Matrix sketching framework for linear mixed models in association studies.

Linear mixed models (LMMs) have been widely used in genome-wide association studies to control for population stratification and cryptic relatedness. However, estimating LMM parameters is computationally expensive, necessitating large-scale matrix operations to build the genetic relationship matrix (GRM). Over the past 25 years, Randomized Linear Algebra has provided alternative approaches to such matrix operations by leveraging , which often results in provably accurate fast and efficient approximations.

High-throughput genetic engineering of nonmodel and undomesticated bacteria via iterative site-specific genome integration.

Efficient genome engineering is critical to understand and use microbial functions. Despite recent development of tools such as CRISPR-Cas gene editing, efficient integration of exogenous DNA with well-characterized functions remains limited to model bacteria. Here, we describe serine recombinase-assisted genome engineering, or SAGE, an easy-to-use, highly efficient, and extensible technology that enables selection marker-free, site-specific genome integration of up to 10 DNA constructs, often with efficiency on par with or superior to replicating plasmids.

Bacterial catabolism of acetovanillone, a lignin-derived compound.

Bacterial catabolic pathways have considerable potential as industrial biocatalysts for the valorization of lignin, a major component of plant-derived biomass. Here, we describe a pathway responsible for the catabolism of acetovanillone, a major component of several industrial lignin streams. GD02 was previously isolated for growth on acetovanillone.

Bacterial Transformation of Aromatic Monomers in Softwood Black Liquor.

The valorization of lignin, a major component of plant-derived biomass, is essential to sustainable biorefining. We identified the major monoaromatic compounds present in black liquor, a lignin-rich stream generated in the kraft pulping process, and investigated their bacterial transformation. Among tested solvents, acetone extracted the greatest amount of monoaromatic compounds from softwood black liquor, with guaiacol, vanillin, and acetovanillone, in an approximately 4:3:2 ratio, constituting ~90% of the total extracted monoaromatic content.

Generative Adversarial Networks for Creating Synthetic Free-Text Medical Data: A Proposal for Collaborative Research and Re-use of Machine Learning Models.

Restrictions in sharing Patient Health Identifiers (PHI) limit cross-organizational re-use of free-text medical data. We leverage Generative Adversarial Networks (GAN) to produce synthetic unstructured free-text medical data with low re-identification risk, and assess the suitability of these datasets to replicate machine learning models. We trained GAN models using unstructured free-text laboratory messages pertaining to salmonella, and identified the most accurate models for creating synthetic datasets that reflect the informational characteristics of the original dataset.

Tandem chemical deconstruction and biological upcycling of poly(ethylene terephthalate) to β-ketoadipic acid by Pseudomonas putida KT2440.

Poly(ethylene terephthalate) (PET) is the most abundantly consumed synthetic polyester and accordingly a major source of plastic waste. The development of chemocatalytic approaches for PET depolymerization to monomers offers new options for open-loop upcycling of PET, which can leverage biological transformations to higher-value products. To that end, here we perform four sequential metabolic engineering efforts in Pseudomonas putida KT2440 to enable the conversion of PET glycolysis products via: (i) ethylene glycol utilization by constitutive expression of native genes, (ii) terephthalate (TPA) catabolism by expression of tphA2A3BA1 from Comamonas and tpaK from Rhodococcus jostii, (iii) bis(2-hydroxyethyl) terephthalate (BHET) hydrolysis to TPA by expression of PETase and MHETase from Ideonella sakaiensis, and (iv) BHET conversion to a performance-advantaged bioproduct, β-ketoadipic acid (βKA) by deletion of pcaIJ.

Production of itaconic acid from alkali pretreated lignin by dynamic two stage bioconversion.

Expanding the portfolio of products that can be made from lignin will be critical to enabling a viable bio-based economy. Here, we engineer Pseudomonas putida for high-yield production of the tricarboxylic acid cycle-derived building block chemical, itaconic acid, from model aromatic compounds and aromatics derived from lignin. We develop a nitrogen starvation-detecting biosensor for dynamic two-stage bioproduction in which itaconic acid is produced during a non-growth associated production phase.

