Barbara A. Bailey

Noise and Nonlinearity in Measles Epidemics: Combining Mechanistic and Statistical Approaches to Population Modeling

We present and evaluate an approach to analyzing population dynamics data using semimechanistic models. These models incorporate reliable information on population structure and underlying dynamic mechanisms but use nonparametric surface‐fitting methods to avoid unsupported assumptions about the precise form of rate equations. Using historical data on measles epidemics as a case study, we show how this approach can lead to better forecasts, better characterizations of the dynamics, and a better understanding of the factors causing complex population dynamics relative to either mechanistic models or purely descriptive statistical time‐series models. The semimechanistic models are found to have better forecasting accuracy than either of the model types used in previous analyses when tested on data not used to fit the models. The dynamics are characterized as being both nonlinear and noisy, and the global dynamics are clustered very tightly near the border of stability (dominant Lyapunov exponent λ ≈ 0). However, locally in state space the dynamics oscillate between strong short‐term stability and strong short‐term chaos (i.e., between negative and positive local Lyapunov exponents). There is statistically significant evidence for short‐term chaos in all data sets examined. Thus the nonlinearity in these systems is characterized by the variance over state space in local measures of chaos versus stability rather than a single summary measure of the overall dynamics as either chaotic or nonchaotic.

Lytic to temperate switching of viral communities

Microbial viruses can control host abundances via density-dependent lytic predator–prey dynamics. Less clear is how temperate viruses, which coexist and replicate with their host, influence microbial communities. Here we show that virus-like particles are relatively less abundant at high host densities. This suggests suppressed lysis where established models predict lytic dynamics are favoured. Meta-analysis of published viral and microbial densities showed that this trend was widespread in diverse ecosystems ranging from soil to freshwater to human lungs. Experimental manipulations showed viral densities more consistent with temperate than lytic life cycles at increasing microbial abundance. An analysis of 24 coral reef viromes showed a relative increase in the abundance of hallmark genes encoded by temperate viruses with increased microbial abundance. Based on these four lines of evidence, we propose the Piggyback-the-Winner model wherein temperate dynamics become increasingly important in ecosystems with high microbial densities; thus ‘more microbes, fewer viruses’.

Clinical Insights from Metagenomic Analysis of Sputum Samples from Patients with Cystic Fibrosis

ABSTRACT As DNA sequencing becomes faster and cheaper, genomics-based approaches are being explored for their use in personalized diagnoses and treatments. Here, we provide a proof of principle for disease monitoring using personal metagenomic sequencing and traditional clinical microbiology by focusing on three adults with cystic fibrosis (CF). The CF lung is a dynamic environment that hosts a complex ecosystem composed of bacteria, viruses, and fungi that can vary in space and time. Not surprisingly, the microbiome data from the induced sputum samples we collected revealed a significant amount of species diversity not seen in routine clinical laboratory cultures. The relative abundances of several species changed as clinical treatment was altered, enabling the identification of the climax and attack communities that were proposed in an earlier work. All patient microbiomes encoded a diversity of mechanisms to resist antibiotics, consistent with the characteristics of multidrug-resistant microbial communities that are commonly observed in CF patients. The metabolic potentials of these communities differed by the health status and recovery route of each patient. Thus, this pilot study provides an example of how metagenomic data might be used with clinical assessments for the development of treatments tailored to individual patients.

PHACTS, a computational approach to classifying the lifestyle of phages

Motivation: Bacteriophages have two distinct lifestyles: virulent and temperate. The virulent lifestyle has many implications for phage therapy, genomics and microbiology. Determining which lifestyle a newly sequenced phage falls into is currently determined using standard culturing techniques. Such laboratory work is not only costly and time consuming, but also cannot be used on phage genomes constructed from environmental sequencing. Therefore, a computational method that utilizes the sequence data of phage genomes is needed. Results: Phage Classification Tool Set (PHACTS) utilizes a novel similarity algorithm and a supervised Random Forest classifier to make a prediction whether the lifestyle of a phage, described by its proteome, is virulent or temperate. The similarity algorithm creates a training set from phages with known lifestyles and along with the lifestyle annotation, trains a Random Forest to classify the lifestyle of a phage. PHACTS predictions are shown to have a 99% precision rate. Availability and implementation: PHACTS was implemented in the PERL programming language and utilizes the FASTA program (Pearson and Lipman, 1988) and the R programming language library ‘Random Forest’ (Liaw and Weiner, 2010). The PHACTS software is open source and is available as downloadable stand-alone version or can be accessed online as a user-friendly web interface. The source code, help files and online version are available at http://www.phantome.org/PHACTS/. Contact: katelyn@rohan.sdsu.edu; redwards@sciences.sdsu.edu Supplementary information: Supplementary data are available at Bioinformatics online.

