A White-Box Machine Learning Approach for Revealing Antibiotic Mechanisms of Action

Protocol for hybrid flux balance, statistical, and machine learning analysis of multi-omic data from the cyanobacterium Synechococcus sp. PCC 7002

Combining a computational framework for flux balance analysis with machine learning improves the accuracy of predicting metabolic activity across conditions, while enabling mechanistic interpretation. This protocol presents a guide to condition-specific metabolic modeling that integrates regularized flux balance analysis with machine learning approaches to extract key features from transcriptomic and fluxomic data. We demonstrate the protocol as applied to Synechococcus sp. PCC 7002; we also outline how it can be adapted to any species or community with available multi-omic data. For complete details on the use and execution of this protocol, please refer to Vijayakumar et al. (2020).

Lanza V, Rodríguez-Beltrán J, Galán JC, San Millán A, Cantón R, Coque TM. Evolutionary Pathways and Trajectories in Antibiotic Resistance

Evolution is the hallmark of life. Descriptions of the evolution of microorganisms have provided a wealth of information, but knowledge regarding "what happened" has precluded a deeper understanding of "how" evolution has proceeded, as in the case of antimicrobial resistance. The difficulty in answering the "how" question lies in the multihierarchical dimensions of evolutionary processes, nested in complex networks, encompassing all units of selection, from genes to communities and ecosystems. At the simplest ontological level (as resistance genes), evolution proceeds by random (mutation and drift) and directional (natural selection) processes; however, sequential pathways of adaptive variation can occasionally be observed, and under fixed circumstances (particular fitness landscapes), evolution is predictable. At the highest level (such as that of plasmids, clones, species, microbiotas), the systems' degrees of freedom increase dramatically, related to the variable dispersal, fragmentation, relatedness, or coalescence of bacterial populations, depending on heterogeneous and changing niches and selective gradients in complex environments. Evolutionary trajectories of antibiotic resistance find their way in these changing landscapes subjected to random variations, becoming highly entropic and therefore unpredictable. However, experimental, phylogenetic, and ecogenetic analyses reveal preferential frequented paths (highways) where antibiotic resistance flows and propagates, allowing some understanding of evolutionary dynamics, modeling and designing interventions. Studies on antibiotic resistance have an applied aspect in improving individual health, One Health, and Global Health, as well as an academic value for understanding evolution. Most importantly, they have a heuristic significance as a model to reduce the negative influence of anthropogenic effects on the environment.

The E. coli Whole-Cell Modeling Project

The Escherichia coli whole-cell modeling project seeks to create the most detailed computational model of an E. coli cell in order to better understand and predict the behavior of this model organism. Details about the approach, framework, and current version of the model are discussed. Currently, the model includes the functions of 43% of characterized genes, with ongoing efforts to include additional data and mechanisms. As additional information is incorporated in the model, its utility and predictive power will continue to increase, which means that discovery efforts can be accelerated by community involvement in the generation and inclusion of data. This project will be an invaluable resource to the E. coli community that could be used to verify expected physiological behavior, to predict new outcomes and testable hypotheses for more efficient experimental design iterations, and to evaluate heterogeneous data sets in the context of each other through deep curation.

Hfq Regulates Efflux Pump Expression and Purine Metabolic Pathway to Increase Trimethoprim Resistance in Aeromonas veronii

Aeromonas veronii (A. veronii) is a zoonotic pathogen. It causes clinically a variety of diseases such as dysentery, bacteremia, and meningitis, and brings huge losses to aquaculture. A. veronii has been documented as a multiple antibiotic resistant bacterium. Hfq (host factor for RNA bacteriophage Qβ replication) participates in the regulations of the virulence, adhesion, and nitrogen fixation, effecting on the growth, metabolism synthesis and stress resistance in bacteria. The deletion of hfq gene in A. veronii showed more sensitivity to trimethoprim, accompanying by the upregulations of purine metabolic genes and downregulations of efflux pump genes by transcriptomic data analysis. Coherently, the complementation of efflux pump-related genes acrA and acrB recovered the trimethoprim resistance in Δhfq. Besides, the accumulations of adenosine and guanosine were increased in Δhfq in metabonomic data. The strain Δhfq conferred more sensitive to trimethoprim after appending 1 mM guanosine to M9 medium, while wild type was not altered. These results demonstrated that Hfq mediated trimethoprim resistance by elevating efflux pump expression and degrading adenosine, and guanosine metabolites. Collectively, Hfq is a potential target to tackle trimethoprim resistance in A. veronii infection.

A review: antimicrobial resistance data mining models and prediction methods study for pathogenic bacteria

Antimicrobials have paved the way for medical and social development over the last century and are indispensable for treating infections in humans and animals. The dramatic spread and diversity of antibiotic-resistant pathogens have significantly reduced the efficacy of essentially all antibiotic classes and is a global problem affecting human and animal health. Antimicrobial resistance is influenced by complex factors such as resistance genes and dosing, which are highly nonlinear, time-lagged and multivariate coupled, and the amount of resistance data is large and redundant, making it difficult to predict and analyze. Based on machine learning methods and data mining techniques, this paper reviews (1) antimicrobial resistance data storage and analysis techniques, (2) antimicrobial resistance assessment methods and the associated risk assessment methods for antimicrobial resistance, and (3) antimicrobial resistance prediction methods. Finally, the current research results on antimicrobial resistance and the development trend are summarized to provide a systematic and comprehensive reference for the research on antimicrobial resistance.

MECHANISTIC AND DATA-DRIVEN MODELS OF CELL SIGNALING: TOOLS FOR FUNDAMENTAL DISCOVERY AND RATIONAL DESIGN OF THERAPY

A full understanding of cell signaling processes requires knowledge of protein structure/function relationships, protein-protein interactions, and the abilities of pathways to control phenotypes. Computational models offer a valuable framework for integrating that knowledge to predict the effects of system perturbations and interventions in health and disease. Whereas mechanistic models are well suited for understanding the biophysical basis for signal transduction and principles of therapeutic design, data-driven models are particularly suited to distill complex signaling relationships among samples and between multivariate signaling changes and phenotypes. Both approaches have limitations and provide incomplete representations of signaling biology, but their careful implementation and integration can provide new understanding for how manipulating system variables impacts cellular decisions.

Artificial intelligence for assisting cancer diagnosis and treatment in the era of precision medicine

Over the past decade, artificial intelligence (AI) has contributed substantially to the resolution of various medical problems, including cancer. Deep learning (DL), a subfield of AI, is characterized by its ability to perform automated feature extraction and has great power in the assimilation and evaluation of large amounts of complicated data. On the basis of a large quantity of medical data and novel computational technologies, AI, especially DL, has been applied in various aspects of oncology research and has the potential to enhance cancer diagnosis and treatment. These applications range from early cancer detection, diagnosis, classification and grading, molecular characterization of tumors, prediction of patient outcomes and treatment responses, personalized treatment, automatic radiotherapy workflows, novel anti-cancer drug discovery, and clinical trials. In this review, we introduced the general principle of AI, summarized major areas of its application for cancer diagnosis and treatment, and discussed its future directions and remaining challenges. As the adoption of AI in clinical use is increasing, we anticipate the arrival of AI-powered cancer care.

