Biology subjects

Song, B.

Publications and source records attributed to Song, B..

At least 73 records · Page 4Linked to original sources

bioRxiv2021

Structures of the Omicron Spike trimer with ACE2 and an anti-Omicron antibody

The Omicron variant of SARS-CoV-2 has rapidly become the dominant infective strain and the focus efforts against the ongoing COVID-19 pandemic. Here we report an extensive set of structures of the Omicron spike trimer by its own or in complex with ACE2 and an anti-Omicron antibody. These structures reveal that most Omicron mutations are located on the surface of the spike protein, which confer stronger ACE2 binding by nearly 10 folds but become inactive epitopes resistant to many therapeutic antibodies. Importantly, both RBD and the closed conformation of the Omicron spike trimer are thermodynamically unstable, with the melting temperature of the Omicron RBD decreased by as much as 7{degrees}C, making the spiker trimer prone to random open conformations. An unusual RBD-RBD interaction in the ACE2-spike complex unique to Omicron is observed to support the open conformation and ACE2 binding, serving the basis for the higher infectivity of Omicron. A broad-spectrum therapeutic antibody JMB2002, which has completed Phase 1 clinical trial, is found to interact with the same two RBDs to inhibit ACE2 binding, in a mode that is distinguished from all previous antibodies, thus providing the structural basis for the potent inhibition of Omicron by this antibody. Together with biochemical data, our structures provide crucial insights into higher infectivity, antibody evasion and inhibition of Omicron.

biophysics↗

bioRxiv2021

A novel cryo-electron microscopy support film based on 2D crystal of HFBI protein

Cryo-electron microscopy (cryo-EM) has become the most powerful tool to resolve the high-resolution structures of biomacromolecules in solution. However, the air-water interface induced preferred orientation, dissociation or denaturation of biomacromolecules during cryo-vitrification is still a major limitation factor for many specimens. To solve this bottleneck, we developed a new type of cryo-EM support film using the 2D crystal of hydrophobin I (HFBI) protein. The HFBI-film utilizes its hydrophilic side to adsorb protein particles via electrostatic interactions and keep air-water interface away, allowing thin-enough ice and high-quality data collection. The particle orientation distribution can be optimized by changing the buffer pH. We, for the first time, solved the cryo-EM structure of catalase (2.28-[A]) that exhibited strong preferred orientation using conventional cryo-vitrification protocol. We further proved the HFBI-film is suitable to solve the high-resolution structures of small proteins including aldolase (150 kDa, 3.34-[A]) and hemoglobin (64 kDa, 3.6-[A]). Our work implied that the HFBI-film will have a wide application in the future to increase the successful rate and efficiency of cryo-EM.

biophysics↗

bioRxiv2021

De-noising Spatial Expression Profiling Data Based on in situ Position and Image Information

Spatial transcriptomics (ST) technologies provide gene expression close to or even superior to single-cell resolution while retaining the physical locations of sequencing and often also providing matched pathology images. However, the expression data captured by ST technologies suffer from high noise levels, as a result of the shallow coverage in each sequencing unit. The extra experimental steps for preserving the spatial locations of sequencing could result in even more severe noises, compared to regular single-cell RNA-sequencing (scRNA-seq). Fortunately, such noises could be largely removed by leveraging information from the physical locations of sequencing, and the tissue and cellular organization reflected by corresponding pathology images. In this work, we demonstrated the extensive levels of noise in ST data. We developed a mathematical model, named Sprod, to impute accurate ST gene expression based on latent space and graph learning of matched location and imaging data. We comprehensively validated Sprod and demonstrated its advantages over prior methods for removing drop-outs in scRNA-seq data. We further showed that, after adequate imputation by Sprod, differential expression analyses, pseudotime analyses, and cell-to-cell interaction inferences yield significantly more informative results. Overall, we envision denoising by Sprod to become a key first step to empower ST technologies for biomedical discoveries and innovations.

bioinformatics↗

bioRxiv2021

The Practical Haplotype Graph, a platform for storing and using pangenomes for imputation

