Bacterial colony size growth estimation by deep learning.

The bacterial growth rate is important for pathogenicity and food safety. Therefore, the study of bacterial growth rate over time can provide important data from a medical and veterinary point of view. We trained convolutional neural networks (CNNs) on manually annotated solid medium cultures to detect bacterial colonies as accurately as possible. Predictions of bacterial colony size and growth rate were estimated from image sequences of independent Staphylococcus aureus cultures using trained CNNs. A simple linear model for control cultures with less than 150 colonies estimated that the mean growth rate was 60.3 [Formula: see text] for the first 24 h. Analyzing with a mixed effect model that also takes into account the effect of culture, smaller values of change in colony size were obtained (control: 51.0 [Formula: see text], rifampicin pretreated: 36.5[Formula: see text]). An increase in the number of neighboring colonies clearly reduces the colony growth rate in the control group but less typically in the rifampicin-pretreated group. Based on our results, CNN-based bacterial colony detection and the subsequent analysis of bacterial colony growth dynamics might become an accurate and efficient tool for bacteriological work and research.

A survey on antimicrobial resistance genes of frequently used probiotic bacteria, 1901 to 2022.

BackgroundAntimicrobial resistance (AMR) is caused by AMR determinants, mainly genes (ARGs) in the bacterial genome. Bacteriophages, integrative mobile genetic elements (iMGEs) or plasmids can allow ARGs to be exchanged among bacteria by horizontal gene transfer (HGT). Bacteria, including bacteria with ARGs, can be found in food. Thus, it is conceivable that in the gastrointestinal tract, bacteria from the gut flora could take up ARGs from food.AimThe study objective was to gain insight into the ARG set carried by commonly used probiotic bacteria that may enter the human body with non-fermented foods, fermented foods, or probiotic dietary supplements (FFPs) and to assess ARG mobility.MethodsNext generation sequencing whole genome data from 579 isolates of 12 commonly employed probiotic bacterial species were collected from a public repository. Using bioinformatical tools, ARGs were analysed and linkage with mobile genetic elements assessed.ResultsResistance genes were found in eight bacterial species. The ratios of ARG positive/negative samples per species were: Bifidobacterium animalis (65/0), Lactiplantibacillus plantarum (18/194), Lactobacillus delbrueckii (1/40), Lactobacillus helveticus (2/64), Lactococcus lactis (74/5), Leucoconstoc mesenteroides (4/8), Levilactobacillus brevis (1/46), Streptococcus thermophilus (4/19). In 66% (112/169) of the ARG-positive samples, at least one ARG could be linked to plasmids or iMGEs. No bacteriophage-linked ARGs were found.ConclusionThe finding of potentially mobile ARGs in probiotic strains for human consumption raises awareness of a possibility of ARG HGT in the gastrointestinal tract. In addition to existing recommendations, screening FFP bacterial strains for ARG content and mobility characteristics might be considered.

Mutations Associated with Decreased Susceptibility to Seven Antimicrobial Families in Field and Laboratory-Derived Mycoplasma bovis Strains.

The molecular mechanisms of resistance to fluoroquinolones, tetracyclines, an aminocyclitol, macrolides, a lincosamide, a phenicol, and pleuromutilins were investigated in Mycoplasma bovis For the identification of mutations responsible for the high MICs of certain antibiotics, whole-genome sequencing of 35 M. bovis field isolates and 36 laboratory-derived antibiotic-resistant mutants was performed. In vitro resistant mutants were selected by serial passages of M. bovis in broth medium containing subinhibitory concentrations of the antibiotics. Mutations associated with high fluoroquinolones MICs were found at positions 244 to 260 and at positions 232 to 250 (according to Escherichia coli numbering) of the quinolone resistance-determining regions of the gyrA and parC genes, respectively. Alterations related to elevated tetracycline MICs were described at positions 962 to 967, 1058, 1195, 1196, and 1199 of genes encoding the 16S rRNA and forming the primary tetracycline binding site. Single transversion at position 1192 of the rrs1 gene resulted in a spectinomycin MIC of 256 µg/ml. Mutations responsible for high macrolide, lincomycin, florfenicol, and pleuromutilin antibiotic MICs were identified in genes encoding 23S rRNA. Understanding antibiotic resistance mechanisms is an important tool for future developments of genetic-based diagnostic assays for the rapid detection of resistant M. bovis strains.

