Modification of kitchen blenders into controllable laboratory mixers for mechanochemical synthesis of atomically thin materials

Graphene and related two-dimensional materials (2DMs) have shown promise across numerous technology areas including flexible electronics, energy storage and pollution remediation. Research into novel applications of these atomically thin materials relies on access to synthesis techniques for producing 2DMs with suitable quality and quantity. Liquid-phase exfoliation is a mechanochemical approach that can achieve this and produce defect-free nanomaterial dispersions which are compatible for downstream use (e.g. inkjet printing, coatings). Here, using kitchen blenders to deliver shear-driven exfoliation, we develop a range of inexpensive hardware solutions that can allow researchers to synthesise 2DMs using a controllable, sustainable and scalable process. Extensive modifications were necessary as the onboard electronics lack the experimental controls (temperature, speed, characterisation) for scientific research and precision synthesis. The technical aspects (including the many lessons learned) of the modifications are discussed and a simple selection process is proposed for creating bespoke mechanochemical processors for any application in the hope that this encourages experimentation. Specific builds with detailed notes, cost breakdown and associated files are provided in the Open Science Framework (OSF) repository, OpenLPE associated with this article.

Scalable production of chitosan sub-micron particles by membrane ionotropic gelation process

Ionotropic gelation (IG) is a highly attractive method for the synthesis of natural water-soluble polymeric nanoparticles (NPs) and sub-micron particles (sMP) due to its relatively simple procedure and the absence of organic solvents. The method involves the electrostatic interaction between two ionic species of opposite charge. Although it is well studied at the laboratory scale, the difficulty to achieve size control in conventional bench-top process is actually a critical aspect of the technology. The aim of this work is to study the membrane dispersion technology in combination with IG as a suitable scalable method for the production of chitosan sub-micron particles (CS-sMPs). The two phases, one containing chitosan (CS) and the other containing sodium tripolyphosphate (TPP), were put in contact using a tubular hydrophobic glass membrane with a pore diameter of 1 μm. TPP (dispersed phase) was permeated through the membrane pores into the lumen side along which the CS solution (the continuous phase) flowed in batch recirculation or continuous single-pass operation mode. The influence of chemical variables (i.e. pH, concentration and mass ratio of polyelectrolyte species, emulsifier) and fluid-dynamic parameters (i.e. polyelectrolyte solution flow rate and their relative mass ratio) was studied to precisely tune the size of CS-Ps.

Solution-processable, robust and sustainable cooler via nano-structured engineering

Passive daytime radiative cooling (PDRC) materials simultaneously featuring aesthetic and safety distinctions demonstrate versatile applications beyond cooling buildings, while the integrated advantages of high strength, morphological reconfigurability, and sustainability remain challenging for the conventional PDRC materials. Herein, we designed a robust, custom-shaped and eco-friendly cooler via a scalable solution-processable strategy, involving the nano-scale assembly of nano cellulose (NC) and inorganic nanoparticle (e.g., ZrO₂, SiO₂, BaSO₄, and hydroxyapatite). The robust cooler shows an interesting "brick-and-mortar" structure, where the NC constructs interwoven framework (as brick structure) and the inorganic nanoparticle uniformly locates in the skeleton (as mortar structure), collectively contributing to high mechanical strength (>80 MPa) and flexibility. In addition, the structural and chemical distinctions enable our cooler to show a high solar reflectance (>96 %) and mid-infrared emissivity (>0.9), demonstrating a sub-ambient average temperature drop of 8.8 °C in long-term outdoor environments. The high-performance cooler with robustness, scalability and environmental friendliness, serves as a competitive participant toward the advanced PDRC materials in our low-carbon society.

A green and efficient synthetic methodology towards the synthesis of 1-allyl-6-chloro-4-oxo-1,4-dihydroquinoline-3-carboxamide derivatives

Quinolone is a privileged scaffold in medicinal chemistry and 4-Quinolone-3-Carboxamides have been reported to harbor vast therapeutic potential. However, conversion of N-1 substituted 4-Quinolone 3-Carboxylate to its corresponding carbamates is highly restrictive. This motivated us to adopt a much simpler, scalable and efficient methodology for the synthesis of highly pure N-1 substituted 4- Quinolone-3-Carboxamides with excellent yields. Our adopted methodology not only provides a robust pathway for the convenient synthesis of N-1 substituted 4- Quinolone-3-Carboxamides which can then be explored for their therapeutic potential, this may also be adaptable for the derivatization of other such less reactive carboxylate species.