Nutrient Scarcity in a New Defined Medium Reveals Metabolic Resistance to Antibiotics in the Fish Pathogen Piscirickettsia salmonis

Extensive use of antibiotics has been the primary treatment for the Salmonid Rickettsial Septicemia, a salmonid disease caused by the bacterium Piscirickettsia salmonis. Occurrence of antibiotic resistance has been explored in various P. salmonis isolates using different assays; however, P. salmonis is a nutritionally demanding intracellular facultative pathogen; thus, assessing its antibiotic susceptibility with standardized and validated protocols is essential. In this work, we studied the pathogen response to antibiotics using a genomic, a transcriptomic, and a phenotypic approach. A new defined medium (CMMAB) was developed based on a metabolic model of P. salmonis. CMMAB was formulated to increase bacterial growth in nutrient-limited conditions and to be suitable for performing antibiotic susceptibility tests. Antibiotic resistance was evaluated based on a comprehensive search of antibiotic resistance genes (ARGs) from P. salmonis genomes. Minimum inhibitory concentration assays were conducted to test the pathogen susceptibility to antibiotics from drug categories with predicted ARGs. In all tested P. salmonis strains, resistance to erythromycin, ampicillin, penicillin G, streptomycin, spectinomycin, polymyxin B, ceftazidime, and trimethoprim was medium-dependent, showing resistance to higher antibiotic concentrations in the CMMAB medium. The mechanism for antibiotic resistance to ampicillin in the defined medium was further explored and was proven to be associated to a decrease in the bacterial central metabolism, including the TCA cycle, the pentose-phosphate pathway, energy production, and nucleotide metabolism, and it was not associated with decreased growth rate of the bacterium or with the expression of any predicted ARG. Our results suggest that nutrient scarcity plays a role in the bacterial antibiotic resistance, protecting against the detrimental effects of antibiotics, and thus, we propose that P. salmonis exhibits a metabolic resistance to ampicillin when growing in a nutrient-limited medium.

Functions predict horizontal gene transfer and the emergence of antibiotic resistance

Phylogenetic distance, shared ecology, and genomic constraints are often cited as key drivers governing horizontal gene transfer (HGT), although their relative contributions are unclear. Here, we apply machine learning algorithms to a curated set of diverse bacterial genomes to tease apart the importance of specific functional traits on recent HGT events. We find that functional content accurately predicts the HGT network [area under the receiver operating characteristic curve (AUROC) = 0.983], and performance improves further (AUROC = 0.990) for transfers involving antibiotic resistance genes (ARGs), highlighting the importance of HGT machinery, niche-specific, and metabolic functions. We find that high-probability not-yet detected ARG transfer events are almost exclusive to human-associated bacteria. Our approach is robust at predicting the HGT networks of pathogens, including Acinetobacter baumannii and Escherichia coli, as well as within localized environments, such as an individual’s gut microbiome.

Insights into the mechanism of the interference of sulfadiazine on soil microbial community and function

The accumulation of sulfonamides in the soil environment possessed the potential to change soil microbial community and function. Metabolomics is capable of providing insights into the carbon metabolic pool and molecular mechanisms associated with external stressors. Here we evaluated alternations in soil bacterial community and soil metabolites profiles under sulfadiazine (SDZ) exposure and proposed a potential mechanism that SDZ accumulation in soil affected soil organic matter (SOM) cycling. Sequencing analysis showed that the relative abundance of bacterial species associated with carbon cycling significantly decreased under high concentrations of SDZ exposure. Untargeted metabolomics analysis showed that 78 metabolites were significantly changed with the presence of SDZ in soil. The combination of functional predictions and pathway analysis both demonstrated that high concentrations of SDZ exposure could cause disturbance in anabolism and catabolism. Moreover, the noticeable decline in the relative content of carbohydrates under high concentrations of SDZ exposure might weaken physical separation and provide more chances for microbes to degrade SOM. The above results provided evidence that SDZ accumulation in soil held the potential to disturb SOM cycling. These findings spread our understanding about the environmental risk of antibiotic in the soil environment beyond the dissemination of antibiotic resistance.

Biologically informed deep neural network for prostate cancer discovery

The determination of molecular features that mediate clinically aggressive phenotypes in prostate cancer remains a major biological and clinical challenge^1,2. Recent advances in interpretability of machine learning models as applied to biomedical problems may enable discovery and prediction in clinical cancer genomics^3-5. Here we developed P-NET-a biologically informed deep learning model-to stratify patients with prostate cancer by treatment-resistance state and evaluate molecular drivers of treatment resistance for therapeutic targeting through complete model interpretability. We demonstrate that P-NET can predict cancer state using molecular data with a performance that is superior to other modelling approaches. Moreover, the biological interpretability within P-NET revealed established and novel molecularly altered candidates, such as MDM4 and FGFR1, which were implicated in predicting advanced disease and validated in vitro. Broadly, biologically informed fully interpretable neural networks enable preclinical discovery and clinical prediction in prostate cancer and may have general applicability across cancer types.

Path to improving the life cycle and quality of genome-scale models of metabolism

Genome-scale models of metabolism (GEMs) are key computational tools for the systems-level study of metabolic networks. Here, we describe the "GEM life cycle," which we subdivide into four stages: inception, maturation, specialization, and amalgamation. We show how different types of GEM reconstruction workflows fit in each stage and proceed to highlight two fundamental bottlenecks for GEM quality improvement: GEM maturation and content removal. We identify common characteristics contributing to increasing quality of maturing GEMs drawing from past independent GEM maturation efforts. We then shed some much-needed light on the latent and unrecognized but pervasive issue of content removal, demonstrating the substantial effects of model pruning on its solution space. Finally, we propose a novel framework for content removal and associated confidence-level assignment which will help guide future GEM development efforts, reduce duplication of effort across groups, potentially aid automated reconstruction platforms, and boost the reproducibility of model development.

Antimicrobial susceptibility testing by measuring bacterial oxygen consumption on an integrated platform

Cellular respiration is a fundamental feature of metabolic activity and oxygen consumption can be considered as a reliable indicator of bacterial aerobic respiration, including for facultative anaerobic bacteria like E. coli. Addressing the emerging global health challenge of antimicrobial resistance, we performed antimicrobial susceptibility testing using the bacterial oxygen consumption rate (OCR) as a phenotypic indicator. We demonstrated that microbial exposure to antibiotics showed systematic OCR variations, which enabled determining minimum inhibitory concentrations for three clinically relevant antibiotics, ampicillin, ciprofloxacin, and gentamicin, within a few hours. Our study was performed by using photoluminescence-based oxygen sensing in a microchamber format, which enabled reducing the sample volume to a few hundred microliters. OCR modeling based on exponential bacterial growth allowed estimating the bacterial doubling time for various culture conditions (different types of media, different culture temperature and antibiotic concentrations). Furthermore, correlating metabolic heat production data, as obtained by nanocalorimetry in the same type of microchamber, and OCR measurements provided further insight on the actual metabolic state and activity of a microbial sample. This approach represents a new path towards more comprehensive microbiological studies performed on integrated miniaturized systems.

