Biomathematics and Statistics Scotland BioSS

Bucksburn, United Kingdom

Biomathematics and Statistics Scotland BioSS

Bucksburn, United Kingdom
Time filter
Source Type

NEWTON A.C.,James Hutton Institute | GUY D.C.,James Hutton Institute | PREEDY K.,Biomathematics and Statistics Scotland BioSS
Journal of Agricultural Science | Year: 2017

A range of wheat cultivars, including elite cultivars, older cultivars and some preferred by organic growers, were trialled under high and low nitrogen (N) conventional and organic conditions to determine whether cultivars that yield highly under organic conditions have the same relative yield under conventional conditions. A range of cultivar mixtures was also assessed to see whether these gave yield advantages or superiority in either farming system. The conventional trials were grown with and without full fungicide programmes, which largely controlled disease. Amongst the cultivars, Alchemy showed superior yield under organic conditions as did Pegassos, but under conventional conditions Pegassos was always one of the low-ranking cultivars. Under conventional conditions the more recent cultivars Alchemy, Glasgow and Istabraq yielded highly, while an older one, Consort, yielded highly under low fertilizer conditions, and both Ambrosia and Deben also yielded highly generally. Fungicide and high N favoured the disease-susceptible, high-yield cultivars such as Glasgow whereas Consort, an older susceptible cultivar, was favoured by fungicide and low N. Together this demonstrates that whilst the yield characteristics of some elite germplasm are also expressed under organic conditions, at least one cultivar that yielded poorly under conventional conditions showed adaptation towards the organic conditions of these trials. Other cultivars yielding poorly under conventional conditions also gave poor yield under organic conditions. The equal proportion mixtures of cultivars grown under conventional conditions showed no evidence of differences in yield from the mean of the component cultivars grown separately, but combinations of Glasgow, Alchemy and Istabraq gave consistently high yield. Copyright © Cambridge University Press 2017

Ostertag L.M.,University of Aberdeen | Ostertag L.M.,UK Institute of Food Research | Kroon P.A.,UK Institute of Food Research | Wood S.,University of Aberdeen | And 5 more authors.
Molecular Nutrition and Food Research | Year: 2013

We examined whether flavan-3-ol-enriched dark chocolate, compared with standard dark and white chocolate, beneficially affects platelet function in healthy subjects, and whether this relates to flavan-3-ol bioavailability. Methods and results: A total of 42 healthy subjects received an acute dose of flavan-3-ol-enriched dark, standard dark or white chocolate, in random order. Blood and urine samples were obtained just before and 2 and 6 h after consumption for measurements of platelet function, and bioavailability and excretion of flavan-3-ols. Flavan-3-ol-enriched dark chocolate significantly decreased adenosine diphosphate-induced platelet aggregation and P-selectin expression in men (all p ≤ 0.020), decreased thrombin receptor-activating peptide-induced platelet aggregation and increased thrombin receptor-activating peptide-induced fibrinogen binding in women (both p ≤ 0.041), and increased collagen/epinephrine-induced ex vivo bleeding time in men and women (p ≤ 0.042). White chocolate significantly decreased adenosine diphosphate-induced platelet P-selectin expression (p = 0.002) and increased collagen/epinephrine-induced ex vivo bleeding time (p = 0.042) in men only. Differences in efficacy by which flavan-3-ols affect platelet function were only partially explained by concentrations of flavan-3-ols and their metabolites in plasma or urine. Conclusion: Flavan-3-ols in dark chocolate, but also compounds in white chocolate, can improve platelet function, dependent on gender, and may thus beneficially affect atherogenesis. © 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Lobley G.E.,University of Aberdeen | Holtrop G.,Biomathematics and Statistics Scotland BioSS | Horgan G.W.,Biomathematics and Statistics Scotland BioSS | Bremner D.M.,University of Aberdeen | And 2 more authors.
British Journal of Nutrition | Year: 2015