RResolver: efficient short-read repeat resolution within ABySS

BACKGROUND

De novo genome assembly is essential to modern genomics studies. As it is not biased by a reference, it is also a useful method for studying genomes with high variation, such as cancer genomes. De novo short-read assemblers commonly use de Bruijn graphs, where nodes are sequences of equal length k, also known as k-mers. Edges in this graph are established between nodes that overlap by [Formula: see text] bases, and nodes along unambiguous walks in the graph are subsequently merged. The selection of k is influenced by multiple factors, and optimizing this value results in a trade-off between graph connectivity and sequence contiguity. Ideally, multiple k sizes should be used, so lower values can provide good connectivity in lesser covered regions and higher values can increase contiguity in well-covered regions. However, current approaches that use multiple k values do not address the scalability issues inherent to the assembly of large genomes.

RESULTS

Here we present RResolver, a scalable algorithm that takes a short-read de Bruijn graph assembly with a starting k as input and uses a k value closer to that of the read length to resolve repeats. RResolver builds a Bloom filter of sequencing reads which is used to evaluate the assembly graph path support at branching points and removes paths with insufficient support. RResolver runs efficiently, taking only 26 min on average for an ABySS human assembly with 48 threads and 60 GiB memory. Across all experiments, compared to a baseline assembly, RResolver improves scaffold contiguity (NGA50) by up to 15% and reduces misassemblies by up to 12%.

CONCLUSIONS

RResolver adds a missing component to scalable de Bruijn graph genome assembly. By improving the initial and fundamental graph traversal outcome, all downstream ABySS algorithms greatly benefit by working with a more accurate and less complex representation of the genome. The RResolver code is integrated into ABySS and is available at https://github.com/bcgsc/abyss/tree/master/RResolver .

Semi-supervised learning with natural language processing for right ventricle classification in echocardiography-a scalable approach

We created a deep learning model, trained on text classified by natural language processing (NLP), to assess right ventricular (RV) size and function from echocardiographic images. We included 12,684 examinations with corresponding written reports for text classification. After manual annotation of 1489 reports, we trained an NLP model to classify the remaining 10,651 reports. A view classifier was developed to select the 4-chamber or RV-focused view from an echocardiographic examination (n = 539). The final models were two image classification models trained on the predicted labels from the combined manual annotation and NLP models and the corresponding echocardiographic view to assess RV function (training set n = 11,008) and size (training set n = 9951. The text classifier identified impaired RV function with 99% sensitivity and 98% specificity and RV enlargement with 98% sensitivity and 98% specificity. The view classification model identified the 4-chamber view with 92% accuracy and the RV-focused view with 73% accuracy. The image classification models identified impaired RV function with 93% sensitivity and 72% specificity and an enlarged RV with 80% sensitivity and 85% specificity; agreement with the written reports was substantial (both κ = 0.65). Our findings show that models for automatic image assessment can be trained to classify RV size and function by using model-annotated data from written echocardiography reports. This pipeline for auto-annotation of the echocardiographic images, using a NLP model with medical reports as input, can be used to train an image-assessment model without manual annotation of images and enables fast and inexpensive expansion of the training dataset when needed.

Scalable Moiré Lattice with Oriented TMD Monolayers

Moiré lattice in artificially stacked monolayers of two-dimensional (2D) materials effectively modulates the electronic structures of materials, which is widely highlighted. Formation of the electronic Moiré superlattice promises the prospect of uniformity among different moiré cells across the lattice, enabling a new platform for novel properties, such as unconventional superconductivity, and scalable quantum emitters. Recently, epitaxial growth of the monolayer transition metal dichalcogenide (TMD) is achieved on the sapphire substrate by chemical vapor deposition (CVD) to realize scalable growth of highly-oriented monolayers. However, fabrication of the scalable Moiré lattice remains challenging due to the lack of essential manipulation of the well-aligned monolayers for clean interface quality and precise twisting angle control. Here, scalable and highly-oriented monolayers of TMD are realized on the sapphire substrates by using the customized CVD process. Controlled growth of the epitaxial monolayers is achieved by promoting the rotation of the nuclei-like domains in the initial growth stage, enabling aligned domains for further grain growth in the steady-state stage. A full coverage and distribution of the highly-oriented domains are verified by second-harmonic generation (SHG) microscopy. By developing the method for clean monolayer manipulation, hetero-stacked bilayer (epi-WS₂/epi-MoS₂) is fabricated with the specific angular alignment of the two major oriented monolayers at the edge direction of 0°/ ± 60°. On account of the optimization for scalable Moiré lattice with a high-quality interface, the observation of interlayer exciton at low temperature illustrates the feasibility of scalable Moiré superlattice based on the oriented monolayers.

