Cullen Buie

Multidimensional scaling (MDS) is a widely used dimensionality reduction technique in microbial ecology data analysis that captures the multivariate structure of the data while preserving pairwise distances between samples. While improvements in MDS have enhanced the ability to reveal group-specific data patterns, these MDS-based methods require prior assumptions for inference, limiting their application in general microbiome analysis. In this study, we introduce a new MDS-based ordination method, "F-informed MDS," which configures the data distribution based on the F-statistic, the ratio of dispersion between groups sharing common and different characteristics. Using semisynthetic datasets, we demonstrate that the proposed method is robust to hyperparameter selection while maintaining statistical significance throughout the ordination process. Various quality metrics for evaluating dimensionality reduction confirm that F-informed MDS is comparable to state-of-the-art methods in preserving both local and global data structures. Its application to a diatom-associated bacterial community suggests the role of this new method in interpreting the community's response to the host. Our approach offers a well-founded refinement of MDS that aligns with statistical test results, which can be beneficial for broader multidimensional data analyses in microbiology and ecology. This new visualization tool can be incorporated into standard microbiome data analyses.

Accurate colony counting is crucial for assessing microbial growth in high-throughput workflows. However, existing automated counting solutions struggle with the issue of merged colonies, a common occurrence in high-throughput plating. To overcome this limitation, we propose MCount, the only known solution that incorporates both contour information and regional algorithms for colony counting. By optimizing the pairing of contours with regional candidate circles, MCount can accurately infer the number of merged colonies. We evaluate MCount on a precisely labeled Escherichia coli dataset of 960 images (15,847 segments) and achieve an average error rate of 3.99%, significantly outperforming existing published solutions such as NICE (16.54%), AutoCellSeg (33.54%), and OpenCFU (50.31%). MCount is user-friendly as it only requires two hyperparameters. To further facilitate deployment in scenarios with limited labeled data, we propose statistical methods for selecting the hyperparameters using few labeled or even unlabeled data points, all of which guarantee consistently low error rates. MCount presents a promising solution for accurate and efficient colony counting in application workflows requiring high throughput, particularly in cases with merged colonies.

Bifidobacteria represent a dominant constituent of human gut microbiomes during infancy, influencing nutrition, immune development, and resistance to infection. Despite interest in bifidobacteria as a live biotic therapy, our understanding of colonization, host-microbe interactions, and the health-promoting effects of bifidobacteria is limited. To address these major knowledge gaps, we used a large-scale genetic approach to create a mutant fitness compendium in Bifidobacterium breve. First, we generated a high-density randomly barcoded transposon insertion pool and used it to determine fitness requirements during colonization of germ-free mice and chickens with multiple diets and in response to hundreds of in vitro perturbations. Second, to enable mechanistic investigation, we constructed an ordered collection of insertion strains covering 1,462 genes. We leveraged these tools to reveal community- and diet-specific requirements for colonization and to connect the production of immunomodulatory molecules to growth benefits. These resources will catalyze future investigations of this important beneficial microbe.

Phytoplankton account for nearly half of global photosynthetic carbon fixation, and the fate of that carbon is regulated in large part by microbial food web processing. We currently lack a mechanistic understanding of how interactions among heterotrophic bacteria impact the fate of photosynthetically fixed carbon. Here, we used a set of bacterial isolates capable of growing on exudates from the diatom Phaeodactylum tricornutum to investigate how bacteria-bacteria interactions affect the balance between exudate remineralization and incorporation into biomass. With exometabolomics and genome-scale metabolic modeling, we estimated the degree of resource competition between bacterial pairs. In a sequential spent media experiment, we found that pairwise interactions were more beneficial than predicted based on resource competition alone, and 30% exhibited facilitative interactions. To link this to carbon fate, we used single-cell isotope tracing in a custom cultivation system to compare the impact of different "primary" bacterial strains in close proximity to live P. tricornutum on a distal "secondary" strain. We found that a primary strain with a high degree of competition decreased secondary strain carbon drawdown by 51% at the single-cell level, providing a quantitative metric for the "cost" of competition on algal carbon fate. Additionally, a primary strain classified as facilitative based on sequential interactions increased total algal-derived carbon assimilation by 7.6 times, integrated over all members, compared to the competitive primary strain. Our findings suggest that the degree of interaction between bacteria along a spectrum from competitive to facilitative is directly linked to algal carbon drawdown.

