Search

In this study, we explore how transformer models, which are known for their attention mechanisms, can improve pathogen prediction in pastured poultry farming. By combining farm management practices with microbiome data, our model outperforms traditional prediction methods in terms of the F1 score—an evaluation metric for model performance—thus fulfilling an essential need in predictive microbiology. Additionally, the emphasis is on making our model’s predictions explainable. We introduce a novel approach for identifying feature importance using the model’s attention matrix and the PageRank algorithm, offering insights that enhance our comprehension of established techniques such as DeepLIFT. Our results showcase the efficacy of transformer models in pathogen prediction for food safety and mark a noteworthy contribution to the progress of explainable AI within the biomedical sciences. This study sheds light on the impact of effective farm management practices and highlights the importance of technological advancements in ensuring food safety.

Polyamines are polycations derived from amino acids that play an important role in proliferation and growth in almost all living cells. In Streptococcus pneumoniae (the pneumococcus), modulation of polyamine metabolism not only plays an important regulatory role in central metabolism, but also impacts virulence factors such as the capsule and stress responses that affect survival in the host. However, functional annotation of enzymes from the polyamine biosynthesis pathways in the pneumococcus is based predominantly on computational prediction. In this study, we cloned SP_0166, predicted to be a pyridoxal-dependent decarboxylase, from the Orn/Lys/Arg family pathway in S. pneumoniae TIGR4 and expressed and purified the recombinant protein. We performed biochemical characterization of the recombinant SP_0166 and confirmed the substrate specificity. For polyamine analysis, we developed a simultaneous quantitative method using hydrophilic interaction liquid chromatography (HILIC)-based liquid chromatography–tandem mass spectrometry (LC–MS/MS) without derivatization. SP_0166 has apparent Km, kcat, and kcat/Km values of 11.3 mM, 715,053 min−1, and 63,218 min−1 mM−1, respectively, with arginine as a substrate at pH 7.5. We carried out inhibition studies of SP_0166 enzymatic activity with arginine as a substrate using chemical inhibitors DFMO and DFMA. DFMO is an irreversible inhibitor of ornithine decarboxylase activity, while DFMA inhibits arginine decarboxylase activity. Our findings confirm that SP_0166 is inhibited by DFMA and DFMO, impacting agmatine production. The use of arginine as a substrate revealed that the synthesis of putrescine by agmatinase and N-carbamoylputrescine by agmatine deiminase were both affected and inhibited by DFMA. This study provides experimental validation that SP_0166 is an arginine decarboxylase in pneumococci.

Streptococcus pneumoniae (Spn), a Gram-positive bacterium, poses a significant threat to human health, causing mild respiratory infections to severe invasive conditions. Despite the availability of vaccines, challenges persist due to serotype replacement and antibiotic resistance, emphasizing the need for alternative therapeutic strategies. This study explores the intriguing role of polyamines, ubiquitous, small organic cations, in modulating virulence factors, especially the capsule, a crucial determinant of Spn’s pathogenicity. Using chemical inhibitors, difluoromethylornithine (DFMO) and AMXT 1501, this research unveils distinct regulatory effects on the gene expression of the Spn D39 serotype in response to altered polyamine homeostasis. DFMO inhibits polyamine biosynthesis, disrupting pathways associated with glucose import and the interconversion of sugars. In contrast, AMXT 1501, targeting polyamine transport, enhances the expression of polyamine and glucose biosynthesis genes, presenting a novel avenue for regulating the capsule independent of glucose availability. Despite ample glucose availability, AMXT 1501 treatment downregulates the glycolytic pathway, fatty acid synthesis, and ATP synthase, crucial for energy production, while upregulating two-component systems responsible for stress management. This suggests a potential shutdown of energy production and capsule biosynthesis, redirecting resources towards stress management. Following DFMO and AMXT 1501 treatments, countermeasures, such as upregulation of stress response genes and ribosomal protein, were observed but appear to be insufficient to overcome the deleterious effects on capsule production. This study highlights the complexity of polyamine-mediated regulation in S. pneumoniae, particularly capsule biosynthesis. Our findings offer valuable insights into potential therapeutic targets for modulating capsules in a polyamine-dependent manner, a promising avenue for intervention against S. pneumoniae infections.