The Upper Respiratory Tract as a Microbial Source for Pulmonary Infections in Cystic Fibrosis. Parallels from Island Biogeography

A continuously mixed series of microbial communities inhabits various points of the respiratory tract, with community composition determined by distance from colonization sources, colonization rates, and extinction rates. Ecology and evolution theory developed in the context of biogeography is relevant to clinical microbiology and could reframe the interpretation of recent studies comparing communities from lung explant samples, sputum samples, and oropharyngeal swabs. We propose an island biogeography model of the microbial communities inhabiting different niches in human airways. Island biogeography as applied to communities separated by time and space is a useful parallel for exploring microbial colonization of healthy and diseased lungs, with the potential to inform our understanding of microbial community dynamics and the relevance of microbes detected in different sample types. In this perspective, we focus on the intermixed microbial communities inhabiting different regions of the airways of patients with cystic fibrosis.

Diagnostic classification of intrinsic functional connectivity highlights somatosensory, default mode, and visual regions in autism

Despite consensus on the neurological nature of autism spectrum disorders (ASD), brain biomarkers remain unknown and diagnosis continues to be based on behavioral criteria. Growing evidence suggests that brain abnormalities in ASD occur at the level of interconnected networks; however, previous attempts using functional connectivity data for diagnostic classification have reached only moderate accuracy. We selected 252 low-motion resting-state functional MRI (rs-fMRI) scans from the Autism Brain Imaging Data Exchange (ABIDE) including typically developing (TD) and ASD participants (n = 126 each), matched for age, non-verbal IQ, and head motion. A matrix of functional connectivities between 220 functionally defined regions of interest was used for diagnostic classification, implementing several machine learning tools. While support vector machines in combination with particle swarm optimization and recursive feature elimination performed modestly (with accuracies for validation datasets <70%), diagnostic classification reached a high accuracy of 91% with random forest (RF), a nonparametric ensemble learning method. Among the 100 most informative features (connectivities), for which this peak accuracy was achieved, participation of somatosensory, default mode, visual, and subcortical regions stood out. Whereas some of these findings were expected, given previous findings of default mode abnormalities and atypical visual functioning in ASD, the prominent role of somatosensory regions was remarkable. The finding of peak accuracy for 100 interregional functional connectivities further suggests that brain biomarkers of ASD may be regionally complex and distributed, rather than localized.

Relationships Between Safety and Both Congestion and Number of Lanes on Urban Freeways

This paper first explores the relationship between safety and congestion and then examines the relationship between safety and the number of lanes on urban freeways. The relationship between safety and congestion on urban freeways was explored with the use of safety performance functions (SPFs) calibrated for multilane freeways in Colorado, California, and Texas. The focus of most SPF modeling efforts to date has been on the statistical technique and the underlying probability distribution, with only limited consideration given to the nature of the phenomenon itself. In this study neural networks have been used to identify the underlying relationship between safety and exposure. The modeling process was informed by the consideration of the traffic operations parameters described by the Highway Capacity Manual. The shape of the SPF is best described by a sigmoid reflecting a dose-response type of relationship between safety and traffic demand on urban freeways. Relating safety to the degree of congestion suggests that safety deteriorates with the degradation in the quality of service expressed through the level of service. Practitioners generally believe that additional capacity afforded by additional lanes is associated with more safety. How much safety and for what time period are generally not considered. Comparison of SPFs of multilane freeways suggests that adding lanes may initially result in a temporary safety improvement that disappears as congestion increases. As annual average daily traffic increases, the slope of SPF, described by its first derivative, becomes steeper, reflecting that accidents are increasing at a faster rate than would be expected from a freeway with fewer lanes.

Assessing Coral Reefs on a Pacific-Wide Scale Using the Microbialization Score

The majority of the worlds coral reefs are in various stages of decline. While a suite of disturbances (overfishing, eutrophication, and global climate change) have been identified, the mechanism(s) of reef system decline remain elusive. Increased microbial and viral loading with higher percentages of opportunistic and specific microbial pathogens have been identified as potentially unifying features of coral reefs in decline. Due to their relative size and high per cell activity, a small change in microbial biomass may signal a large reallocation of available energy in an ecosystem; that is the microbialization of the coral reef. Our hypothesis was that human activities alter the energy budget of the reef system, specifically by altering the allocation of metabolic energy between microbes and macrobes. To determine if this is occurring on a regional scale, we calculated the basal metabolic rates for the fish and microbial communities at 99 sites on twenty-nine coral islands throughout the Pacific Ocean using previously established scaling relationships. From these metabolic rate predictions, we derived a new metric for assessing and comparing reef health called the microbialization score. The microbialization score represents the percentage of the combined fish and microbial predicted metabolic rate that is microbial. Our results demonstrate a strong positive correlation between reef microbialization scores and human impact. In contrast, microbialization scores did not significantly correlate with ocean net primary production, local chla concentrations, or the combined metabolic rate of the fish and microbial communities. These findings support the hypothesis that human activities are shifting energy to the microbes, at the expense of the macrobes. Regardless of oceanographic context, the microbialization score is a powerful metric for assessing the level of human impact a reef system is experiencing.