Accelerating antibiotic discovery through artificial intelligence

By targeting invasive organisms, antibiotics insert themselves into the ancient struggle of the host-pathogen evolutionary arms race. As pathogens evolve tactics for evading antibiotics, therapies decline in efficacy and must be replaced, distinguishing antibiotics from most other forms of drug development. Together with a slow and expensive antibiotic development pipeline, the proliferation of drug-resistant pathogens drives urgent interest in computational methods that promise to expedite candidate discovery. Strides in artificial intelligence (AI) have encouraged its application to multiple dimensions of computer-aided drug design, with increasing application to antibiotic discovery. This review describes AI-facilitated advances in the discovery of both small molecule antibiotics and antimicrobial peptides. Beyond the essential prediction of antimicrobial activity, emphasis is also given to antimicrobial compound representation, determination of drug-likeness traits, antimicrobial resistance, and de novo molecular design. Given the urgency of the antimicrobial resistance crisis, we analyze uptake of open science best practices in AI-driven antibiotic discovery and argue for openness and reproducibility as a means of accelerating preclinical research. Finally, trends in the literature and areas for future inquiry are discussed, as artificially intelligent enhancements to drug discovery at large offer many opportunities for future applications in antibiotic development.

AMiGA: Software for Automated Analysis of Microbial Growth Assays

The analysis of microbial growth is one of the central methods in the field of microbiology. Microbial growth dynamics can be characterized by meaningful parameters, including carrying capacity, exponential growth rate, and growth lag. However, microbial assays with clinical isolates, fastidious organisms, or microbes under stress often produce atypical growth shapes that do not follow the classical microbial growth pattern. Here, we introduce the analysis of microbial growth assays (AMiGA) software, which streamlines the analysis of growth curves without any assumptions about their shapes. AMiGA can pool replicates of growth curves and infer summary statistics for biologically meaningful growth parameters. In addition, AMiGA can quantify death phases and characterize diauxic shifts. It can also statistically test for differential growth under distinct experimental conditions. Altogether, AMiGA streamlines the organization, analysis, and visualization of microbial growth assays. IMPORTANCE Our current understanding of microbial physiology relies on the simple method of measuring microbial populations' sizes over time and under different conditions. Many advances have increased the throughput of those assays and enabled the study of nonlab-adapted microbes under diverse conditions that widely affect their growth dynamics. Our software provides an all-in-one tool for estimating the growth parameters of microbial cultures and testing for differential growth in a high-throughput and user-friendly fashion without any underlying assumptions about how microbes respond to their growth conditions.

Machine Learning of Bacterial Transcriptomes Reveals Responses Underlying Differential Antibiotic Susceptibility

In vitro antibiotic susceptibility testing often fails to accurately predict in vivo drug efficacies, in part due to differences in the molecular composition between standardized bacteriologic media and physiological environments within the body. Here, we investigate the interrelationship between antibiotic susceptibility and medium composition in Escherichia coli K-12 MG1655 as contextualized through machine learning of transcriptomics data. Application of independent component analysis, a signal separation algorithm, shows that complex phenotypic changes induced by environmental conditions or antibiotic treatment are directly traced to the action of a few key transcriptional regulators, including RpoS, Fur, and Fnr. Integrating machine learning results with biochemical knowledge of transcription factor activation reveals medium-dependent shifts in respiration and iron availability that drive differential antibiotic susceptibility. By extension, the data generation and data analytics workflow used here can interrogate the regulatory state of a pathogen under any measured condition and can be applied to any strain or organism for which sufficient transcriptomics data are available.

IMPORTANCE Antibiotic resistance is an imminent threat to global health. Patient treatment regimens are often selected based on results from standardized antibiotic susceptibility testing (AST) in the clinical microbiology lab, but these in vitro tests frequently misclassify drug effectiveness due to their poor resemblance to actual host conditions. Prior attempts to understand the combined effects of drugs and media on antibiotic efficacy have focused on physiological measurements but have not linked treatment outcomes to transcriptional responses on a systems level. Here, application of machine learning to transcriptomics data identified medium-dependent responses in key regulators of bacterial iron uptake and respiratory activity. The analytical workflow presented here is scalable to additional organisms and conditions and could be used to improve clinical AST by identifying the key regulatory factors dictating antibiotic susceptibility.

A cross-level information transmission network for hierarchical omics data integration and phenotype prediction from a new genotype

MOTIVATION

An unsolved fundamental problem in biology is to predict phenotypes from a new genotype under environmental perturbations. The emergence of multiple omics data provides new opportunities but imposes great challenges in the predictive modeling of genotype-phenotype associations. Firstly, the high-dimensionality of genomics data and the lack of coherent labeled data often make the existing supervised learning techniques less successful. Secondly, it is challenging to integrate heterogeneous omics data from different resources. Finally, few works have explicitly modeled the information transmission from DNA to phenotype, which involves multiple intermediate molecular types. Higher-level features (e.g. gene expression) usually have stronger discriminative and interpretable power than lower-level features (e.g. somatic mutation).

RESULTS

We propose a novel Cross-LEvel Information Transmission (CLEIT) network framework to address the above issues. CLEIT aims to represent the asymmetrical multi-level organization of the biological system by integrating multiple incoherent omics data and to improve the prediction power of low-level features. CLEIT first learns the latent representation of the high-level domain then uses it as ground-truth embedding to improve the representation learning of the low-level domain in the form of contrastive loss. Besides, CLEIT can leverage the unlabeled heterogeneous omics data to improve the generalizability of the predictive model. We demonstrate the effectiveness and significant performance boost of CLEIT in predicting anti-cancer drug sensitivity from somatic mutations via the assistance of gene expressions when compared with state-of-the-art methods. CLEIT provides a general framework to model information transmissions and integrate multi-modal data in a multi-level system.

AVAILABILITYAND IMPLEMENTATION

The source code is freely available at https://github.com/XieResearchGroup/CLEIT.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Genome-Scale Metabolic Models and Machine Learning Reveal Genetic Determinants of Antibiotic Resistance in Escherichia coli and Unravel the Underlying Metabolic Adaptation Mechanisms

Antimicrobial resistance (AMR) is becoming one of the largest threats to public health worldwide, with the opportunistic pathogen Escherichia coli playing a major role in the AMR global health crisis. Unravelling the complex interplay between drug resistance and metabolic rewiring is key to understand the ability of bacteria to adapt to new treatments and to the development of new effective solutions to combat resistant infections. We developed a computational pipeline that combines machine learning with genome-scale metabolic models (GSMs) to elucidate the systemic relationships between genetic determinants of resistance and metabolism beyond annotated drug resistance genes. Our approach was used to identify genetic determinants of 12 AMR profiles for the opportunistic pathogenic bacterium E. coli. Then, to interpret the large number of identified genetic determinants, we applied a constraint-based approach using the GSM to predict the effects of genetic changes on growth, metabolite yields, and reaction fluxes. Our computational platform leads to multiple results. First, our approach corroborates 225 known AMR-conferring genes, 35 of which are known for the specific antibiotic. Second, integration with the GSM predicted 20 top-ranked genetic determinants (including accA, metK, fabD, fabG, murG, lptG, mraY, folP, and glmM) essential for growth, while a further 17 top-ranked genetic determinants linked AMR to auxotrophic behavior. Third, clusters of AMR-conferring genes affecting similar metabolic processes are revealed, which strongly suggested that metabolic adaptations in cell wall, energy, iron and nucleotide metabolism are associated with AMR. The computational solution can be used to study other human and animal pathogens.