Impact Evaluation of Score Classes and Annotation Regions in Deep Learning-Based Dairy Cow Body Condition Prediction.

Body condition scoring is a simple method to estimate the energy supply of dairy cattle. Our study aims to investigate the accuracy with which supervised machine learning, specifically a deep convolutional neural network (CNN), can be used to retrieve body condition score (BCS) classes estimated by an expert. We recorded images of animals' rumps in three large-scale farms using a simple action camera. The images were annotated with classes and three different-sized bounding boxes by an expert. A CNN pretrained model was fine-tuned on 12 and 3 BCS classes. Training in 12 classes with a 0 error range, the Cohen's kappa value yielded minimal agreement between the model predictions and ground truth. Allowing an error range of 0.25, we obtained minimum or weak agreement. With an error range of 0.5, we had strong or almost perfect agreement. The kappa values for the approach trained on three classes show that we can classify all animals into BCS categories with at least moderate agreement. Furthermore, CNNs trained on 3 BCS classes showed a remarkably higher proportion of strong agreement than those trained in 12 classes. The prediction precision when training with various annotation region sizes showed no meaningful differences. The weights of our trained CNNs are freely available, supporting similar works.

Antimicrobial resistance gene lack in tick-borne pathogenic bacteria.

Tick-borne infections, including those of bacterial origin, are significant public health issues. Antimicrobial resistance (AMR), which is one of the most pressing health challenges of our time, is driven by specific genetic determinants, primarily by the antimicrobial resistance genes (ARGs) of bacteria. In our work, we investigated the occurrence of ARGs in the genomes of tick-borne bacterial species that can cause human infections. For this purpose, we processed short/long reads of 1550 bacterial isolates of the genera Anaplasma (n = 20), Bartonella (n = 131), Borrelia (n = 311), Coxiella (n = 73), Ehrlichia (n = 13), Francisella (n = 959) and Rickettsia (n = 43) generated by second/third generation sequencing that have been freely accessible at the NCBI SRA repository. From Francisella tularensis, 98.9% of the samples contained the FTU-1 beta-lactamase gene. However, it is part of the F. tularensis representative genome as well. Furthermore, 16.3% of them contained additional ARGs. Only 2.2% of isolates from other genera (Bartonella: 2, Coxiella: 8, Ehrlichia: 1, Rickettsia: 2) contained any ARG. We found that the odds of ARG occurrence in Coxiella samples were significantly higher in isolates related to farm animals than from other sources. Our results describe a surprising lack of ARGs in these bacteria and suggest that Coxiella species in farm animal settings could play a role in the spread of AMR.

Annotated dataset for deep-learning-based bacterial colony detection.

Quantifying bacteria per unit mass or volume is a common task in various fields of microbiology (e.g., infectiology and food hygiene). Most bacteria can be grown on culture media. The unicellular bacteria reproduce by dividing into two cells, which increases the number of bacteria in the population. Methodologically, this can be followed by culture procedures, which mostly involve determining the number of bacterial colonies on the solid culture media that are visible to the naked eye. However, it is a time-consuming and laborious professional activity. Addressing the automation of colony counting by convolutional neural networks in our work, we have cultured 24 bacteria species of veterinary importance with different concentrations on solid media. A total of 56,865 colonies were annotated manually by bounding boxes on the 369 digital images of bacterial cultures. The published dataset will help developments that use artificial intelligence to automate the counting of bacterial colonies.