Bacterial cell genetic engineering is fundamental for research aiming to learn more about bacterial species for a broad range of applications. One method of intracellular delivery of foreign DNA during the genetic engineering process is the use of electroporation to create pores along the bacterial cell membrane. Current methods for assessing pore formation do not directly measure cell permeabilization or enable same-day assessment. In this thesis, a novel fast-screening protocol combining SYTOX green, microfluidics, and fluorescence imaging is evaluated for its capability to assess multiple conditions for cell permeabilization within a single day. By imaging bulk suspensions of post-electroporated cells stained with intracellularly delivered SYTOX, multiple electroporation conditions can be rapidly screened for cell permeabilization. This fast-screening protocol utilizes standard microbiology equipment and low-cost microfluidic imaging chambers, lowering the barrier to adoption and significantly reducing experimental time compared to conventional protocols involving foreign DNA delivery. Importantly, by decoupling permeabilization assessment from foreign DNA uptake, this method isolates the effect of membrane permeabilization from confounding factors such as restriction-modification systems. As a result, it provides a more accurate qualitative and quantitative assessment of bacterial membrane disruption. This approach enables same-day evaluation of electroporation conditions regardless of bacterial growth rate, potentially accelerating the optimization process for intracellular delivery in gene editing applications.

The field of biosensing is undergoing remarkable advancements, spurred by the pressing demand for innovative solutions across various domains, including disease diagnosis, health monitoring, environmental surveillance, and biomarker discovery. This progress is particularly evident in biosensor applications, where technologies are evolving rapidly to meet the diverse and complex needs of healthcare, environmental science, and beyond.

Abstract Microbial communities regulate the transformations of carbon in aquatic systems through metabolic interactions and food-web dynamics that can alter the balance of photosynthesis and respiration. Direct competition for resources is thought to drive microbial community assembly in algal systems, but other interaction modes that may shape communities are more challenging to isolate. Through untargeted metabolomics and metabolic modeling, we predicted the degree of resource competition between bacterial pairs when growing on model diatom Phaeodactylum tricornutum- derived substrates. In a subsequent sequential media experiment, we found that pairwise interactions were consistently more cooperative than predicted based on resource competition alone, indicating an unexpected role for cooperation in algal carbon processing. To link this directly to algal carbon fate, we chose a representative cooperative and competitive ‘influencer’ isolate and a model ‘recipient’ and applied single-cell isotope tracing in a custom porous microplate cultivation system. In the presence of live algae, the recipient drew down more algal carbon in the presence of the cooperative influencer compared to the competitive influencer, supporting the sequential experiment results. We also found that total carbon assimilation into bacterial biomass, integrated over influencer and recipient, was significantly higher for the cooperative interaction. Our findings support the notion that non-competitive interactions are critical for predicting algal carbon fate. Significance Statement Microbial interactions have widely been studied in the context of host resources but testing and measuring direct interactions in a lab has been particularly challenging. By combining untargeted metabolomics, sequential/(co-)culture, and metabolic modeling, we demonstrate that the presence of an unexpected interaction mode in a live system and show how it impacts the flow of host-derived resources. This top-down approach can help identify novel bacterial interactions that play a crucial role in microbial community-host ecosystems, which may have an impact in holobiont phenotypes including alga, fungal, or plant systems.