IMPORTANCEEscherichia coli is a major public health concern given its increasing level of antibiotic resistance worldwide and extraordinary capacity to acquire and spread resistance via horizontal gene transfer with surrounding species and via mutations in its existing genome. E. coli also exhibits a large amount of metabolic pathway redundancy, which promotes resistance via metabolic adaptability. In this study, we developed a computational approach that integrates machine learning with metabolic modeling to understand the correlation between AMR and metabolic adaptation mechanisms in this model bacterium. Using our approach, we identified AMR genetic determinants associated with cell wall modifications for increased permeability, virulence factor manipulation of host immunity, reduction of oxidative stress toxicity, and changes to energy metabolism. Unravelling the complex interplay between antibiotic resistance and metabolic rewiring may open new opportunities to understand the ability of E. coli, and potentially of other human and animal pathogens, to adapt to new treatments.

Plant Natural Flavonoids Against Multidrug Resistant Pathogens

The increasing emergence and dissemination of multidrug resistant (MDR) bacterial pathogens accelerate the desires for new antibiotics. Natural products dominate the preferred chemical scaffolds for the discovery of antibacterial agents. Here, the potential of natural flavonoids from plants against MDR bacteria, is demonstrated. Structure-activity relationship analysis shows the prenylation modulates the activity of flavonoids and obtains two compounds, α-mangostin (AMG) and isobavachalcone (IBC). AMG and IBC not only display rapid bactericidal activity against Gram-positive bacteria, but also restore the susceptibility of colistin against Gram-negative pathogens. Mechanistic studies generally show such compounds bind to the phospholipids of bacterial membrane, and result in the dissipation of proton motive force and metabolic perturbations, through distinctive modes of action. The efficacy of AMG and IBC in four models associated with infection or contamination, is demonstrated. These results suggest that natural products of plants may be a promising and underappreciated reservoir to circumvent the existing antibiotic resistance.

Towards a deeper understanding of microbial communities: integrating experimental data with dynamic models

Microbial communities and their functions are shaped by complex networks of interactions among microbes and with their environment. While the critical roles microbial communities play in numerous environments have become increasingly appreciated, we have a very limited understanding of their interactions and how these interactions combine to generate community-level behaviors. This knowledge gap hinders our ability to predict community responses to perturbations and to design interventions that manipulate these communities to our benefit. Dynamic models are promising tools to address these questions. We review existing modeling techniques to construct dynamic models of microbial communities at different scales and suggest ways to leverage multiple types of models and data to facilitate our understanding and engineering of microbial communities.

Synergy of alanine and gentamicin to reduce nitric oxide for elevating killing efficacy to antibiotic-resistant Vibrio alginolyticus

The present study explored the cooperative effect of both alanine (Ala) and gentamicin (Gent) on metabolic mechanisms by which exogenous Ala potentiates Gent to kill antibiotic-resistant Vibrio alginolyticus. To test this, GC-MS-based metabolomics was used to characterize Ala-, Gent- and both-induced metabolic profiles, identifying nitric oxide (NO) production pathway as the most key clue to understand metabolic mechanisms. Gent, Ala and both led to low, lower and lowest activity of total nitric oxide synthase (tNOS) and level of NO, respectively. NOS promoter L-arginine and inhibitor N^G-Monomethyl-L-arginine inhibited and promoted the killing, respectively, with the elevation and decrease of NOS activity and NO level. The present study further showed that CysJ is the enzyme-producing NO in V. alginolyticus. These results indicate that the cooperative effect of Ala and Gent causes the lowest NO, which plays a key role in Ala-potentiated Gent-mediated killing.

Explainable Artificial Intelligence Reveals Novel Insight into Tumor Microenvironment Conditions Linked with Better Prognosis in Patients with Breast Cancer

We investigated the data-driven relationship between immune cell composition in the tumor microenvironment (TME) and the ≥5-year survival rates of breast cancer patients using explainable artificial intelligence (XAI) models. We acquired TCGA breast invasive carcinoma data from the cbioPortal and retrieved immune cell composition estimates from bulk RNA sequencing data from TIMER2.0 based on EPIC, CIBERSORT, TIMER, and xCell computational methods. Novel insights derived from our XAI model showed that B cells, CD8+ T cells, M0 macrophages, and NK T cells are the most critical TME features for enhanced prognosis of breast cancer patients. Our XAI model also revealed the inflection points of these critical TME features, above or below which ≥5-year survival rates improve. Subsequently, we ascertained the conditional probabilities of ≥5-year survival under specific conditions inferred from the inflection points. In particular, the XAI models revealed that the B cell fraction (relative to all cells in a sample) exceeding 0.025, M0 macrophage fraction (relative to the total immune cell content) below 0.05, and NK T cell and CD8+ T cell fractions (based on cancer type-specific arbitrary units) above 0.075 and 0.25, respectively, in the TME could enhance the ≥5-year survival in breast cancer patients. The findings could lead to accurate clinical predictions and enhanced immunotherapies, and to the design of innovative strategies to reprogram the breast TME.

Metabolomics-Guided Elucidation of Plant Abiotic Stress Responses in the 4IR Era: An Overview

Plants are constantly challenged by changing environmental conditions that include abiotic stresses. These are limiting their development and productivity and are subsequently threatening our food security, especially when considering the pressure of the increasing global population. Thus, there is an urgent need for the next generation of crops with high productivity and resilience to climate change. The dawn of a new era characterized by the emergence of fourth industrial revolution (4IR) technologies has redefined the ideological boundaries of research and applications in plant sciences. Recent technological advances and machine learning (ML)-based computational tools and omics data analysis approaches are allowing scientists to derive comprehensive metabolic descriptions and models for the target plant species under specific conditions. Such accurate metabolic descriptions are imperatively essential for devising a roadmap for the next generation of crops that are resilient to environmental deterioration. By synthesizing the recent literature and collating data on metabolomics studies on plant responses to abiotic stresses, in the context of the 4IR era, we point out the opportunities and challenges offered by omics science, analytical intelligence, computational tools and big data analytics. Specifically, we highlight technological advancements in (plant) metabolomics workflows and the use of machine learning and computational tools to decipher the dynamics in the chemical space that define plant responses to abiotic stress conditions.