Antimicrobial resistance determinants in silage.

Animal products may play a role in developing and spreading antimicrobial resistance in several ways. On the one hand, residues of antibiotics not adequately used in animal farming can enter the human body via food. However, resistant bacteria may also be present in animal products, which can transfer the antimicrobial resistance genes (ARG) to the bacteria in the consumer's body by horizontal gene transfer. As previous studies have shown that fermented foods have a meaningful ARG content, it is indicated that such genes may also be present in silage used as mass feed in the cattle sector. In our study, we aspired to answer what ARGs occur in silage and what mobility characteristics they have? For this purpose, we have analyzed bioinformatically 52 freely available deep sequenced silage samples from shotgun metagenome next-generation sequencing. A total of 16 perfect matched ARGs occurred 54 times in the samples. More than half of these ARGs are mobile because they can be linked to integrative mobile genetic elements, prophages or plasmids. Our results point to a neglected but substantial ARG source in the food chain.

A glimpse of antimicrobial resistance gene diversity in kefir and yoghurt.

Antimicrobial resistance (AMR) is a global threat gaining more and more practical significance every year. The main determinants of AMR are the antimicrobial resistance genes (ARGs). Since bacteria can share genetic components via horizontal gene transfer, even non-pathogenic bacteria may provide ARG to any pathogens which they become physically close to (e.g. in the human gut). In addition, fermented food naturally contains bacteria in high amounts. In this study, we examined the diversity of ARG content in various kefir and yoghurt samples (products, grains, bacterial strains) using a unified metagenomic approach. We found numerous ARGs of commonly used fermenting bacteria. Even with the strictest filter restrictions, we identified ARGs undermining the efficacy of aminocoumarins, aminoglycosides, carbapenems, cephalosporins, cephamycins, diaminopyrimidines, elfamycins, fluoroquinolones, fosfomycins, glycylcyclines, lincosamides, macrolides, monobactams, nitrofurans, nitroimidazoles, penams, penems, peptides, phenicols, rifamycins, tetracyclines and triclosan. In the case of gene lmrD, we detected genetic environment providing mobility of this ARG. Our findings support the theory that during the fermentation process, the ARG content of foods can grow due to bacterial multiplication. The results presented suggest that the starting culture strains of fermented foods should be monitored and selected in order to decrease the intake of ARGs via foods.

Development of molecular methods for the rapid detection of antibiotic susceptibility of Mycoplasma bovis.

Determining the antibiotic susceptibility profile of Mycoplasma bovis isolates in vitro provides the basis for the appropriate choice of antibiotics in the therapy. Traditionally, the antibiotic susceptibility examination of mycoplasmas is technically demanding, time-consuming and rarely performed in diagnostic laboratories. The aim of the present study was to develop rapid molecular assays to determine mutations responsible for elevated minimal inhibitory concentrations (MICs) to fluoroquinolones, tetracyclines, aminocyclitols, macrolides, lincosamides, phenicols and pleuromutilins in M. bovis. The nine mismatch amplification mutation assays (MAMA) and seven high resolution melt (HRM) tests designed in the present study enable the simultaneous detection of these genetic markers. The sensitivity of the assays varied between 102-105 copy numbers/reaction. Cross-reactions with other mycoplasmas occurring in cattle were detected in assays targeting universal regions (e.g. 16S rRNA). Nevertheless, results of the novel method were in accordance with sequence and MICs data of the M. bovis pure cultures. Also, the tests of clinical samples containing high amount of M. bovis DNA were congruent even in the presence of other Mycoplasma spp. The presented method is highly cost-effective and can provide an antibiogram to 12 antibiotics in approximately 3-4 days when previous isolation of M. bovis is applied. In order to assure the proper identification of the genetic markers at issue, the regions examined by the MAMA and HRM tests are overlapping. In conclusion, the developed assays have potential to be used in routine diagnostics for the detection of antibiotic susceptibility in M. bovis.