Quiz-Based Microlearning at Scale: a Rapid Educational Response to COVID-19

Virtual quizzing is a viable model for continuing education at a large scale, particularly during the COVID-19 pandemic. By leveraging technology, microlearning encourages mobile education that is engaging, flexible, and accessible. Learners reported that this format was effective and preferable to traditional methods of education, suggesting further opportunity for innovation.

Expansion of Human Pluripotent Stem Cells in Stirred Tank Bioreactors

Bioreactor technolology enables the expansion of mammalian cells, which can be translated to processes compatible with Current Good Manufacturing Practice (cGMP) regulations. Cells are introduced to the bioreactor vessel, wherein key parameters such as temperature, pH, and oxygen levels are tightly controlled to facilitate growth over time. Here, we describe the microcarrier-based expansion of human pluripotent stem cells in a 3 L stirred tank bioreactor.

Boosting zinc ion energy storage capability of inert MnO cathode by defect engineering

Aqueous zinc ion battery constitutes a safe, stable and promising next-generation energy storage device, but suffers the lack of suitable host compounds for zinc ion storage. Development of a facile way to emerging cathode materials is strongly requested toward superior electrochemical activities and practical applications. Herein, defect engineering, i.e., simultaneous introduction of nitrogen dopant and oxygen vacancy into commercial and low-cost MnO, is proposed as a positive strategy to activate the originally inert phase for kinetically propelling its zinc ion storage capability. Both experimental characterization and theoretical calculations demonstrate that the nitrogen dopant significantly improves the electric conductivity of electrochemical inert MnO. Simultaneously, the oxygen vacancy creates sufficient large inserted channels and available activated adsorption sites for zinc ions storage. These synergistic structural advantages obviously ameliorate the electrochemical performance of inert MnO. Therefore, even without any conductive agent additive, the as-prepared material shows high specific capacity, superb rate capability, prolonged cycling stability and attractive energy density, which are dramatically superior to those of the pristine MnO as well as many other host cathode materials. This work presents fresh insights on the role of defect engineering in the enhancement of the intrinsic electrochemical reactivity of inert cathode, and an effective strategy for scalable fabrication of high-performance cathode for zinc ion battery.

Development and validation of a photobioreactor for uniform distribution of light intensity along the optical path based on numerical simulation

A theoretical approach was followed to optimize the design of a cylindrical photobioreactor for wastewater treatment based on algal culture. In particular, the problem of uneven light distribution that impairs algal growth was minimized by optimizing the area of uniform illumination distribution for a bioreactor design that can be enlarged without affecting its performance. The theoretical analysis was based on modeled simulations to determine the best configuration and illumination mode. The Monte Carlo method was used to simulate the illumination distribution inside the bioreactor, and the relationships between the width of the area with uniform illumination and related parameters were explored. Based on these theoretical considerations and predictions, an actual experimental photobioreactor was built containing a working area (where culture of Chlorella pyrenoidosa was enabled) and a catchment area for effluent. The performance of this bioreactor was tested with synthetic wastewater as a substrate. The light distribution was found to be relatively uniform inside the bioreactor, supporting excellent algal growth and resulting in maximum removal rates of 84.41% for total nitrogen, 99.73% for total phosphorus, 85.03% for NH₄⁺-N, and 75.94% for chemical oxygen demand (COD) over a period of 32 days of operation. The presented approach provides new insights for improving the efficiency and scalability of photobioreactors and promotes their development for wastewater treatment and resource utilization.

Bayesian sparse multiple regression for simultaneous rank reduction and variable selection

We develop a Bayesian methodology aimed at simultaneously estimating low-rank and row-sparse matrices in a high-dimensional multiple-response linear regression model. We consider a carefully devised shrinkage prior on the matrix of regression coefficients which obviates the need to specify a prior on the rank, and shrinks the regression matrix towards low-rank and row-sparse structures. We provide theoretical support to the proposed methodology by proving minimax optimality of the posterior mean under the prediction risk in ultra-high dimensional settings where the number of predictors can grow sub-exponentially relative to the sample size. A one-step post-processing scheme induced by group lasso penalties on the rows of the estimated coefficient matrix is proposed for variable selection, with default choices of tuning parameters. We additionally provide an estimate of the rank using a novel optimization function achieving dimension reduction in the covariate space. We exhibit the performance of the proposed methodology in an extensive simulation study and a real data example.