MotivationPangenomes provide novel insights for population and quantitative genetics, genomics, and breeding not available from studying a single reference genome. Instead, a species is better represented by a pangenome or collection of genomes. Unfortunately, managing and using pangenomes for genomically diverse species is computationally and practically challenging. We developed a trellis graph representation anchored to the reference genome that represents most pangenomes well and can be used to impute complete genomes from low density sequence or variant data. ResultsThe Practical Haplotype Graph (PHG) is a pangenome pipeline, database (PostGRES & SQLite), data model (Java, Kotlin, or R), and Breeding API (BrAPI) web service. The PHG has already been able to accurately represent diversity in four major crops including maize, one of the most genomically diverse species, with up to 1000-fold data compression. Using simulated data, we show that, at even 0.1X coverage, with appropriate reads and sequence alignment, imputation results in extremely accurate haplotype reconstruction. The PHG is a platform and environment for the understanding and application of genomic diversity. AvailabilityAll resources listed here are freely available. The PHG Docker used to generate the simulation results is https://hub.docker.com/ as maizegenetics/phg:0.0.27. PHG source code is at https://bitbucket.org/bucklerlab/practicalhaplotypegraph/src/master/. The code used for the analysis of simulated data is at https://bitbucket.org/bucklerlab/phg-manuscript/src/master/. The PHG database of NAM parent haplotypes is in the CyVerse data store (https://de.cyverse.org/de/) and named /iplant/home/shared/panzea/panGenome/PHG_db_maize/phg_v5Assemblies_20200608.db. Contactpjb39@cornell.edu

bioinformatics↗

bioRxiv2021

A deep learning-based iterative digital pathology annotation tool

Well-annotated exemplars are an important prerequisite for supervised deep learning schemes. Unfortunately, generating these annotations is a cumbersome and laborious process, due to the large amount of time and effort needed. Here we present a deep-learning-based iterative digital pathology annotation tool that is both easy to use by pathologists and easy to integrate into machine vision systems. Our pathology image annotation tool greatly reduces annotation time from hours to a few minutes, while maintaining high fidelity with human-expert manual annotations. Here we demonstrate that our active learning tool can be used for a variety of pathology annotation tasks including masking tumor, stroma, and lymphocyte-rich regions, among others. This annotation automation system was validated on 90 unseen digital pathology images with tumor content from the CAMELYON16 database and it was found that pathologists gold standard masks were re-produced successfully using our tool. That is, an average of 2.7 positive selections (mouse clicks) and 8.0 negative selections (mouse clicks) were sufficient to generate tumor masks similar to pathologists gold standard in CAMELYON16 test WSIs. Furthermore, the developed image annotation tool has been used to build gold standard masks for hundreds of TCGA digital pathology images. This set was used to train a convolutional neural network for identification of tumor epithelium. The developed pan-cancer deep neural network was then tested on TCGA and internal data with comparable performance. The validated pathology image annotation tool described herein has the potential to be of great value in facilitating accurate, rapid pathological analysis of tumor biopsies.

pathology↗

bioRxiv2021

Multispecies phase diagram reveals biophysical principles of bacterial biofilm architectures

Bacterial biofilms are among the most abundant multicellular structures on Earth and play essential roles in a wide range of ecological, medical, and industrial processes. However, general principles that govern the emergence of biofilm architecture across different species remain unknown. Here, we combine experiments, simulations and statistical analysis to identify shared biophysical mechanisms that determine biofilm architecture development at the single-cell level, for the species Vibrio cholerae, Escherichia coli, Salmonella enterica, and Pseudomonas aeruginosa. Our data-driven analysis reveals that despite the many molecular differences between these species, the biofilm architecture differences can be described by only two control parameters: cellular aspect ratio and cell density. Further experiments using single-species mutants for which the cell aspect ratio and the cell density are systematically varied, and mechanistic simulations, show that tuning these two control parameters reproduces biofilm architectures of different species. Altogether, our results show that early-stage biofilm architecture is determined by mechanical cell-cell interactions which are conserved across different species and, therefore, provide a unifying understanding of biofilm architecture development.

microbiology↗

bioRxiv2021

AnchorWave: sensitive alignment of genomes with high diversity, structural polymorphism and whole-genome duplication variation

