resistance genes, the remaining 24 were new predictions that have not yet been experimentally tested.
The researchers then badyzed the predictions of the algorithm and identified combinations of alleles that could interact with each other and make a strain resistant to antibiotics. They also mapped these alleles on crystalline structures of M. tuberculosis Proteins (published in the Protein Data Bank). They discovered that some of these alleles appeared in some structural regions of proteins.
"We performed interaction and structure badyzes to deepen and develop more complex hypotheses about how these genes might contribute to antibiotic resistance phenotypes," said first author, Erol Kavvas, a Ph.D. bioengineering. student in the Palsson research group. "These results could facilitate future experimental research to determine whether the structural grouping of these alleles plays a role in attributing antibiotic resistance."
The results of this study are all calculated. The team is therefore looking to work with experimental researchers to test whether the 24 new genes predicted by the algorithm actually confer antibiotic resistance at home. M. tuberculosis.
Future studies will involve the application of the team's automatic learning to the main infectious bacteria, ESKAPE pathogens: Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa and Enterobacter species. In a next step, the team integrates genome-wide metabolic network models into their machine learning approach to understand the mechanisms underlying the evolution of antibiotic resistance.
Source of the story:
Material provided by University of California – San Diego. Original written by Liezel Labios. Note: Content can be changed for style and length.
Source link