Challenges and opportunities in biological funneling of heterogeneous and toxic substrates beyond lignin

Significant developments in the understanding and manipulation of microbial metabolism have enabled the use of engineered biological systems toward a more sustainable energy and materials economy. While developments in metabolic engineering have primarily focused on the conversion of carbohydrates, substantial opportunities exist for using these same principles to extract value from more heterogeneous and toxic waste streams, such as those derived from lignin, biomass pyrolysis, or industrial waste. Funneling heterogeneous substrates from these streams toward valuable products, termed biological funneling, presents new challenges in balancing multiple catabolic pathways competing for shared cellular resources and engineering against perturbation from toxic substrates. Solutions to many of these challenges have been explored within the field of lignin valorization. This perspective aims to extend beyond lignin to highlight the challenges and discuss opportunities for use of biological systems to upgrade previously inaccessible waste streams.

Applications of Machine Learning to the Problem of Antimicrobial Resistance: an Emerging Model for Translational Research

Antimicrobial resistance (AMR) remains one of the most challenging phenomena of modern medicine. Machine learning (ML) is a subfield of artificial intelligence that focuses on the development of algorithms that learn how to accurately predict outcome variables using large sets of predictor variables that are typically not hand selected and are minimally curated. Models are parameterized using a training data set and then applied to a test data set on which predictive performance is evaluated. The application of ML algorithms to the problem of AMR has garnered increasing interest in the past 5 years due to the exponential growth of experimental and clinical data, heavy investment in computational capacity, improvements in algorithm performance, and increasing urgency for innovative approaches to reducing the burden of disease. Here, we review the current state of research at the intersection of ML and AMR with an emphasis on three domains of work. The first is the prediction of AMR using genomic data. The second is the use of ML to gain insight into the cellular functions disrupted by antibiotics, which forms the basis for understanding mechanisms of action and developing novel anti-infectives. The third focuses on the application of ML for antimicrobial stewardship using data extracted from the electronic health record. Although the use of ML for understanding, diagnosing, treating, and preventing AMR is still in its infancy, the continued growth of data and interest ensures it will become an important tool for future translational research programs.

A forecast for large-scale, predictive biology: Lessons from meteorology

Quantitative systems biology, in which predictive mathematical models are constructed to guide the design of experiments and predict experimental outcomes, is at an exciting transition point, where the foundational scientific principles are becoming established, but the impact is not yet global. The next steps necessary for mathematical modeling to transform biological research and applications, in the same way it has already transformed other fields, is not completely clear. The purpose of this perspective is to forecast possible answers to this question-what needs to happen next-by drawing on the experience gained in another field, specifically meteorology. We review here a number of lessons learned in weather prediction that are directly relevant to biological systems modeling, and that we believe can enable the same kinds of global impact in our field as atmospheric modeling makes today.

A Prediction Method for Animal-Derived Drug Resistance Trend Using a Grey-BP Neural Network Combination Model

There is an increasing drug resistance of animal-derived pathogens, seriously posing a huge threat to the health of animals and humans. Traditional drug resistance testing methods are expensive, have low efficiency, and are time-consuming, making it difficult to evaluate overall drug resistance. To develop a better approach to detect drug resistance, a small sample of Escherichia coli resistance data from 2003 to 2014 in Chengdu, Sichuan Province was used, and multiple regression interpolation was applied to impute missing data based on the time series. Next, cluster analysis was used to classify anti-E. coli drugs. According to the classification results, a GM(1,1)-BP model was selected to analyze the changes in the drug resistance of E. coli, and a drug resistance prediction system was constructed based on the GM(1,1)-BP Neural Network model. The GM(1,1)-BP Neural Network model showed a good prediction effect using a small sample of drug resistance data, with a determination coefficient R² of 0.7830 and an RMSE of only 0.0527. This model can be applied for the prediction of drug resistance trends of other animal-derived pathogenic bacteria, and provides the scientific and technical means for the effective assessment of bacterial resistance.

Growth and Stress Tolerance Comprise Independent Metabolic Strategies Critical for Staphylococcus aureus Infection

Staphylococcus aureus is an important pathogen that leads to high morbidity and mortality. Although S. aureus produces many factors important for pathogenesis, few have been validated as playing a role in the pathogenesis of S. aureus pneumonia. To gain a better understanding of the genetic elements required for S. aureus pathogenesis in the airway, we performed an unbiased genome-wide transposon sequencing (Tn-seq) screen in a model of acute murine pneumonia. We identified 136 genes important for bacterial survival during infection, with a high proportion involved in metabolic processes. Phenotyping 80 individual deletion mutants through diverse in vitro and in vivo assays demonstrated that metabolism is linked to several processes, which include biofilm formation, growth, and resistance to host stressors. We further validated the importance of 23 mutations in pneumonia. Multivariate and principal-component analyses identified two key metabolic mechanisms enabling infection in the airway, growth (e.g., the ability to replicate and form biofilms) and resistance to host stresses. As deep validation of these hypotheses, we investigated the role of pyruvate carboxylase, which was important across multiple infection models and confirmed a connection between growth and resistance to host cell killing. Pathogenesis is conventionally understood in terms of the host-pathogen interactions that enable a pathogen to neutralize a host’s immune response. We demonstrate with the important bacterial pathogen S. aureus that microbial metabolism influences key traits important for in vivo infection, independent from host immunomodulation.

Requirements for implementation of artificial intelligence in the practice of gastrointestinal pathology

Tremendous advances in artificial intelligence (AI) in medical image analysis have been achieved in recent years. The integration of AI is expected to cause a revolution in various areas of medicine, including gastrointestinal (GI) pathology. Currently, deep learning algorithms have shown promising benefits in areas of diagnostic histopathology, such as tumor identification, classification, prognosis prediction, and biomarker/genetic alteration prediction. While AI cannot substitute pathologists, carefully constructed AI applications may increase workforce productivity and diagnostic accuracy in pathology practice. Regardless of these promising advances, unlike the areas of radiology or cardiology imaging, no histopathology-based AI application has been approved by a regulatory authority or for public reimbursement. Thus, implying that there are still some obstacles to be overcome before AI applications can be safely and effectively implemented in real-life pathology practice. The challenges have been identified at different stages of the development process, such as needs identification, data curation, model development, validation, regulation, modification of daily workflow, and cost-effectiveness balance. The aim of this review is to present challenges in the process of AI development, validation, and regulation that should be overcome for its implementation in real-life GI pathology practice.

The Axes of Life: A roadmap for understanding dynamic multiscale systems

The biological challenges facing humanity are complex, multi-factorial, and are intimately tied to the future of our health, welfare, and stewardship of the Earth. Tackling problems in diverse areas, such as agriculture, ecology, and health care require linking vast data sets that encompass numerous components and spatio-temporal scales. Here, we provide a new framework and a road map for using experiments and computation to understand dynamic biological systems that span multiple scales. We discuss theories that can help understand complex biological systems and highlight the limitations of existing methodologies and recommend data generation practices. The advent of new technologies such as big data analytics and artificial intelligence can help bridge different scales and data types. We recommend ways to make such models transparent, compatible with existing theories of biological function, and to make biological data sets readable by advanced machine learning algorithms. Overall, the barriers for tackling pressing biological challenges are not only technological, but also sociological. Hence, we also provide recommendations for promoting interdisciplinary interactions between scientists.