Salmonella spp., a leading cause of foodborne illness, is a formidable global menace due to escalating antimicrobial resistance (AMR). The evaluation of minimum inhibitory concentration (MIC) for antimicrobials is critical for characterizing AMR. The current whole genome sequencing (WGS)-based approaches for predicting MIC are hindered by both computational and feature identification constraints. We propose an innovative methodology called the “Genome Feature Extractor Pipeline” that integrates traditional machine learning (random forest, RF) with deep learning models (multilayer perceptron (MLP) and DeepLift) for WGS-based MIC prediction. We used a dataset from the National Antimicrobial Resistance Monitoring System (NARMS), comprising 4500 assembled genomes of nontyphoidal Salmonella, each annotated with MIC metadata for 15 antibiotics. Our pipeline involves the batch downloading of annotated genomes, the determination of feature importance using RF, Gini-index-based selection of crucial 10-mers, and their expansion to 20-mers. This is followed by an MLP network, with four hidden layers of 1024 neurons each, to predict MIC values. Using DeepLift, key 20-mers and associated genes influencing MIC are identified. The 10 most significant 20-mers for each antibiotic are listed, showcasing our ability to discern genomic features affecting Salmonella MIC prediction with enhanced precision. The methodology replaces binary indicators with k-mer counts, offering a more nuanced analysis. The combination of RF and MLP addresses the limitations of the existing WGS approach, providing a robust and efficient method for predicting MIC values in Salmonella that could potentially be applied to other pathogens.

Bovine Respiratory Disease (BRD) is a multifactorial condition affecting cattle worldwide resulting in high rates of morbidity and mortality. The disease can be triggered by Bovine Herpesvirus-1 (BoHV-1) infection, stress, and the subsequent proliferation and lung colonization by commensal bacteria such as Mannheimia haemolytica, ultimately inducing severe pneumonic inflammation. Due to its polymicrobial nature, the study of BRD microbes requires co-infection models. While several past studies have mostly focused on the effects of co-infection on host gene expression, we focused on the relationship between BRD pathogens during co-infection, specifically on M. haemolytica’s effect on BoHV-1 replication. This study shows that M. haemolytica negatively impacts BoHV-1 replication in a dose-dependent manner in different in vitro models. The negative effect was observed at very low bacterial doses while increasing the viral dose counteracted this effect. Viral suppression was also dependent on the time at which each microbe was introduced to the cell culture. While acidification of the culture medium did not grossly affect cell viability, it significantly inhibited viral replication. We conclude that M. haemolytica and BoHV-1 interaction is dose and time-sensitive, wherein M. haemolytica proliferation induces significant viral suppression when the viral replication program is not fully established.

Zoonotic diseases or zoonoses are infections due to the natural transmission of pathogens between species (animals and humans). More than 70% of emerging infectious diseases are attributed to animal origin. Artificial Intelligence (AI) models have been used for studying zoonotic pathogens and the factors that contribute to their spread. The aim of this literature survey is to synthesize and analyze machine learning, and deep learning approaches applied to study zoonotic diseases to understand predictive models to help researchers identify the risk factors, and develop mitigation strategies. Based on our survey findings, machine learning and deep learning are commonly used for the prediction of both foodborne and zoonotic pathogens as well as the factors associated with the presence of the pathogens.

