Rapid and accurate identification of Escherichia coli STEC O157:H7 by mass spectrometry, artificial intelligence and detection of specific biomarkers peaks.

The different pathotypes of Escherichia can produce a large number of human diseases. Surveillance becomes complex since their differentiation are not easy. Particularly, the detection of Shiga toxin-producing Escherichia coli (STEC) serotype O157:H7 consists of stool culture of a diarrheal sample in enrichment and/or selective media, identification of presumptive colonies and confirmation by Multiplex PCR technique for the genotypic characterization of serogroup O157 and Shiga toxins (stx1 and stx2), in addition to the traditional biochemical identification. All of these procedures are laborious, require a certain level of training, are time consuming and expensive. Among the currently most widely used methodologies, MALDI-TOF MS mass spectrometry (matrix-assisted laser desorption/ionization with time-of-flight mass detection), allows a quick and easy way to obtain a protein spectrum of a microorganism, not only in order to identify the genus and species, but also the discovery of potential biomarker peaks of a certain characteristic. In the present work, the information obtained from 60 clinical isolates was used to detect peptide fingerprints of STEC O157:H7 and other diarrheagenic E. coli. The differences found in the protein profiles of the different pathotypes established the foundations for the development and evaluation of classification models through automated training. The application of the Biomarkers in combination with the predictive models on a new set of samples (n=142), achieved 99.3% of correct classifications, allowing the distinction between STEC O157:H7 isolates from the other diarrheal Escherichia coli. Therefore, given that STEC O157:H7 is the main causal agent of haemolytic uremic syndrome and based on the performance values obtained in the present work (Sensitivity=98.5% and Specificity=100%), this development could be a useful tool for diagnosis of the disease in clinical microbiology laboratories. ### Competing Interest Statement The authors have declared no competing interest.