Microbial, host and xenobiotic diversity in the cystic fibrosis sputum metabolome.

Cystic fibrosis (CF) lungs are filled with thick mucus that obstructs airways and facilitates chronic infections. Pseudomonas aeruginosa is a significant pathogen of this disease that produces a variety of toxic small molecules. We used molecular networking-based metabolomics to investigate the chemistry of CF sputa and assess how the microbial molecules detected reflect the microbiome and clinical culture history of the patients. Metabolites detected included xenobiotics, P. aeruginosa specialized metabolites and host sphingolipids. The clinical culture and microbiome profiles did not correspond to the detection of P. aeruginosa metabolites in the same samples. The P. aeruginosa molecules that were detected in sputum did not match those from laboratory cultures. The pseudomonas quinolone signal (PQS) was readily detectable from cultured strains, but absent from sputum, even when its precursor molecules were present. The lack of PQS production in vivo is potentially due to the chemical nature of the CF lung environment, indicating that culture-based studies of this pathogen may not explain its behavior in the lung. The most differentially abundant molecules between CF and non-CF sputum were sphingolipids, including sphingomyelins, ceramides and lactosylceramide. As these highly abundant molecules contain the inflammatory mediator ceramide, they may have a significant role in CF hyperinflammation. This study demonstrates that the chemical makeup of CF sputum is a complex milieu of microbial, host and xenobiotic molecules. Detection of a bacterium by clinical culturing and 16S rRNA gene profiling do not necessarily reflect the active production of metabolites from that bacterium in a sputum sample.

Subdiffusive motion of bacteriophage in mucosal surfaces increases the frequency of bacterial encounters

Significance Bacteriophages (phages) are viruses that infect and kill bacteria. Being inanimate, phages rely on diffusion to search for bacterial prey. Here we demonstrate that a phage that adheres weakly to mucus exhibits subdiffusive motion, not normal diffusion, in mucosal surfaces. Supporting theory and experiments revealed that subdiffusive motion increases bacterial encounter rates for phages when bacterial concentration is low. To the best of our knowledge, no other predator has been shown to effectively use a subdiffusive search mechanism. Mucosal surfaces are vulnerable to infection. Mucus-adherent phages reduce bacterial infection of lifelike mucosal surfaces more effectively than nonadherent phages. These findings provide a basis for engineering adherent phages to manipulate mucosal surface microbiomes for protection from infection and other purposes. Bacteriophages (phages) defend mucosal surfaces against bacterial infections. However, their complex interactions with their bacterial hosts and with the mucus-covered epithelium remain mostly unexplored. Our previous work demonstrated that T4 phage with Hoc proteins exposed on their capsid adhered to mucin glycoproteins and protected mucus-producing tissue culture cells in vitro. On this basis, we proposed our bacteriophage adherence to mucus (BAM) model of immunity. Here, to test this model, we developed a microfluidic device (chip) that emulates a mucosal surface experiencing constant fluid flow and mucin secretion dynamics. Using mucus-producing human cells and Escherichia coli in the chip, we observed similar accumulation and persistence of mucus-adherent T4 phage and nonadherent T4∆hoc phage in the mucus. Nevertheless, T4 phage reduced bacterial colonization of the epithelium >4,000-fold compared with T4∆hoc phage. This suggests that phage adherence to mucus increases encounters with bacterial hosts by some other mechanism. Phages are traditionally thought to be completely dependent on normal diffusion, driven by random Brownian motion, for host contact. We demonstrated that T4 phage particles displayed subdiffusive motion in mucus, whereas T4∆hoc particles displayed normal diffusion. Experiments and modeling indicate that subdiffusive motion increases phage–host encounters when bacterial concentration is low. By concentrating phages in an optimal mucus zone, subdiffusion increases their host encounters and antimicrobial action. Our revised BAM model proposes that the fundamental mechanism of mucosal immunity is subdiffusion resulting from adherence to mucus. These findings suggest intriguing possibilities for engineering phages to manipulate and personalize the mucosal microbiome.