Machine learning identifies signatures of host adaptation in the bacterial pathogen Salmonella enterica

A guide to machine learning for bacterial host attribution using genome sequence data

With the ever-expanding number of available sequences from bacterial genomes, and the expectation that this data type will be the primary one generated from both diagnostic and research laboratories for the foreseeable future, then there is both an opportunity and a need to evaluate how effectively computational approaches can be used within bacterial genomics to predict and understand complex phenotypes, such as pathogenic potential and host source. This article applied various quantitative methods such as diversity indexes, pangenome-wide association studies (GWAS) and dimensionality reduction techniques to better understand the data and then compared how well unsupervised and supervised machine learning (ML) methods could predict the source host of the isolates. The study uses the example of the pangenomes of 1203 Salmonella enterica serovar Typhimurium isolates in order to predict 'host of isolation' using these different methods. The article is aimed as a review of recent applications of ML in infection biology, but also, by working through this specific dataset, it allows discussion of the advantages and drawbacks of the different techniques. As with all such sub-population studies, the biological relevance will be dependent on the quality and diversity of the input data. Given this major caveat, we show that supervised ML has the potential to add real value to interpretation of bacterial genomic data, as it can provide probabilistic outcomes for important phenotypes, something that is very difficult to achieve with the other methods.

Predicting antimicrobial resistance in Pseudomonas aeruginosa with machine learning-enabled molecular diagnostics

Limited therapy options due to antibiotic resistance underscore the need for optimization of current diagnostics. In some bacterial species, antimicrobial resistance can be unambiguously predicted based on their genome sequence. In this study, we sequenced the genomes and transcriptomes of 414 drug-resistant clinical Pseudomonas aeruginosa isolates. By training machine learning classifiers on information about the presence or absence of genes, their sequence variation, and expression profiles, we generated predictive models and identified biomarkers of resistance to four commonly administered antimicrobial drugs. Using these data types alone or in combination resulted in high (0.8-0.9) or very high (> 0.9) sensitivity and predictive values. For all drugs except for ciprofloxacin, gene expression information improved diagnostic performance. Our results pave the way for the development of a molecular resistance profiling tool that reliably predicts antimicrobial susceptibility based on genomic and transcriptomic markers. The implementation of a molecular susceptibility test system in routine microbiology diagnostics holds promise to provide earlier and more detailed information on antibiotic resistance profiles of bacterial pathogens and thus could change how physicians treat bacterial infections.

An African Salmonella Typhimurium ST313 sublineage with extensive drug-resistance and signatures of host adaptation

Bloodstream infections by Salmonella enterica serovar Typhimurium constitute a major health burden in sub-Saharan Africa (SSA). These invasive non-typhoidal (iNTS) infections are dominated by isolates of the antibiotic resistance-associated sequence type (ST) 313. Here, we report emergence of ST313 sublineage II.1 in the Democratic Republic of the Congo. Sublineage II.1 exhibits extensive drug resistance, involving a combination of multidrug resistance, extended spectrum β-lactamase production and azithromycin resistance. ST313 lineage II.1 isolates harbour an IncHI2 plasmid we name pSTm-ST313-II.1, with one isolate also exhibiting decreased ciprofloxacin susceptibility. Whole genome sequencing reveals that ST313 II.1 isolates have accumulated genetic signatures potentially associated with altered pathogenicity and host adaptation, related to changes observed in biofilm formation and metabolic capacity. Sublineage II.1 emerged at the beginning of the 21st century and is involved in on-going outbreaks. Our data provide evidence of further evolution within the ST313 clade associated with iNTS in SSA.

Machine Learning Approaches for Epidemiological Investigations of Food-Borne Disease Outbreaks

Foodborne diseases (FBDs) are infections of the gastrointestinal tract caused by foodborne pathogens (FBPs) such as bacteria [Salmonella, Listeria monocytogenes and Shiga toxin-producing E. coli (STEC)] and several viruses, but also parasites and some fungi. Artificial intelligence (AI) and its sub-discipline machine learning (ML) are re-emerging and gaining an ever increasing popularity in the scientific community and industry, and could lead to actionable knowledge in diverse ranges of sectors including epidemiological investigations of FBD outbreaks and antimicrobial resistance (AMR). As genotyping using whole-genome sequencing (WGS) is becoming more accessible and affordable, it is increasingly used as a routine tool for the detection of pathogens, and has the potential to differentiate between outbreak strains that are closely related, identify virulence/resistance genes and provide improved understanding of transmission events within hours to days. In most cases, the computational pipeline of WGS data analysis can be divided into four (though, not necessarily consecutive) major steps: de novo genome assembly, genome characterization, comparative genomics, and inference of phylogeny or phylogenomics. In each step, ML could be used to increase the speed and potentially the accuracy (provided increasing amounts of high-quality input data) of identification of the source of ongoing outbreaks, leading to more efficient treatment and prevention of additional cases. In this review, we explore whether ML or any other form of AI algorithms have already been proposed for the respective tasks and compare those with mechanistic model-based approaches.

Computational Health Engineering Applied to Model Infectious Diseases and Antimicrobial Resistance Spread

Infectious diseases are the primary cause of mortality worldwide. The dangers of infectious disease are compounded with antimicrobial resistance, which remains the greatest concern for human health. Although novel approaches are under investigation, the World Health Organization predicts that by 2050, septicaemia caused by antimicrobial resistant bacteria could result in 10 million deaths per year. One of the main challenges in medical microbiology is to develop novel experimental approaches, which enable a better understanding of bacterial infections and antimicrobial resistance. After the introduction of whole genome sequencing, there was a great improvement in bacterial detection and identification, which also enabled the characterization of virulence factors and antimicrobial resistance genes. Today, the use of in silico experiments jointly with computational and machine learning offer an in depth understanding of systems biology, allowing us to use this knowledge for the prevention, prediction, and control of infectious disease. Herein, the aim of this review is to discuss the latest advances in human health engineering and their applicability in the control of infectious diseases. An in-depth knowledge of host–pathogen–protein interactions, combined with a better understanding of a host’s immune response and bacterial fitness, are key determinants for halting infectious diseases and antimicrobial resistance dissemination.