Randomly barcoded transposon mutant libraries are powerful tools for studying gene function and organization, assessing gene essentiality and pathways, discovering potential therapeutic targets, and understanding the physiology of gut bacteria and their interactions with the host. However, construction of high-quality libraries with uniform representation can be challenging. In this review, we survey various strategies for barcoded library construction, including transposition systems, methods of transposon delivery, optimal library size, and transconjugant selection schemes. We discuss the advantages and limitations of each approach, as well as factors to consider when selecting a strategy. In addition, we highlight experimental and computational advances in arraying condensed libraries from mutant pools. We focus on examples of successful library construction in gut bacteria and their application to gene function studies and drug discovery. Given the need for understanding gene function and organization in gut bacteria, we provide a comprehensive guide for researchers to construct randomly barcoded transposon mutant libraries.

The critical role of the intestinal microbiota in human health and disease is well recognized. Nevertheless, there are still large gaps in our understanding of the functions and mechanisms encoded in the genomes of most members of the gut microbiota. Genome-scale libraries of transposon mutants are a powerful tool to help us address this gap. Recent advances in barcoded transposon mutagenesis have dramatically lowered the cost of mutant fitness determination in hundreds of in vitro and in vivo experimental conditions. In an accompanying review, we discuss recent advances and caveats for the construction of pooled and arrayed barcoded transposon mutant libraries in human gut commensals. In this review, we discuss how these libraries can be used across a wide range of applications, the technical aspects involved, and expectations for such screens.

MG1655 strain. Measured zeta potentials for these samples were in agreement with literature values obtained by conventional measurement methods. Taken together, our data demonstrate the power of this workflow to broadly enable critical measurements of particle and bacterial zeta potential for numerous applications.

Abstract Bifidobacteria commonly represent a dominant constituent of human gut microbiomes during infancy, influencing nutrition, immune development, and resistance to infection. Despite interest as a probiotic therapy, predicting the nutritional requirements and health-promoting effects of Bifidobacteria is challenging due to major knowledge gaps. To overcome these deficiencies, we used large-scale genetics to create a compendium of mutant fitness in Bifidobacterium breve ( Bb ). We generated a high density, randomly barcoded transposon insertion pool in Bb , and used this pool to determine Bb fitness requirements during colonization of germ-free mice and chickens with multiple diets and in response to hundreds of in vitro perturbations. To enable mechanistic investigation, we constructed an ordered collection of insertion strains covering 1462 genes. We leveraged these tools to improve models of metabolic pathways, reveal unexpected host- and diet-specific requirements for colonization, and connect the production of immunomodulatory molecules to growth benefits. These resources will greatly reduce the barrier to future investigations of this important beneficial microbe.

Bacteria zeta potential, ζ, is one of the key parameters for the deposition of bacteria on an electrode, also called electrophoretic deposition (EPD) and depends on the strain, suspension composition and pH. Measuring directly with a zetasizer is challenging due to the non-spherical shape of bacteria and the need for a lower conductivity medium which could affect the value of ζ. Electrokinetic measurements are usually performed with microchannels in PDMS [1] or in PMMA [2], both of which require expensive tools for the fabrication of either the mold or the microchannel itself. Reducing the cost and time needed to perform such measurements would enable the broader use of this technique. We anticipate solving this challenge with commercially available microfluidic chips as will be outlined in this study. Under an electric field E in a microchannel, the electrokinetic velocity of particles in a fluid is linked to two phenomena: electrophoresis (EP) due to the zeta potential of the particle and electroosmosis (EO) due to the zeta potential of the channel wall. PMMA microchannels (Microfluidic ChipShop) with 200 µl reservoirs together with a voltage sequencer (Labsmith) controlled with LabVIEW were used to characterize the velocity of 1 µm and 2 µm polystyrene beads with different surface functionalizations (nonfunctionalized polystryrene (PS), Sulfate-modified (SU-modified), Carboxylate-modified (CO-modified) and Amine-modified (AM-modified) from Invitrogen and Magsphere suspended in 10 mM HEPES (pH 7) by applying different voltages between the inlet and outlet platinum electrodes for 90 s. Recorded time-lapse fluorescent image sequences were analyzed with ImageJ and TrackMate. Electrokinetic velocities obtained by Particle Tracking Velocimetry (PTV) at different voltages were fitted with a linear fit with a high coefficient of determination allowing the extraction of the different particle zeta potentials, which are in good agreement with the control measurements performed with a Zetasizer Ultra (Malvern Panalytical) [3]. Further, this methodology was used to characterize a prevalent strain of E. coli, yielding values in agreement with the literature. The use of commercially available microfluidic chips in electrokinetic characterization could alleviate the need for microfabrication and allow the wider adoption of this characterization method. These measurements are ultimately critical to better understand the role of the zeta potential in EPD of microbes for an array of applications. [1] S. Antunez-Vela, et al., Anal Chem , 2020 [2] Q. Wang et al., Science Advances , 2019 [3] J. Cottet, et al . In preparation