Engineered Pseudomonas putida simultaneously catabolizes five major components of corn stover lignocellulose: Glucose, xylose, arabinose, p-coumaric acid, and acetic acid.

Valorization of all major lignocellulose components, including lignin, cellulose, and hemicellulose is critical for an economically viable bioeconomy. In most biochemical conversion approaches, the standard process separately upgrades sugar hydrolysates and lignin. Here, we present a new process concept based on an engineered microbe that could enable simultaneous upgrading of all lignocellulose streams, which has the ultimate potential to reduce capital cost and enable new metabolic engineering strategies.

Generalization of Machine Learning Approaches to Identify Notifiable Conditions from a Statewide Health Information Exchange.

Healthcare analytics is impeded by a lack of machine learning (ML) model generalizability, the ability of a model to predict accurately on varied data sources not included in the model's training dataset. We leveraged free-text laboratory data from a Health Information Exchange network to evaluate ML generalization using Notifiable Condition Detection (NCD) for public health surveillance as a use case. We 1) built ML models for detecting syphilis, salmonella, and histoplasmosis; 2) evaluated generalizability of these models across data from holdout lab systems, and; 3) explored factors that influence weak model generalizability.

Randomized Linear Algebra Approaches to Estimate the Von Neumann Entropy of Density Matrices.

The , named after John von Neumann, is an extension of the classical concept of entropy to the field of quantum mechanics. From a numerical perspective, von Neumann entropy can be computed simply by computing all eigenvalues of a density matrix, an operation that could be prohibitively expensive for large-scale density matrices. We present and analyze three randomized algorithms to approximate von Neumann entropy of real density matrices: our algorithms leverage recent developments in the Randomized Numerical Linear Algebra (RandNLA) literature, such as randomized trace estimators, provable bounds for the power method, and the use of random projections to approximate the eigenvalues of a matrix.

An Adversorial Approach to Enable Re-Use of Machine Learning Models and Collaborative Research Efforts Using Synthetic Unstructured Free-Text Medical Data.

We leverage Generative Adversarial Networks (GAN) to produce synthetic free-text medical data with low re-identification risk, and apply these to replicate machine learning solutions. We trained GAN models to generate free-text cancer pathology reports. Decision models were trained using synthetic datasets reported performance metrics that were statistically similar to models trained using original test data.

Minimum Effective Dose of Cattle and Sheep BSE for Oral Sheep Infection.

The minimum dose required to cause infection of Romney and Suffolk sheep of the ARQ/ARQ or ARQ/ARR prion protein gene genotypes following oral inoculation with Romney or Suffolk a sheep Bovine spongiform encephalopathy (BSE)-derived or cattle BSE-derived agent was investigated using doses ranging from 0.0005g to 5g. ARQ/ARQ sheep which were methionine (M) / threonine (T) heterozygous or T/T homozygous at codon 112 of the Prnp gene, dosed ARQ/ARR sheep and undosed controls did not show any evidence of infection.

Dynamics of the natural transmission of bovine spongiform encephalopathy within an intensively managed sheep flock.

Sheep are susceptible to the bovine spongiform encephalopathy (BSE) agent and in the UK they may have been exposed to BSE via contaminated meat and bone meal. An experimental sheep flock was established to determine whether ovine BSE could be naturally transmitted under conditions of intensive husbandry. The flock consisted of 113 sheep of different breeds and susceptible PRNP genotypes orally dosed with BSE, 159 sheep subsequently born to them and 125 unchallenged sentinel controls.

Nazarov reactions intercepted by (4 + 3) cycloadditions with oxygen-substituted dienes.

The oxyallyl cation intermediate from the Lewis acid mediated Nazarov reaction of an allenyl vinyl ketone was intercepted by acyclic, 2-silyloxy-substituted butadienes by highly regioselective (4 + 3) cycloadditions. Stereoselectivity was often modest, but in some instances steric interactions were responsible for high selectivity. The results are consistent with concerted (4 + 3) cycloadditions.