Millions of species are currently being sequenced and their genomes are being compared. Many of them have more complex genomes than model systems and raised novel challenges for genome alignment. Widely used local alignment strategies often produce limited or incongruous results when applied to genomes with dispersed repeats, long indels, and highly diverse sequences. Moreover, alignment using many-to-many or reciprocal best hit approaches conflicts with well-studied patterns between species with different rounds of whole-genome duplication or polyploidy levels. Here we introduce AnchorWave, which performs whole-genome duplication informed collinear anchor identification between genomes and performs base-pair resolution global alignments for collinear blocks using the wavefront algorithm and a 2-piece affine gap cost strategy. This strategy enables AnchorWave to precisely identify multi-kilobase indels generated by transposable element (TE) presence/absence variants (PAVs). When aligning two maize genomes, AnchorWave successfully recalled 87% of previously reported TE PAVs between two maize lines. By contrast, other genome alignment tools showed almost zero power for TE PAV recall. AnchorWave precisely aligns up to three times more of the genome than the closest competitive approach, when comparing diverse genomes. Moreover, AnchorWave recalls transcription factor binding sites (TFBSs) at a rate of 1.05-74.85 fold higher than other tools, while with significantly lower false positive alignments. AnchorWave shows obvious improvement when applied to genomes with dispersed repeats, active transposable elements, high sequence diversity and whole-genome duplication variation. Significance statementOne fundamental analysis needed to interpret genome assemblies is genome alignment. Yet, accurately aligning regulatory and transposon regions outside of genes remains challenging. We introduce AnchorWave, which implements a genome duplication informed longest path algorithm to identify collinear regions and performs base-pair resolved, end-to-end alignment for collinear blocks using an efficient 2-piece affine gap cost strategy. AnchorWave improves alignment of partially synthetic and real genomes under a number of scenarios: genomes with high similarity, large genomes with high TE activity, genomes with many inversions, and alignments between species with deeper evolutionary divergence and different whole-genome duplication histories. Potential use cases for the method include genome comparison for evolutionary analysis of non-genic sequences and population genetics of taxa with complex genomes.

genomics↗

bioRxiv2021

A multiple genome alignment workflow shows the impact of repeat masking and parameter tuning on alignment of functional regions in plants

Alignments of multiple genomes are a cornerstone of comparative genomics, but generating these alignments remains technically challenging and often impractical. We developed the msa_pipeline workflow (https://bitbucket.org/bucklerlab/msa_pipeline) based on the LAST aligner to allow practical and sensitive multiple alignment of diverged plant genomes with minimal user inputs. Our workflow only requires a set of genomes in FASTA format as input. The workflow outputs multiple alignments in MAF format, and includes utilities to help calculate genome-wide conservation scores. As high repeat content and genomic divergence are substantial challenges in plant genome alignment, we also explored the impact of different masking approaches and alignment parameters using genome assemblies of 33 grass species. Compared to conventional masking with RepeatMasker, a k-mer masking approach increased the alignment rate of CDS and non-coding functional regions by 25% and 14% respectively. We further found that default alignment parameters generally perform well, but parameter tuning can increase the alignment rate for non-coding functional regions by over 52% compared to default LAST settings. Finally, by increasing alignment sensitivity from the default baseline, parameter tuning can increase the number of non-coding sites that can be scored for conservation by over 76%.

genomics↗

bioRxiv2021

A novel method for generating 3D constructs with branched vascular networks using multi-materials bioprinting and direct surgical anastomosis

Vessels pervade almost all body tissues, and significantly influence the pathophysiology of human body. Previous attempts to establish multi-scale vascular connection and function in 3D model tissues using bioprinting have had limited success due to the incoordination between cell-laden materials and stability of the perfusion channel. Here, we report a methodology to fabricate centimetre-scale vascularized soft tissue with high viability and accuracy using multi-materials bioprinting involving inks with low viscosity and a customized multistage-temperature-control printer. The tissue formed was perfused with branched vasculature with well-formed 3D capillary network and lumen, which would potentially supply the cellular components with sufficient nutrients in the matrix. Furthermore, the same methodology was applied for generating liver-like tissue with the objective to fabricate and mimic a mature and functional liver tissue, with increased functionality in terms of synthesis of liver specific proteins after in vitro perfusion and in vivo subperitoneal transplantation in mice. Moreover, to establish immediate blood perfusion, an elastic layer was printed wrapping sacrificial ink to support the direct surgical anastomosis of the carotid artery to the jugular vein. Our findings highlight the support extended by vasculature network in soft hydrogels which helps to sustain the thick and dense cellularization in engineered tissues.

bioengineering↗

bioRxiv2020

Gene expression evolution in pattern-triggered immunity within Arabidopsis thaliana and across Brassicaceae species