Due to nutritional benefits and perceived humane ways of treating the animals, the demand for antibiotic-free pastured poultry chicken has continued to be steadily rise. Despite the non-usage of antibiotics in pastured poultry broiler production, antibiotic resistance (AR) is reported in zoonotic poultry pathogens. However, factors that drive multidrug resistance (MDR) in pastured poultry are not well understood. In this study, we used machine learning and deep learning approaches to predict farm management practices and physicochemical properties of feces and soil that drive MDR in zoonotic poultry pathogens. Antibiotic use in agroecosystems is known to contribute to resistance. Evaluation of the development of resistance in environments that are free of antibiotics such as the all-natural, antibiotic-free, pastured poultry production systems described here is critical to understand the background AR in the absence of any selection pressure, i.e., basal levels of resistance. We analyzed 1635 preharvest (feces and soil) samples collected from forty-two pastured poultry flocks and eleven farms in the Southeastern United States. CDC National Antimicrobial Resistance Monitoring System guidelines were used to determine antimicrobial/multidrug resistance profiles of Salmonella, Listeria, and Campylobacter. A combination of two traditional machine learning (RandomForest and XGBoost) and three deep learning (Multi-layer Perceptron, Generative Adversarial Network, and Auto-Encoder) approaches identified critical farm management practices and environmental variables that drive multidrug resistance in poultry pathogens in broiler production systems that represents background resistance. This study enumerates management practices that contribute to AR and makes recommendations to potentially mitigate multidrug resistance and the prevalence of Salmonella and Listeria in pastured poultry.

Infections due to Streptococcus pneumoniae, a commensal in the nasopharynx, still claim a significant number of lives worldwide. Genome plasticity, antibiotic resistance, and limited serotype coverage of the available polysaccharide-based conjugate vaccines confounds therapeutic interventions to limit the spread of this pathogen. Pathogenic mechanisms that allow successful adaption and persistence in the host could be potential innovative therapeutic targets. Polyamines are ubiquitous polycationic molecules that regulate many cellular processes. We previously reported that deletion of polyamine transport operon potABCD, which encodes a putrescine/spermidine transporter (ΔpotABCD), resulted in an unencapsulated attenuated phenotype. Here, we characterize the transcriptome, metabolome, and stress responses of polyamine transport-deficient S. pneumoniae. Compared with the wild-type strain, the expression of genes involved in oxidative stress responses and the nucleotide sugar metabolism was reduced, while expression of genes involved in the Leloir, tagatose, and pentose phosphate pathways was higher in ΔpotABCD. A metabolic shift towards the pentose phosphate pathway will limit the synthesis of precursors of capsule polysaccharides. Metabolomics results show reduced levels of glutathione and pyruvate in the mutant. Our results also show that the potABCD operon protects pneumococci against hydrogen peroxide and nitrosative stress. Our findings demonstrate the importance of polyamine transport in pneumococcal physiology that could impact in vivo fitness. Thus, polyamine transport in pneumococci represents a novel target for therapeutic interventions.

Polyamines such as putrescine, cadaverine, and spermidine are small cationic molecules that play significant roles in cellular processes, including bacterial stress responses and host–pathogen interactions. Streptococcus pneumoniae is an opportunistic human pathogen, which causes several diseases that account for significant morbidity and mortality worldwide. As it transits through different host niches, S. pneumoniae is exposed to and must adapt to different types of stress in the host microenvironment. We earlier reported that S. pneumoniae TIGR4, which harbors an isogenic deletion of an arginine decarboxylase (ΔspeA), an enzyme that catalyzes the synthesis of agmatine in the polyamine synthesis pathway, has a reduced capsule. Here, we report the impact of arginine decarboxylase deletion on pneumococcal stress responses. Our results show that ΔspeA is more susceptible to oxidative, nitrosative, and acid stress compared to the wild-type strain. Gene expression analysis by qRT-PCR indicates that thiol peroxidase, a scavenger of reactive oxygen species and aguA from the arginine deiminase system, could be important for peroxide stress responses in a polyamine-dependent manner. Our results also show that speA is essential for endogenous hydrogen peroxide and glutathione production in S. pneumoniae. Taken together, our findings demonstrate the critical role of arginine decarboxylase in pneumococcal stress responses that could impact adaptation and survival in the host.