Influence of breed and genotype on the onset and distribution of infectivity and disease-associated prion protein in sheep following oral infection with the bovine spongiform encephalopathy agent.

The onset and distribution of infectivity and disease-specific prion protein (PrP(d)) accumulation was studied in Romney and Suffolk sheep of the ARQ/ARQ, ARQ/ARR and ARR/ARR prion protein gene (Prnp) genotypes (where A stands for alanine, R for arginine and Q for glutamine at codons 136, 154 and 171 of PrP), following experimental oral infection with cattle-derived bovine spongiform encephalopathy (BSE) agent. Groups of sheep were killed at regular intervals and a wide range of tissues taken for mouse bioassay or immunohistochemistry (IHC), or both. Bioassay results for infectivity were mostly coincident with those of PrP(d) detection by IHC both in terms of tissues and time post infection.

A qualitative exploration of child clinical psychologists' understanding of user involvement.

'Service User Involvement' is a key directive for mental health services. This is thought to be especially complex in child services-despite evidence that it can be achieved-because of the need to use developmentally-appropriate tools. Children are in a multi-faceted position of disempowerment when they enter mental health services; attempts to involve them in these services are entangled with intricate power issues.

The evaluation of exposure risks for natural transmission of scrapie within an infected flock.

Background: Although the epidemiology of scrapie has been broadly understood for many years, attempts to introduce voluntary or compulsory controls to eradicate the disease have frequently failed. Lack of precision in defining the risk factors on farm has been one of the challenges to designing control strategies. This study attempted to define which parts of the annual flock management cycle represented the greatest risk of infection to naive lambs exposed to the farm environment at different times.

Accumulation and dissemination of prion protein in experimental sheep scrapie in the natural host.

Background: In order to study the sites of uptake and mechanisms of dissemination of scrapie prions in the natural host under controlled conditions, lambs aged 14 days and homozygous for the VRQ allele of the PrP gene were infected by the oral route. Infection occurred in all lambs with a remarkably short and highly consistent incubation period of approximately 6 months. Challenge of lambs at approximately eight months of age resulted in disease in all animals, but with more variable incubation periods averaging significantly longer than those challenged at 14 days.

Pruritus is a common feature in sheep infected with the BSE agent.

Background: The variability in the clinical or pathological presentation of transmissible spongiform encephalopathies (TSEs) in sheep, such as scrapie and bovine spongiform encephalopathy (BSE), has been attributed to prion protein genotype, strain, breed, clinical duration, dose, route and type of inoculum and the age at infection. The study aimed to describe the clinical signs in sheep infected with the BSE agent throughout its clinical course to determine whether the clinical signs were as variable as described for classical scrapie in sheep. The clinical signs were compared to BSE-negative sheep to assess if disease-specific clinical markers exist.

Diagnosis of preclinical scrapie in live sheep by the immunohistochemical examination of rectal biopsies.

In most sheep infected with a transmissible spongiform encephalopathy (tse) the disease-associated prion protein (PrP(d)) accumulates in tissues of the lymphoreticular system, suggesting that it might be detected in biopsy specimens. A procedure has been developed to obtain biopsy specimens of rectal mucosa in which PrP(d) has been detected by immunohistochemistry in preclinically infected sheep of all susceptible PrP genotypes. It is probable that PrP(d) increases with the age of sheep or period of incubation.

Tissue distribution of bovine spongiform encephalopathy infectivity in Romney sheep up to the onset of clinical disease after oral challenge.

Sixty Romney sheep of three prion protein genotypes were dosed orally at six months of age with an inoculum prepared from the brains of cattle clinically affected with BSE, and 15 sheep were left undosed as controls. They were randomly assigned within genotype to groups and were sequentially euthanased and examined postmortem at intervals of six or 12 months, depending on their predicted susceptibility. Tissue pools prepared from the three, four or five dosed animals in each group were inoculated into groups of 20 RIII mice as a bioassay for infectivity.