Integration of machine learning and genome-scale metabolic modeling identifies multi-omics biomarkers for radiation resistance

Resistance to ionizing radiation, a first-line therapy for many cancers, is a major clinical challenge. Personalized prediction of tumor radiosensitivity is not currently implemented clinically due to insufficient accuracy of existing machine learning classifiers. Despite the acknowledged role of tumor metabolism in radiation response, metabolomics data is rarely collected in large multi-omics initiatives such as The Cancer Genome Atlas (TCGA) and consequently omitted from algorithm development. In this study, we circumvent the paucity of personalized metabolomics information by characterizing 915 TCGA patient tumors with genome-scale metabolic Flux Balance Analysis models generated from transcriptomic and genomic datasets. Metabolic biomarkers differentiating radiation-sensitive and -resistant tumors are predicted and experimentally validated, enabling integration of metabolic features with other multi-omics datasets into ensemble-based machine learning classifiers for radiation response. These multi-omics classifiers show improved classification accuracy, identify clinical patient subgroups, and demonstrate the utility of personalized blood-based metabolic biomarkers for radiation sensitivity. The integration of machine learning with genome-scale metabolic modeling represents a significant methodological advancement for identifying prognostic metabolite biomarkers and predicting radiosensitivity for individual patients.

Multimodal regularised linear models with flux balance analysis for mechanistic integration of omics data

MOTIVATION

High-throughput biological data, thanks to technological advances, have become cheaper to collect, leading to the availability of vast amounts of omic data of different types. In parallel, the in silico reconstruction and modelling of metabolic systems is now acknowledged as a key tool to complement experimental data on a large scale. The integration of these model- and data-driven information is therefore emerging as a new challenge in systems biology, with no clear guidance on how to better take advantage of the inherent multi-source and multi-omic nature of these data types while preserving mechanistic interpretation.

RESULTS

Here we investigate different regularisation techniques for high-dimensional data derived from the integration of gene expression profiles with metabolic flux data, extracted from strain-specific metabolic models, to improve cellular growth rate predictions. To this end, we propose ad-hoc extensions of previous regularisation frameworks including group, view-specific and principal component regularisation, and experimentally compare them using data from 1,143 Saccharomyces cerevisiae strains. We observe a divergence between methods in terms of regression accuracy and integration effectiveness based on the type of regularisation employed. In multi-omic regression tasks, when learning from experimental and model-generated omic data, our results demonstrate the competitiveness and ease of interpretation of multimodal regularised linear models compared to data-hungry methods based on neural networks.

AVAILABILITY

All data, models, and code produced in this work are available on GitHub at https://github.com/Angione-Lab/HybridGroupIPFLasso_pc2Lasso.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

C. elegans: A biosensor for host-microbe interactions

Microbes are an integral part of life on this planet. Microbes and their hosts influence each other in an endless dance that shapes how the meta-organism interacts with its environment. Although great advances have been made in microbiome research over the past 20 years, the mechanisms by which both hosts and their microbes interact with each other and the environment are still not well understood. The nematode Caenorhabditis elegans has been widely used as a model organism to study a remarkable number of human-like processes. Recent evidence shows that the worm is a powerful tool to investigate in fine detail the complexity that exists in microbe-host interactions. By combining the large array of genetic tools available for both organisms together with deep phenotyping approaches, it has been possible to uncover key effectors in the complex relationship between microbes and their hosts. In this perspective, we survey the literature for insightful discoveries in the microbiome field using the worm as a model. We discuss the latest conceptual and technological advances in the field and highlight the strengths that make C. elegans a valuable biosensor tool for the study of microbe-host interactions.

Deep Learning for Reintegrating Biology

The goal of this vision paper is to investigate the possible role that advanced machine learning techniques, especially deep learning, could play in the reintegration of various biological disciplines. To achieve this goal, a series of operational, but admittedly very simplistic, conceptualizations have been introduced: Life has been taken as a multidimensional phenomenon that inhabits three physical dimensions (time, space, and scale) and biological research as establishing connection between different points in the domain of life. Each of these points hence denotes a position in time, space, and scale at which a life phenomenon of interest takes place. Using these conceptualizations, fragmentation of biology can be seen as the result of too few and especially too short-ranged connections. Reintegrating biology could then be accomplished by establishing more, longer ranged connections. Deep learning methods appear to be very well suited for addressing this particular need at this particular time. Not withstanding the numerous unsubstantiated claims regarding the capabilities of AI, deep learning networks represent a major advance in the ability to find complex relationships inside large data sets that would have not been accessible with traditional data analytic methods or to a human observer. In addition, ongoing advances in the automation of taking measurements from phenomena on all levels of biological organization, continue to increase the number of large quantitative data sets that are available. These increasingly common data sets could serve as anchor points for making long-range connections by virtue of deep learning. However, connections within the domain of life are likely to be structured in a highly nonuniform fashion and hence it is necessary to develop methods, e.g., theoretical, computational, and experimental, to determine linkage of biological data sets most likely to provide useful insights on a biological problem using deep learning. Finally, specific deep learning approaches and architectures should be developed to match the needs of reintegrating biology.

The pursuit of mechanism of action: uncovering drug complexity in TB drug discovery

Whole cell-based phenotypic screens have become the primary mode of hit generation in tuberculosis (TB) drug discovery during the last two decades. Different drug screening models have been developed to mirror the complexity of TB disease in the laboratory. As these culture conditions are becoming more and more sophisticated, unraveling the drug target and the identification of the mechanism of action (MOA) of compounds of interest have additionally become more challenging. A good understanding of MOA is essential for the successful delivery of drug candidates for TB treatment due to the high level of complexity in the interactions between Mycobacterium tuberculosis (Mtb) and the TB drug used to treat the disease. There is no single "standard" protocol to follow and no single approach that is sufficient to fully investigate how a drug restrains Mtb. However, with the recent advancements in -omics technologies, there are multiple strategies that have been developed generally in the field of drug discovery that have been adapted to comprehensively characterize the MOAs of TB drugs in the laboratory. These approaches have led to the successful development of preclinical TB drug candidates, and to a better understanding of the pathogenesis of Mtb infection. In this review, we describe a plethora of efforts based upon genetic, metabolomic, biochemical, and computational approaches to investigate TB drug MOAs. We assess these different platforms for their strengths and limitations in TB drug MOA elucidation in the context of Mtb pathogenesis. With an emphasis on the essentiality of MOA identification, we outline the unmet needs in delivering TB drug candidates and provide direction for further TB drug discovery.