Rapid scale-up and production of active-loaded PEGylated liposomes

Manufacturing of liposomal nanomedicines (e.g. Doxil®/Caelyx®) is a challenging and slow process based on multiple-vessel and batch processing techniques. As a result, the translation of these nanomedicines from bench to bedside has been limited. Microfluidic-based manufacturing offers the opportunity to address this issue, and de-risk the wider adoption of nanomedicines. Here we demonstrate the applicability of microfluidics for continuous manufacturing of PEGylated liposomes encapsulating ammonium sulfate (250 mM). Doxorubicin was subsequently active-loaded into these pre-formed liposomes. Critical process parameters and material considerations demonstrated to influence the liposomal product attributes included solvent selection and lipid concentration, flow rate ratio, and temperature and duration used for drug loading. However, the total flow rate did not affect the liposome product characteristics, allowing high production speeds to be adopted. The final liposomal product comprised of 80-100 nm vesicles (PDI < 0.2) encapsulating ≥ 90% doxorubicin, with matching release profiles to the innovator product and is stable for at least 6 months. Additionally, vincristine and acridine orange were active-loaded into these PEGylated liposomes (≥ 90% and ~100 nm in size) using the same process. These results demonstrate the ability to produce active-loaded PEGylated liposomes with high encapsulation efficiencies and particle sizes which support tumour targeting.

TreeSwift: A massively scalable Python tree package

Phylogenetic trees are essential to evolutionary biology, and numerous methods exist that attempt to extract phylogenetic information applicable to a wide range of disciplines, such as epidemiology and metagenomics. Currently, the three main Python packages for trees are Bio.Phylo, DendroPy, and the ETE Toolkit, but as dataset sizes grow, parsing and manipulating ultra-large trees becomes impractical for these tools. To address this issue, we present TreeSwift, a user-friendly and massively scalable Python package for traversing and manipulating trees that is ideal for algorithms performed on ultra-large trees.

Developing a Scalable STEM Career Development Program for Elementary School-Aged Students

This article describes an approach to designing a scalable career development curriculum for elementary school students using minimal-cost and readily available resources. Content experts, veterinary medical students, university staff, teachers, community partners, evaluation experts, and a children's book illustrator developed a library of low-cost, culturally responsive, fun, and educationally engaging lessons to expose elementary school-aged students to scientific knowledge and careers in veterinary medicine. The home team piloted and evaluated the approach at a local community center. Teams in eight other states were provided materials to pilot and assess the program. Seven of those teams successfully piloted the program and provided evidence of child engagement. Although models, props, and other costly supplies enhance delivery of Science Technology Engineering and Math lessons, our experience with the delivery of this curriculum was proof of concept that a low-cost curricular model is one strategy to facilitate scaling and sustainability of an engaging veterinary science curriculum.

Integration of juvenile habitat quality and river connectivity models to understand and prioritise the management of barriers for Atlantic salmon populations across spatial scales

Diadromous fish populations are strongly affected by in-stream barriers that cause river network fragmentation, constraining productivity or preventing completion of their lifecycle. Removal or reduction of barrier impacts is a restoration measure associated with unambiguous benefits. Management of barriers is therefore often prioritised above other restoration actions. Barrier management is prioritised at local and national scales depending on funding. However, barrier prioritisation is potentially sub-optimal because existing tools do not consider habitat quality. Furthermore, effects of partial barriers (those passable under certain conditions) are uncertain, depending on location and potential cumulative effects. A framework is presented for assessing effects of impassable manmade barriers (IMBs) on longitudinal river network connectivity (percentage of upstream habitat accessible from the river mouth) for Atlantic salmon across spatial scales, using Scotland as an example. The framework integrates juvenile habitat quality and network connectivity models to (1) provide information necessary for local and national prioritisation of barriers, and (2) assess potential effects of passable manmade barriers (PMBs) within a sensitivity framework. If only IMBs are considered, high levels of longitudinal connectivity are observed across most of Scotland's rivers. Barrier prioritisation is sensitive to habitat weighting: not accounting for habitat quality can lead to over- or underestimating the importance of IMBs. Prioritisation is also highly sensitive to the passability of PMBs: if passability drops to <97% (combined up- and downstream passability), the mean effect of PMBs becomes greater than IMBs at the national level. Moreover, impacts on catchment connectivity, and thus production (number of juvenile salmon produced by the river), could be severe, suggesting a better understanding of the passability of PMBs is important for future management of migration barriers. The presented framework can be transferred to other catchments, regions, or countries where necessary data are available, making it a valuable tool to the broader restoration community.