Plants recognize surrounding microbes by sensing microbe-associated molecular patterns (MAMPs) to activate pattern-triggered immunity (PTI). Despite their significance for microbial control, the evolution of PTI responses remains largely uncharacterized. Employing comparative transcriptomics of six Arabidopsis thaliana accessions and three additional Brassicaceae species for PTI responses to the MAMP flg22, we identified a set of genes with expression changes under purifying selection in the Brassicaceae species and genes exhibiting species-specific expression signatures. Variation in flg22-triggered transcriptome and metabolome responses across Brassicaceae species was incongruent with their phylogeny while expression changes were strongly conserved within A. thaliana, suggesting directional selection for some species-specific gene expression. We found the enrichment of WRKY transcription factor binding sites in 5-regulatory regions in conserved and species-specific responsive genes, linking the emergence of WRKY-binding sites with the evolution of gene responses in PTI. Our findings advance our understanding of transcriptome evolution during biotic stress.

plant biology↗

bioRxiv2020

Physiological characterization of nitrate ammonifying bacteria isolated from rice paddy soils via a newly developed high-throughput screening method

Dissimilatory nitrate/nitrite reduction to ammonium (DNRA) has recently gained attention as a nitrogen retention pathway that may potentially be harnessed to alleviate nitrogen loss resulting from denitrification. Until recently, ecophysiology of DNRA bacteria inhabiting agricultural soils has remained largely unexplored, due to the difficulty in targeted enrichment and isolation of DNRA microorganisms. In this study, >100 microbial isolates capable of DNRA have been isolated from rice paddy soil with apparent dominance of denitrification using a novel high-throughput screening method. Six of these isolates, each assigned to a disparate genus, was examined to improve understanding of DNRA physiology. All isolates carried nrfA and/or nirB, and an isolate affiliated to Bacillus possessed a clade II nosZ gene and was capable of N2O reduction. A common prominent physiological feature observed in all DNRA isolates was NO2- accumulation observed before NH4+ production, which was further examined with Citrobacter sp. DNRA3 (possessing nrfA and nirB) and Enterobacter sp. DNRA5 (possessing only nirB). In both organisms, NO2--to-NH4+ reduction was inhibited by submillimolar NO3-, and nrfA or nirB transcription was down-regulated when NO3- was being reduced to NO2-. Both batch and chemostat incubations of these isolates with excess organic electron donors produced NH4+ from reduction of NO3-; however, incubation with excess NO3- resulted in NO2- buildup but no substantial NH4+ production, presumably due to NO3- presence. This previously overlooked link between NO3- repression of NO2--to-NH4+ reduction and the C-to-N ratio regulation of DNRA activity may be a key mechanism underpinning denitrification-vs-DNRA competition in soil. IMPORTANCEDissimilatory nitrate/nitrite reduction to ammonium (DNRA) is an anaerobic microbial pathway that competes with denitrification for common substrates NO3- and NO2-. Unlike denitrification leading to nitrogen loss and N2O emission, DNRA reduces NO3- and NO2- to NH4+, a reactive nitrogen with higher tendency to be retained in soil matrix. Therefore, stimulation of DNRA has often been proposed as a strategy to improve fertilizer efficiency and reduce greenhouse gas emissions. Such attempts have been hampered by lack of insights into soil DNRA ecophysiology. Here, we have developed a novel high-throughput screening method for isolating DNRA-catalyzing organisms from agricultural soils without apparent DNRA activity. Physiological characteristics of six DNRA isolates were closely examined, disclosing a previously overlooked link between NO3- repression of NO2--to-NH4+ reduction and the C-to-N ratio regulation of DNRA activity, which may be key to understanding why significant DNRA activity is rarely observed in nitrogen-rich agricultural soils.

microbiology↗

bioRxiv2020

Glucose Production and Glucose Disposal Integrated by an Intrinsic Inverse Coupling in Vivo: hypothesis of inter-organ reflex on glucose regulation

Glucose production (GP) and glucose disposal (Rd) are two decisive and fundamental parameters in glucose turnover and in glucose homeostasis regulation. In conventional theory, GP and Rd were responsive to regulatory factors respectively and independently of each other. Even though GP and Rd responded in reverse to insulin, GP for suppression and Rd for elevation, these inverse alterations used to be attributed to insulin multiple functions both on hepatic GP, directly or indirectly, and on whole-body glucose Rd. However, in the present study, we found GP and Rd were inversely coupled intrinsically no matter which side was the target of insulin by comparison of Rd and GP data pairs between peripheral vein insulin infusion protocol and portal vein insulin infusion protocol in rats. Furthermore, neither circulating NEFA nor HFD induced resistance broke the GP-Rd inverse coupling, but both of them reduced the responses of both GP and Rd to insulin. In conclusion, we provide the evidence that GP and Rd are two coupled parameters in vivo and they alter in reverse simultaneously, the mechanism under which needs further investigation but we tend to believe an inter-organ neural reflex was involved.