Bovine herpesvirus 1 (BoHV-1) is one of several microbes that contributes to the development of the bovine respiratory disease (BRD) and can also induce abortions in cattle. As other alpha-herpesvirinae subfamily members, BoHV-1 efficiently replicates in many cell types and subsequently establishes a life-long latent infection in sensory neurons. BoHV-1 encodes more than 70 proteins that are expressed in a well-defined manner during productive infection. However, in silico open reading frame (ORF) prediction of the BoHV-1 genome suggests that the virus may encode more than one hundred proteins. In this study we used mass spectrometry followed by proteogenomic mapping to reveal the existence of 92 peptides that map to previously un-annotated regions of the viral genome. Twenty-one of the newly termed “intergenic peptides” were predicted to have a viable ORF around them. Twelve of these produced an mRNA transcript as demonstrated by strand-specific RT-PCR. We further characterized the 5′ and 3′ termini of one mRNA transcript, ORF-A, and detected a 55 kDa protein produced during active infection using a custom-synthesized antibody. We conclude that the coding potential of BoHV-1 is underestimated.

Invasive infections caused by Streptococcus pneumoniae, a commensal in the nasopharynx, pose significant risk to human health. Limited serotype coverage by the available polysaccharide-based conjugate vaccines coupled with increasing incidence of antibiotic resistance complicates therapeutic strategies. Bacterial physiology and metabolism that allows pathogens to adapt to the host are a promising avenue for the discovery of novel therapeutics. Intracellular polyamine concentrations are tightly regulated by biosynthesis, transport and degradation. We previously reported that deletion of cadA, a gene that encodes for lysine decarboxylase, an enzyme that catalyzes cadaverine synthesis results in an attenuated phenotype. Here, we report the impact of cadA deletion on pneumococcal capsule and protein expression. Our data show that genes for polyamine biosynthesis and transport are downregulated in ∆cadA. Immunoblot assays show reduced capsule in ∆cadA. Reduced capsule synthesis could be due to reduced transcription and availability of precursors for synthesis. The capsule is the predominant virulence factor in pneumococci and is critical for evading opsonophagocytosis and its loss in ∆cadA could explain the reported attenuation in vivo. Results from this study show that capsule synthesis in pneumococci is regulated by polyamine metabolism, which can be targeted for developing novel therapies.

Bovine herpesvirus (BoHV) type 1 is an important agricultural pathogen that infects cattle and other ruminants worldwide. Acute infection of the oro-respiratory tract leads to immune suppression and allows commensal bacteria to infect an otherwise healthy lower respiratory tract. This condition is known as the Bovine Respiratory Disease (BRD). BoHV-1 latently infects the host for life and periodical stress events re-initiate BRD, translating into high morbidity and large economic losses. To gain a better understanding of the biology of BoHV-1 and the disease it causes, we elucidated the protein composition of extracellular virions using liquid chromatography-mass spectrometry analysis. We detected 33 viral proteins, including the expected proteins of the nucleocapsid and envelope as well as other regulatory proteins present in the viral tegument. In addition to viral proteins, we have also identified packaged proteins of host origin. This constitutes the first proteomic characterization of the BoHV virion.

The distinct stages of cotton fiber development and maturation serve as a single-celled model for studying the molecular mechanisms of plant cell elongation, cell wall development and cellulose biosynthesis. However, this model system of plant cell development is compromised for proteomic studies due to a lack of an efficient protein extraction method during the later stages of fiber development, because of a recalcitrant cell wall and the presence of abundant phenolic compounds. Here, we compared the quality and quantities of proteins extracted from 25 dpa (days post anthesis) fiber with multiple protein extraction methods and present a comprehensive quantitative proteomic study of fiber development from 10 dpa to 25 dpa. Comparative analysis using a label-free quantification method revealed 287 differentially-expressed proteins in the 10 dpa to 25 dpa fiber developmental period. Proteins involved in cell wall metabolism and regulation, cytoskeleton development and carbohydrate metabolism among other functional categories in four fiber developmental stages were identified. Our studies provide protocols for protein extraction from maturing fiber tissues for mass spectrometry analysis and expand knowledge of the proteomic profile of cotton fiber development.