Scalable novel PVDF based nanocomposite foam for direct blood contact and cardiac patch applications

Scalable novel beta phase polyvinylidene fluoride-poly(methyl methacrylate) (PVDF-PMMA) polymer blend based nanocomposite foam with hydroxyapatite (HAp) and titanium dioxide (TiO₂) as nanofillers (β-PVDF-PMMA/HAp/TiO₂) (β-PPHT-f), was prepared by using salt etching assisted solution casting method. The prepared β-PPHT-f nanocomposite material was characterized using XRD, FT-IR, SEM-EDS. The XRD and FTIR results confirmed the formation of β phase of β-PPHT-f. The SEM and EDS results confirmed the formation of high porous structured closed cell type morphology of β-PPHT-f. It also, confirmed the uniform distribution of Ti, Ca, P, N and O, in β-PPHT-f. Contact angle measurements performed using sessile drop method with water and EDTA treated blood (EDTA blood) as probe liquids revealed that β-PPHT-f is hydrophilic with contact angle of 48.2° as well as hemophilic with contact angle of 13.7°. Porosity, fluid absorption and retention investigation by gravimetric analysis revealed that β-PPHT-f was 89.2% porous and can absorb and retain 139.15% and 87.05% of water and blood, respectively. The hemolysis assay performed as per ASTM F756 procedure revealed that β-PPHT-f is non hemolytic. Also, the Leishman stained blood smears prepared from whole blood incubated with β-PPHT-f for 3, 4, 5 and 6 h at 37 °C revealed that the blood cells were not affected by β-PPHT-f, its surface morphology and elemental composition. H9c2 cell line studies on a transparent film prepared using β-PPHT-f revealed that the elemental composition of the nanocomposite favored H9c2 cell adhesion and differentiation. All the characterization results indicate that the newly developed scalable novel β-PPHT-f is hemocompatible and cardiomyocyte compatible, suggesting it as a useful material for direct blood contact and cardiac patch applications.

An improved synthesis of 1-methylcyclopropanol using the Kulinkovich reaction

An improved process for the preparation of 1-methylcyclopropanol using the Kulinkovich reaction is described. The use of titanium tetramethoxide as catalyst resulted in minimal side product formation. The reaction, isolation and purification procedures were optimized so they can be easily implemented in multi-purpose equipment.

Immobilisation of Delta-like 1 ligand for the scalable and controlled manufacture of hematopoietic progenitor cells in a stirred bioreactor

Background

Umbilical cord blood provides a source of hematopoietic stem cells for transplantation with immunological and availability advantages over conventional bone marrow sources. Limited cell numbers and slower engraftment from umbilical cord blood units has led to the clinical development of immobilised Notch ligand Delta-Like 1 to promote ex vivo expansion of a rapidly engrafting cell population. However, current immobilisation methods are not simple to scale in a controlled manner.

Results

Delta-Like 1 was immobilised onto streptavidin coated magnetic particles via a heterobifunctionalised polyethylene glycol linker molecule to provide an easily manipulated format of surface protein presentation. CD34⁺ enriched cord blood cells were treated with Delta-Like 1 immobilised particles, and immunophenotypic markers measured to monitor population distributions using cluster identification, characterization, and regression software. The amenability of the approach to scalability was evaluated in a micro-scale stirred tank bioreactor. Surface concentration of Delta-Like 1 was well controlled used differing stoichiometric reagent ratios. Protein immobilisation was a cost effective process and particles were efficiently removed from the final cell product. Immobilised Delta-Like 1 is functional and stimulates qualitatively similar CD34^hi, CD38^lo, CD90^lo, CD133^hi, CD135^hi progenitor expansion in both static culture and scalable stirred culture platforms.

Conclusions

Immobilised Delta-Like 1 in this form has the potential to improve the manufacturing efficiency and control of final ex vivo expanded cell product through compatibility with highly controlled and characterised suspension culture systems.

Electronic supplementary material

The online version of this article (doi:10.1186/s12896-017-0383-0) contains supplementary material, which is available to authorized users.

A hybrid computational strategy to address WGS variant analysis in >5000 samples

Background

The decreasing costs of sequencing are driving the need for cost effective and real time variant calling of whole genome sequencing data. The scale of these projects are far beyond the capacity of typical computing resources available with most research labs. Other infrastructures like the cloud AWS environment and supercomputers also have limitations due to which large scale joint variant calling becomes infeasible, and infrastructure specific variant calling strategies either fail to scale up to large datasets or abandon joint calling strategies.