physiology↗

bioRxiv2019

Draft Genomes of two Artocarpus plants, Jackfruit (A. heterophyllus) and Breadfruit (A. altilis)

Two of the most economically important plants in the Artocarpus genus are jackfruit (A. heterophyllus Lam.) and breadfruit (A. altilis (Parkinson) Fosberg). Both species are long-lived trees that have been cultivated for thousands of years in their native regions. Today they are grown throughout tropical to subtropical areas as an important source of starch and other valuable nutrients. There are hundreds of breadfruit varieties that are native to Oceania, of which the most commonly distributed types are seedless triploids. Jackfruit is likely native to the western Ghats of India and produces one of the largest tree-borne fruit structures (reaching up to 100 pounds). To date, there is limited genomic information for these two economically important species. Here, we generated 273 Gb and 227 Gb of raw data from jackfruit and breadfruit, respectively. The high-quality reads from jackfruit were assembled into 162,440 scaffolds totaling 982 Mb with 35,858 genes. Similarly, the breadfruit reads were assembled into 180,971 scaffolds totaling 833 Mb with 34,010 genes. A total of 2,822 and 2,034 expanded gene families were found in jackfruit and breadfruit, respectively, enriched in pathways including starch- and sucrose metabolism, photosynthesis and others. The copy number of several starch synthesis related genes were found increased in jackfruit and breadfruit compared to closely related species, and the tissue specific expression might imply their sugar-rich and starch-rich characteristics. Overall, the publication of high-quality genomes for jackfruit and breadfruit provides information about their specific composition and the underlying genes involved in sugar and starch metabolism.

genomics↗

bioRxiv2019

Na2CO3-Responsive Photosynthetic and ROS Scavenging Mechanisms in Chloroplasts of Alkaligrass Revealed by Phosphoproteomics

Alkali-salinity exerts severe osmotic, ionic and high-pH stresses to plants. To understand the alkali-salinity responsive mechanisms underlying photosynthetic modulation and reactive oxygen species (ROS) homeostasis, physiological and diverse quantitative proteomics analyses of alkaligrass (Puccinellia tenuiflora) under Na2CO3 stress were conducted. In addition, Western blot, real-time PCR, and transgenic techniques were applied to validate the proteomic results and test the functions of the Na2CO3-responsive proteins. A total of 104 and 102 Na2CO3-responsive proteins were identified in leaves and chloroplasts, respectively. In addition, 84 Na2CO3-responsive phosphoproteins were identified, including 56 new phosphorylation sites in 56 phosphoproteins from chloroplasts, which are crucial for the regulation of photosynthesis, ion transport, signal transduction and energy homeostasis. A full-length PtFBA encoding an alkaligrass chloroplastic fructose-bisphosphate aldolase (FBA) was overexpressed in wild-type cells of cyanobacterium Synechocystis sp. Strain PCC 6803, leading to enhanced Na2CO3 tolerance. All these results indicate that thermal dissipation, state transition, cyclic electron transport, photorespiration, repair of photosystem (PS) II, PSI activity, and ROS homeostasis were altered in response to Na2CO3 stress, and they have improved our understanding of the Na2CO3-responsive mechanisms in halophytes.

plant biology↗

bioRxiv2019

A stable leading strand polymerase/clamp loader complex is required for normal and perturbed eukaryotic DNA replication

The eukaryotic replisome must faithfully replicate DNA and cope with replication fork blocks and stalling, while simultaneously promoting sister chromatid cohesion. Ctf18-RFC is an alternative PCNA loader that links all these processes together by an unknown mechanism. Here, we use integrative structural biology combined with yeast genetics and biochemistry to highlight the specific functions that Ctf18-RFC plays within the leading strand machinery via an interaction with the catalytic domain of DNA Pol {varepsilon}. We show that a large and unusually flexible interface enables this interaction to occur constitutively throughout the cell cycle and regardless of whether forks are replicating or stalled. We reveal that, by being anchored to the leading strand polymerase, Ctf18-RFC can rapidly signal fork stalling to activate the S phase checkpoint. Moreover, we demonstrate that, independently of checkpoint signaling or chromosome cohesion, Ctf18-RFC functions in parallel to Chl1 and Mrc1 to protect replication forks and cell viability.