Multidimensional scaling (MDS) is a dimensionality reduction technique for microbial ecology data analysis that represents the multivariate structure while preserving pairwise distances between samples. While its improvement has enhanced the ability to reveal data patterns by sample groups, these MDS-based methods require prior assumptions for inference, limiting their application in general microbiome analysis. In this study, we introduce a new MDS-based ordination, $F$-informed MDS, which configures the data distribution based on the $F$-statistic, the ratio of dispersion between groups sharing common and different characteristics. Using simulated compositional datasets, we demonstrate that the proposed method is robust to hyperparameter selection while maintaining statistical significance throughout the ordination process. Various quality metrics for evaluating dimensionality reduction confirm that $F$-informed MDS is comparable to state-of-the-art methods in preserving both local and global data structures. Its application to a diatom-associated bacterial community suggests the role of this new method in interpreting the community response to the host. Our approach offers a well-founded refinement of MDS that aligns with statistical test results, which can be beneficial for broader compositional data analyses in microbiology and ecology. This new visualization tool can be incorporated into standard microbiome data analyses.

Over the past decade, the increased adoption of electroporation-based technologies has led to an expansion of clinical research initiatives. Electroporation has been utilized in molecular biology for mammalian and bacterial transfection; for food sanitation; and in therapeutic settings to increase drug uptake, for gene therapy, and to eliminate cancerous tissues. We begin this article by discussing the biophysics required for understanding the concepts behind the cell permeation phenomenon that is electroporation. We then review nano- and microscale single-cell electroporation technologies before scaling up to emerging in vivo applications.

Lipopolysaccharide (LPS) is the unique feature that composes the outer leaflet of the Gram-negative bacterial cell envelope. Variations in LPS structures affect a number of physiological processes, including outer membrane permeability, antimicrobial resistance, recognition by the host immune system, biofilm formation, and interbacterial competition. Rapid characterization of LPS properties is crucial for studying the relationship between these LPS structural changes and bacterial physiology. However, current assessments of LPS structures require LPS extraction and purification followed by cumbersome proteomic analysis. This paper demonstrates one of the first high-throughput and non-invasive strategies to directly distinguish Escherichia coli with different LPS structures. Using a combination of three-dimensional insulator-based dielectrophoresis (3DiDEP) and cell tracking in a linear electrokinetics assay, we elucidate the effect of structural changes in E. coli LPS oligosaccharides on electrokinetic mobility and polarizability. We show that our platform is sufficiently sensitive to detect LPS structural variations at the molecular level. To correlate electrokinetic properties of LPS with the outer membrane permeability, we further examined effects of LPS structural variations on bacterial susceptibility to colistin, an antibiotic known to disrupt the outer membrane by targeting LPS. Our results suggest that microfluidic electrokinetic platforms employing 3DiDEP can be a useful tool for isolating and selecting bacteria based on their LPS glycoforms. Future iterations of these platforms could be leveraged for rapid profiling of pathogens based on their surface LPS structural identity.