β-Lactams against the Fortress of the Gram-Positive Staphylococcus aureus Bacterium

The biological diversity of the unicellular bacteria-whether assessed by shape, food, metabolism, or ecological niche-surely rivals (if not exceeds) that of the multicellular eukaryotes. The relationship between bacteria whose ecological niche is the eukaryote, and the eukaryote, is often symbiosis or stasis. Some bacteria, however, seek advantage in this relationship. One of the most successful-to the disadvantage of the eukaryote-is the small (less than 1 μm diameter) and nearly spherical Staphylococcus aureus bacterium. For decades, successful clinical control of its infection has been accomplished using β-lactam antibiotics such as the penicillins and the cephalosporins. Over these same decades S. aureus has perfected resistance mechanisms against these antibiotics, which are then countered by new generations of β-lactam structure. This review addresses the current breadth of biochemical and microbiological efforts to preserve the future of the β-lactam antibiotics through a better understanding of how S. aureus protects the enzyme targets of the β-lactams, the penicillin-binding proteins. The penicillin-binding proteins are essential enzyme catalysts for the biosynthesis of the cell wall, and understanding how this cell wall is integrated into the protective cell envelope of the bacterium may identify new antibacterials and new adjuvants that preserve the efficacy of the β-lactams.

Metabolic fitness landscapes predict the evolution of antibiotic resistance

Bacteria evolve resistance to antibiotics by a multitude of mechanisms. A central, yet unsolved question is how resistance evolution affects cell growth at different drug levels. Here, we develop a fitness model that predicts growth rates of common resistance mutants from their effects on cell metabolism. The model maps metabolic effects of resistance mutations in drug-free environments and under drug challenge; the resulting fitness trade-off defines a Pareto surface of resistance evolution. We predict evolutionary trajectories of growth rates and resistance levels, which characterize Pareto resistance mutations emerging at different drug dosages. We also predict the prevalent resistance mechanism depending on drug and nutrient levels: low-dosage drug defence is mounted by regulation, evolution of distinct metabolic sectors sets in at successive threshold dosages. Evolutionary resistance mechanisms include membrane permeability changes and drug target mutations. These predictions are confirmed by empirical growth inhibition curves and genomic data of Escherichia coli populations. Our results show that resistance evolution, by coupling major metabolic pathways, is strongly intertwined with systems biology and ecology of microbial populations.

Digital Insights Into Nucleotide Metabolism and Antibiotic Treatment Failure

Nucleotide metabolism plays a central role in bacterial physiology, producing the nucleic acids necessary for DNA replication and RNA transcription. Recent studies demonstrate that nucleotide metabolism also proactively contributes to antibiotic-induced lethality in bacterial pathogens and that disruptions to nucleotide metabolism contributes to antibiotic treatment failure in the clinic. As antimicrobial resistance continues to grow unchecked, new approaches are needed to study the molecular mechanisms responsible for antibiotic efficacy. Here we review emerging technologies poised to transform understanding into why antibiotics may fail in the clinic. We discuss how these technologies led to the discovery that nucleotide metabolism regulates antibiotic drug responses and why these are relevant to human infections. We highlight opportunities for how studies into nucleotide metabolism may enhance understanding of antibiotic failure mechanisms.

Tuning of a Membrane-Perforating Antimicrobial Peptide to Selectively Target Membranes of Different Lipid Composition

The use of designed antimicrobial peptides as drugs has been impeded by the absence of simple sequence-structure-function relationships and design rules. The likely cause is that many of these peptides permeabilize membranes via highly disordered, heterogeneous mechanisms, forming aggregates without well-defined tertiary or secondary structure. We suggest that the combination of high-throughput library screening with atomistic computer simulations can successfully address this challenge by tuning a previously developed general pore-forming peptide into a selective pore-former for different lipid types. A library of 2916 peptides was designed based on the LDKA template. The library peptides were synthesized and screened using a high-throughput orthogonal vesicle leakage assay. Dyes of different sizes were entrapped inside vesicles with varying lipid composition to simultaneously screen for both pore size and affinity for negatively charged and neutral lipid membranes. From this screen, nine different LDKA variants that have unique activity were selected, sequenced, synthesized, and characterized. Despite the minor sequence changes, each of these peptides has unique functional properties, forming either small or large pores and being selective for either neutral or anionic lipid bilayers. Long-scale, unbiased atomistic molecular dynamics (MD) simulations directly reveal that rather than rigid, well-defined pores, these peptides can form a large repertoire of functional dynamic and heterogeneous aggregates, strongly affected by single mutations. Predicting the propensity to aggregate and assemble in a given environment from sequence alone holds the key to functional prediction of membrane permeabilization.

TAM mediates adaptation of carbapenem-resistant Klebsiella pneumoniae to antimicrobial stress during host colonization and infection

Gram-negative pathogens, such as Klebsiella pneumoniae, remodel their outer membrane (OM) in response to stress to maintain its integrity as an effective barrier and thus to promote their survival in the host. The emergence of carbapenem-resistant K. pneumoniae (CR-Kp) strains that are resistant to virtually all antibiotics is an increasing clinical problem and OM impermeability has limited development of antimicrobial agents because higher molecular weight antibiotics cannot access sites of activity. Here, we demonstrate that TAM (translocation and assembly module) deletion increases CR-Kp OM permeability under stress conditions and enhances sensitivity to high-molecular weight antimicrobials. SILAC-based proteomic analyses revealed mis-localization of membrane proteins in the TAM deficient strain. Stress-induced sensitization enhances clearance of TAM-deficient CR-Kp from the gut lumen following fecal microbiota transplantation and from infection sites following pulmonary or systemic infection. Our study suggests that TAM, as a regulator of OM permeability, represents a potential target for development of agents that enhance the effectiveness of existing antibiotics.

Reactive Oxygen Species in Pathogen Clearance: The Killing Mechanisms, the Adaption Response, and the Side Effects

Reactive oxygen species (ROS) are attractive weapons in both antibiotic-mediated killing and host-mediated killing. However, the involvement of ROS in antibiotic-mediated killing and complexities in host environments challenge the paradigm. In the case of bacterial pathogens, the examples of some certain pathogens thriving under ROS conditions prompt us to focus on the adaption mechanism that pathogens evolve to cope with ROS. Based on these, we here summarized the mechanisms of ROS-mediated killing of either antibiotics or the host, the examples of bacterial adaption that successful pathogens evolved to defend or thrive under ROS conditions, and the potential side effects of ROS in pathogen clearance. A brief section for new antibacterial strategies centered around ROS was also addressed.

A multifaceted cellular damage repair and prevention pathway promotes high-level tolerance to β-lactam antibiotics

Bactericidal antibiotics are powerful agents due to their ability to convert essential bacterial functions into lethal processes. However, many important bacterial pathogens are remarkably tolerant against bactericidal antibiotics due to inducible damage repair responses. The cell wall damage response two-component system VxrAB of the gastrointestinal pathogen Vibrio cholerae promotes high-level β-lactam tolerance and controls a gene network encoding highly diverse functions, including negative control over multiple iron uptake systems. How this system contributes to tolerance is poorly understood. Here, we show that β-lactam antibiotics cause an increase in intracellular free iron levels and collateral oxidative damage, which is exacerbated in the ∆vxrAB mutant. Mutating major iron uptake systems dramatically increases ∆vxrAB tolerance to β-lactams. We propose that VxrAB reduces antibiotic-induced toxic iron and concomitant metabolic perturbations by downregulating iron uptake transporters and show that iron sequestration enhances tolerance against β-lactam therapy in a mouse model of cholera infection. Our results suggest that a microorganism's ability to counteract diverse antibiotic-induced stresses promotes high-level antibiotic tolerance and highlights the complex secondary responses elicited by antibiotics.