Genomic correlates of extraintestinal infection are linked with changes in cell morphology in Campylobacter jejuni

Campylobacter jejuni is the most common cause of bacterial diarrheal disease in the world. Clinical outcomes of infection can range from asymptomatic infection to life-threatening extraintestinal infections. This variability in outcomes for infected patients has raised questions as to whether genetic differences between C. jejuni isolates contribute to their likelihood of causing severe disease. In this study, we compare the genomes of ten C. jejuni isolates that were implicated in extraintestinal infections with reference gastrointestinal isolates, in order to identify unusual patterns of sequence variation associated with infection outcome. We identified a collection of genes that display a higher burden of uncommon mutations in invasive isolates compared with gastrointestinal close relatives, including some that have been previously linked to virulence and invasiveness in C. jejuni. Among the top genes identified were mreB and pgp1, which are both involved in determining cell shape. Electron microscopy confirmed morphological differences in isolates carrying unusual sequence variants of these genes, indicating a possible relationship between extraintestinal infection and changes in cell morphology.

Adding function to the genome of African Salmonella Typhimurium ST313 strain D23580

Salmonella Typhimurium sequence type (ST) 313 causes invasive nontyphoidal Salmonella (iNTS) disease in sub-Saharan Africa, targeting susceptible HIV+, malarial, or malnourished individuals. An in-depth genomic comparison between the ST313 isolate D23580 and the well-characterized ST19 isolate 4/74 that causes gastroenteritis across the globe revealed extensive synteny. To understand how the 856 nucleotide variations generated phenotypic differences, we devised a large-scale experimental approach that involved the global gene expression analysis of strains D23580 and 4/74 grown in 16 infection-relevant growth conditions. Comparison of transcriptional patterns identified virulence and metabolic genes that were differentially expressed between D23580 versus 4/74, many of which were validated by proteomics. We also uncovered the S. Typhimurium D23580 and 4/74 genes that showed expression differences during infection of murine macrophages. Our comparative transcriptomic data are presented in a new enhanced version of the Salmonella expression compendium, SalComD23580: http://bioinf.gen.tcd.ie/cgi-bin/salcom_v2.pl. We discovered that the ablation of melibiose utilization was caused by three independent SNP mutations in D23580 that are shared across ST313 lineage 2, suggesting that the ability to catabolize this carbon source has been negatively selected during ST313 evolution. The data revealed a novel, to our knowledge, plasmid maintenance system involving a plasmid-encoded CysS cysteinyl-tRNA synthetase, highlighting the power of large-scale comparative multicondition analyses to pinpoint key phenotypic differences between bacterial pathovariants.

Prediction of antibiotic resistance in Escherichia coli from large-scale pan-genome data

The emergence of microbial antibiotic resistance is a global health threat. In clinical settings, the key to controlling spread of resistant strains is accurate and rapid detection. As traditional culture-based methods are time consuming, genetic approaches have recently been developed for this task. The detection of antibiotic resistance is typically made by measuring a few known determinants previously identified from genome sequencing, and thus requires the prior knowledge of its biological mechanisms. To overcome this limitation, we employed machine learning models to predict resistance to 11 compounds across four classes of antibiotics from existing and novel whole genome sequences of 1936 E. coli strains. We considered a range of methods, and examined population structure, isolation year, gene content, and polymorphism information as predictors. Gradient boosted decision trees consistently outperformed alternative models with an average accuracy of 0.91 on held-out data (range 0.81-0.97). While the best models most frequently employed gene content, an average accuracy score of 0.79 could be obtained using population structure information alone. Single nucleotide variation data were less useful, and significantly improved prediction only for two antibiotics, including ciprofloxacin. These results demonstrate that antibiotic resistance in E. coli can be accurately predicted from whole genome sequences without a priori knowledge of mechanisms, and that both genomic and epidemiological data can be informative. This paves way to integrating machine learning approaches into diagnostic tools in the clinic.

A k-mer-based method for the identification of phenotype-associated genomic biomarkers and predicting phenotypes of sequenced bacteria

We have developed an easy-to-use and memory-efficient method called PhenotypeSeeker that (a) identifies phenotype-specific k-mers, (b) generates a k-mer-based statistical model for predicting a given phenotype and (c) predicts the phenotype from the sequencing data of a given bacterial isolate. The method was validated on 167 Klebsiella pneumoniae isolates (virulence), 200 Pseudomonas aeruginosa isolates (ciprofloxacin resistance) and 459 Clostridium difficile isolates (azithromycin resistance). The phenotype prediction models trained from these datasets obtained the F1-measure of 0.88 on the K. pneumoniae test set, 0.88 on the P. aeruginosa test set and 0.97 on the C. difficile test set. The F1-measures were the same for assembled sequences and raw sequencing data; however, building the model from assembled genomes is significantly faster. On these datasets, the model building on a mid-range Linux server takes approximately 3 to 5 hours per phenotype if assembled genomes are used and 10 hours per phenotype if raw sequencing data are used. The phenotype prediction from assembled genomes takes less than one second per isolate. Thus, PhenotypeSeeker should be well-suited for predicting phenotypes from large sequencing datasets. PhenotypeSeeker is implemented in Python programming language, is open-source software and is available at GitHub (https://github.com/bioinfo-ut/PhenotypeSeeker/).