Results

We present a high throughput framework including multiple variant callers for single nucleotide variant (SNV) calling, which leverages hybrid computing infrastructure consisting of cloud AWS, supercomputers and local high performance computing infrastructures. We present a novel binning approach for large scale joint variant calling and imputation which can scale up to over 10,000 samples while producing SNV callsets with high sensitivity and specificity. As a proof of principle, we present results of analysis on Cohorts for Heart And Aging Research in Genomic Epidemiology (CHARGE) WGS freeze 3 dataset in which joint calling, imputation and phasing of over 5300 whole genome samples was produced in under 6 weeks using four state-of-the-art callers. The callers used were SNPTools, GATK-HaplotypeCaller, GATK-UnifiedGenotyper and GotCloud. We used Amazon AWS, a 4000-core in-house cluster at Baylor College of Medicine, IBM power PC Blue BioU at Rice and Rhea at Oak Ridge National Laboratory (ORNL) for the computation. AWS was used for joint calling of 180 TB of BAM files, and ORNL and Rice supercomputers were used for the imputation and phasing step. All other steps were carried out on the local compute cluster. The entire operation used 5.2 million core hours and only transferred a total of 6 TB of data across the platforms.

Conclusions

Even with increasing sizes of whole genome datasets, ensemble joint calling of SNVs for low coverage data can be accomplished in a scalable, cost effective and fast manner by using heterogeneous computing platforms without compromising on the quality of variants.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-016-1211-6) contains supplementary material, which is available to authorized users.

Comparability of automated human induced pluripotent stem cell culture: a pilot study

Consistent and robust manufacturing is essential for the translation of cell therapies, and the utilisation automation throughout the manufacturing process may allow for improvements in quality control, scalability, reproducibility and economics of the process. The aim of this study was to measure and establish the comparability between alternative process steps for the culture of hiPSCs. Consequently, the effects of manual centrifugation and automated non-centrifugation process steps, performed using TAP Biosystems’ CompacT SelecT automated cell culture platform, upon the culture of a human induced pluripotent stem cell (hiPSC) line (VAX001024c07) were compared. This study, has demonstrated that comparable morphologies and cell diameters were observed in hiPSCs cultured using either manual or automated process steps. However, non-centrifugation hiPSC populations exhibited greater cell yields, greater aggregate rates, increased pluripotency marker expression, and decreased differentiation marker expression compared to centrifugation hiPSCs. A trend for decreased variability in cell yield was also observed after the utilisation of the automated process step. This study also highlights the detrimental effect of the cryopreservation and thawing processes upon the growth and characteristics of hiPSC cultures, and demonstrates that automated hiPSC manufacturing protocols can be successfully transferred between independent laboratories.

A scalable method for biochemical purification of Salmonella flagellin

•

Highly purified flagellins obtained from liquid bacterial fermentation supernatants.

•

Purification accomplished in four process steps to clarify, bind, wash, polish.

•

Ion-exchange membranes exhibit improved flagellin retention compared to resins.

•

Purified flagellins retain epitope conformation, innate immune biological activity.

Flagellins are the main structural proteins of bacterial flagella and potent stimulators of innate and adaptive immunity in mammals. The flagellins of Salmonella are virulence factors and protective antigens, and form the basis of promising vaccines. Despite broad interest in flagellins as antigens and adjuvants in vaccine formulations, there have been few advances towards the development of scalable and economical purification methods for these proteins. We report here a simple and robust strategy to purify flagellin monomers from the supernatants of liquid growth culture. Phase 1 flagellins from Salmonella enterica serovars Typhimurium (i epitope) and Enteritidis (g,m epitopes) were purified directly from conditioned fermentation growth media using sequential cation- and anion-exchange chromatography coupled with a final tangential flow-filtration step. Conventional porous chromatography resin was markedly less efficient than membrane chromatography for flagellin purification. Recovery after each process step was robust, with endotoxin, nucleic acid and residual host–cell protein effectively removed. The final yield was 200–300 mg/L fermentation culture supernatant, with ∼45–50% overall recovery. A final pH 2 treatment step was instituted to ensure uniformity of flagellin in the monomeric form. Flagellins purified by this method were recognized by monoclonal anti-flagellin antibodies and maintained capacity to activate Toll-like Receptor 5. The process described is simple, readily scalable, uses standard bioprocess methods, and requires only a few steps to obtain highly purified material.