Thymine Sensitizes Gram-Negative Pathogens to Antibiotic Killing

Diminished antibiotic susceptibility of bacterial pathogens is an increasingly serious threat to human and animal health. Alternative strategies are required to combat antibiotic refractory bacteria. Bacterial metabolic state has been shown to play a critical role in its susceptibility to antibiotic killing. However, the adjuvant potential of nucleotides in combination with antibiotics to kill Gram-negative pathogens remains unknown. Herein, we found that thymine potentiated ciprofloxacin killing against both sensitive and resistant-E. coli in a growth phase-independent manner. Similar promotion effects were also observed for other bactericidal antibiotics, including ampicillin and kanamycin, in the fight against four kinds of Gram-negative bacteria. The mechanisms underlying this finding were that exogenous thymine could upregulate bacterial metabolism including increased TCA cycle and respiration, which thereby promote the production of ATP and ROS. Subsequently, metabolically inactive bacteria were converted to active bacteria and restored its susceptibility to antibiotic killing. In Galleria mellonella infection model, thymine effectively improved ciprofloxacin activity against E. coli. Taken together, our results demonstrated that thymine potentiates bactericidal antibiotics activity against Gram-negative pathogens through activating bacterial metabolism, providing a universal strategy to overcome Gram-negative pathogens.

Identifying the effect of vancomycin on health care-associated methicillin-resistant Staphylococcus aureus strains using bacteriological and physiological media

BACKGROUND

The evolving antibiotic-resistant behavior of health care-associated methicillin-resistant Staphylococcus aureus (HA-MRSA) USA100 strains are of major concern. They are resistant to a broad class of antibiotics such as macrolides, aminoglycosides, fluoroquinolones, and many more.

FINDINGS

The selection of appropriate antibiotic susceptibility examination media is very important. Thus, we use bacteriological (cation-adjusted Mueller-Hinton broth) as well as physiological (R10LB) media to determine the effect of vancomycin on USA100 strains. The study includes the profiling behavior of HA-MRSA USA100 D592 and D712 strains in the presence of vancomycin through various high-throughput assays. The US100 D592 and D712 strains were characterized at sub-inhibitory concentrations through growth curves, RNA sequencing, bacterial cytological profiling, and exo-metabolomics high throughput experiments.

CONCLUSIONS

The study reveals the vancomycin resistance behavior of HA-MRSA USA100 strains in dual media conditions using wide-ranging experiments.

How Do Machines Learn? Artificial Intelligence as a New Era in Medicine

With an increased number of medical data generated every day, there is a strong need for reliable, automated evaluation tools. With high hopes and expectations, machine learning has the potential to revolutionize many fields of medicine, helping to make faster and more correct decisions and improving current standards of treatment. Today, machines can analyze, learn, communicate, and understand processed data and are used in health care increasingly. This review explains different models and the general process of machine learning and training the algorithms. Furthermore, it summarizes the most useful machine learning applications and tools in different branches of medicine and health care (radiology, pathology, pharmacology, infectious diseases, personalized decision making, and many others). The review also addresses the futuristic prospects and threats of applying artificial intelligence as an advanced, automated medicine tool.

Classification of the mechanism of toxicity as applied to human cell line ECV304

The objective of this study was to identify the pattern of cytotoxicity testing of the human cell line ECV304 using three techniques of an ensemble learning algorithm (bagging, boosting and stacking). The study of cell morphology of ECV304 cell line was conducted using impedimetric measurement. Three types of toxins were applied to the ECV304 cell line namely 1 mM hydrogen peroxide (H₂O₂), 5% dimethyl sulfoxide and 10 μg Saponin. The measurement was conducted using electrodes and lock-in amplifier to detect impedance changes during cytotoxicity testing within a frequency range 200 and 830 kHz. The results were analysed, processed and extracted using detrended fluctuation analysis to obtain characteristics and features of the cells when exposed to the each of the toxins. Three ensemble algorithms applied showed slightly different results on the performance for classifying the data set from the feature extraction that was performed. However, the results show that the cell reaction to the toxins could be classified.

Application of Artificial Intelligence Technology in Oncology: Towards the Establishment of Precision Medicine

In recent years, advances in artificial intelligence (AI) technology have led to the rapid clinical implementation of devices with AI technology in the medical field. More than 60 AI-equipped medical devices have already been approved by the Food and Drug Administration (FDA) in the United States, and the active introduction of AI technology is considered to be an inevitable trend in the future of medicine. In the field of oncology, clinical applications of medical devices using AI technology are already underway, mainly in radiology, and AI technology is expected to be positioned as an important core technology. In particular, "precision medicine," a medical treatment that selects the most appropriate treatment for each patient based on a vast amount of medical data such as genome information, has become a worldwide trend; AI technology is expected to be utilized in the process of extracting truly useful information from a large amount of medical data and applying it to diagnosis and treatment. In this review, we would like to introduce the history of AI technology and the current state of medical AI, especially in the oncology field, as well as discuss the possibilities and challenges of AI technology in the medical field.

Deep Learning Analysis of Vibrational Spectra of Bacterial Lysate for Rapid Antimicrobial Susceptibility Testing

Rapid antimicrobial susceptibility testing (AST) is an integral tool to mitigate the unnecessary use of powerful and broad-spectrum antibiotics that leads to the proliferation of multi-drug-resistant bacteria. Using a sensor platform composed of surface-enhanced Raman scattering (SERS) sensors with control of nanogap chemistry and machine learning algorithms for analysis of complex spectral data, bacteria metabolic profiles post antibiotic exposure are correlated with susceptibility. Deep neural network models are able to discriminate the responses of Escherichia coli and Pseudomonas aeruginosa to antibiotics from untreated cells in SERS data in 10 min after antibiotic exposure with greater than 99% accuracy. Deep learning analysis is also able to differentiate responses from untreated cells with antibiotic dosages up to 10-fold lower than the minimum inhibitory concentration observed in conventional growth assays. In addition, analysis of SERS data using a generative model, a variational autoencoder, identifies spectral features in the P. aeruginosa lysate data associated with antibiotic efficacy. From this insight, a combinatorial dataset of metabolites is selected to extend the latent space of the variational autoencoder. This culture-free dataset dramatically improves classification accuracy to select effective antibiotic treatment in 30 min. Unsupervised Bayesian Gaussian mixture analysis achieves 99.3% accuracy in discriminating between susceptible versus resistant to antibiotic cultures in SERS using the extended latent space. Discriminative and generative models rapidly provide high classification accuracy with small sets of labeled data, which enormously reduces the amount of time needed to validate phenotypic AST with conventional growth assays. Thus, this work outlines a promising approach toward practical rapid AST.