biochemistry↗

bioRxiv2019

Improving Social Interactive Learning through Dual Brain Stimulation

Social interactive learning denotes the ability to acquire new information from a conspecific - a prerequisite for cultural evolution and survival. As inspired by recent neurophysiological research, here we tested whether social interactive learning can be augmented by exogenously synchronizing oscillatory brain activity across an instructor and a learner engaged in a naturalistic song-learning task. We used a dual brain stimulation protocol entailing the trans-cranial delivery of synchronized electric currents in two individuals simultaneously. When we stimulated inferior frontal brain regions, with 6 Hz alternating currents being in-phase between the instructor and the learner, the dyad exhibited spontaneous and synchronized body movement. Remarkably, this stimulation also led to enhanced learning performance. A mediation analysis further disclosed that interpersonal movement synchrony acted as a partial mediator of the effect of dual brain stimulation on learning performance, i.e. possibly facilitating the effect of dual brain stimulation on learning. Our results provide a causal demonstration that inter-brain synchrony is a sufficient condition to improve real-time information transfer between pairs of individuals.\n\nSignificanceThe study of social behavior, including but not limited to social learning, is undergoing a paradigm shift moving from single- to multi-person brain research. Yet, nearly all evidence in this area is purely correlational: inter-dependencies between brains signals are used to predict success in social behavior. For instance, inter-brain synchrony has been shown to be associated with successful communication, cooperation, and joint attention. Here we took a radically different approach. We stimulated two brains simultaneously, hence manipulating inter-brain synchrony, and measured the resulting effect upon behavior in the context of a social learning task. We report that frequency- and phase-specific dual brain stimulation can lead to the emergence of spontaneous synchronized body movement between an instructor and a learner. Remarkably, this can also augment learning performance.

neuroscience↗

bioRxiv2019

Computational fluid dynamics simulation of changes in the morphology and airflow dynamics of the upper airways in OSAHS patients after treatment with oral appliances

ObjectivesTo explore the changes of morphology and internal airflow in upper airways (UA) after the use of oral appliances (OAs) in patients with obstructive sleep apnea hypopnea syndrome (OSAHS), and investigate the mechanisms by which OAs function as a therapy for OSAHS.\n\nMethodsEight OSAHS patients (all male, aged 37-58, mean age 46.25) underwent CT scans before and after OA use. Then, computational fluid dynamics(CFD) models were built on the base of the CT scans using Mimics and ANSYS ICEM CFD software. The internal airflow of the upper airways was simulated using ANSYS-FLUENT and the results were analyzed using ANSYS-CFD-Post. The data were analyzed to identify the most important changes of biomechanical properties between patients with and without OA intervention. Upper airway morphology and the internal airflow changes were compared using t-tests and Spearman correlation coefficient analysis.\n\nResultsThe narrowest area of upper airways was found to be located in the lower bound of velopharynx, where the volume and pressure were statistically significantly increased (P<0.05) and the air velocity was statistically significantly decreased (P<0.05) in the presence of the OA(P<0.05). After wearing OA, pharyngeal resistance was significantly decreased (P<0.05), from 290.63 to 186.25Pa/L, and the airflow resistance of the pharynx has reduced by 35.9%.\n\nConclusionThe enlargement of the upper airway after wearing the OA changed its airflow dynamics, which decreased the negative pressure and resistance in narrow areas of the upper airways. Thus, the collapsibility of the upper airways was reduced and patency was sustained.

biophysics↗

bioRxiv2019

Flow-induced symmetry breaking in growing bacterial biofilms

Bacterial biofilms represent a major form of microbial life on Earth and serve as a model active nematic system, in which activity results from growth of the rod-shaped bacterial cells. In their natural environments, ranging from human organs to industrial pipelines, biofilms have evolved to grow robustly under significant fluid shear. Despite intense practical and theoretical interest, it is unclear how strong fluid flow alters the local and global architectures of biofilms. Here, we combine highly time-resolved single-cell live imaging with 3D multi-scale modeling to investigate the mechanisms by which flow affects the dynamics of all individual cells in growing biofilms. Our experiments and cell-based simulations reveal three quantitatively different growth phases in strong external flow, and the transitions between them. In the initial stages of biofilm development, flow induces a downstream gradient in cell orientation, causing asymmetrical droplet-like biofilm shapes. In the later developmental stages, when the majority of cells are sheltered from the flow by the surrounding extracellular matrix, buckling-induced cell verticalization in the biofilm core restores radially symmetric biofilm growth, in agreement with predictions of a 3D continuum model.

biophysics↗