High-protein diets are an effective means for weight loss (WL), but the mechanisms are unclear. One hypothesis relates to the release of gut hormones by either protein or amino acids (AA). The present study involved overweight and obese male volunteers (n 18, mean BMI 36·8' kg/m2) who consumed a maintenance diet for 7' d followed by fully randomised 10' d treatments with three iso-energetic WL diets, i.e. with either normal protein (NP, 15' % of energy) or high protein (HP, 30' %) or with a combination of protein and free AA, each 15' % of energy (NPAA). Psychometric ratings of appetite were recorded hourly. On day 10, plasma samples were taken at 30' min intervals over two consecutive 5' h periods (covering post-breakfast and post-lunch) and analysed for AA, glucose and hormones (insulin, total glucose-dependent insulinotropic peptide, active ghrelin and total peptide YY (PYY)) plus leucine kinetics (first 5' h only). Composite hunger was 16' % lower for the HP diet than for the NP diet (P<' 0·01) in the 5' h period after both meals. Plasma essential AA concentrations were greatest within 60' min of each meal for the NPAA diet, but remained elevated for 3-5' h after the HP diet. The three WL diets showed no difference for either fasting concentrations or the postprandial net incremental AUC (net AUCi) for insulin, ghrelin or PYY. No strong correlations were observed between composite hunger scores and net AUCi for either AA or gut peptides. Regulation of hunger may involve subtle interactions, and a range of signals may need to be integrated to produce the overall response. © The Authors 2015.

Lapierre H.,Agriculture and Agri Food Canada | Holtrop G.,Biomathematics and Statistics Scotland BioSS | Calder A.G.,University of Aberdeen | Renaud J.,Agriculture and Agri Food Canada | Lobley G.E.,University of Aberdeen
Journal of Dairy Science | Year: 2012

Rumen-protected forms of Met contain an equimolar mixture of the d- and l-isomers. Only l-Met can be directly used for protein synthesis, but it is unclear how much of the d-isomer can be transformed into l-Met in ruminants. Four lactating dairy cows, with an average milk yield of, received a basal diet (12.5% crude protein, supplying 1,718g/d of metabolizable protein) in 12 equal meals per day plus an abomasal infusion of amino acids (59g/d, casein profile without Met). They were used in 3 consecutive studies to determine utilization of d-Met. First, the cows each received portal vein infusions for d of 5, 10, or 1g/d of dl-Met in a Youden square. On the last day of each period, 6 arterial samples were collected at 45-min intervals. Concentrations of l- and d-Met were determined on a chiral column by gas chromatography-mass spectrometry. Portal infusion of 5, 10, and 1g/d of dl-Met increased plasma total Met concentrations (19.7, 25.0, and 34.4±0.6μM) and the proportion of Met as d (19.4, 30.5, and 37.3±0.7%). The fractional removal of d-Met was 6 to 7 times lower than the fractional removal of L-Met, with mean half-lives of 52 versus 8min, respectively. Second, the same cows were infused for 8h with l[methyl- 2H 3]Met at 1.3mmol/h; at 2h, cows received a bolus injection i.v. of d-[1- 13C]Met (6.8mmol), and arterial samples were collected after 10, 20, 30, 40, 60, 90, 120, 150, 180, 240, 300, 360, 420, and 480min. Expressed relative to l-[ 12C]Met; that is, as tracer:tracee ratios, enrichments of plasma d-[1- 13C]Met and l-[1- 13C]Met averaged 1.77±0.14 and 0.144±0.026, respectively, 10min after the bolus injection and declined exponentially thereafter. A minimum of 75±3% of the d-[1- 13C]Met was transformed into l-[1- 13C]Met. Third, the cows received, in a crossover design, an abomasal infusion for d of either dl-Met or l-Met (1g/d) and, on the last day of each experimental period, blood samples were collected simultaneously from arterial, portal, hepatic, and mammary vessels. Arterial total Met concentrations were higher with dl- versus l-Met infusions (37.4 vs. 25.4±0.5μM), with 37.1±5.0% as d-Met. The mammary gland did not extract any d-Met. Hepatic removal of d-Met was observed, but was numerically lower than the fractional extraction of l-Met. In conclusion, much of the d-Met is transformed into l-Met by the dairy cow but at a slow rate. No uptake of d-Met occurs across the mammary gland but l-Met synthesized from the d-isomer elsewhere in the body can be utilized for milk protein synthesis. © 2012 American Dairy Science Association.

Robinson G.,UK Cryptosporidium Reference Unit | Wright S.,Creative Science Company Science 310263 | Elwin K.,UK Cryptosporidium Reference Unit | Hadfield S.J.,UK Cryptosporidium Reference Unit | And 6 more authors.
International Journal for Parasitology | Year: 2010

To provide re-description of Cryptosporidium cuniculus Inman and Takeuchi, 1979 (synonymous with rabbit genotype), a species closely related to Cryptosporidium hominis, the morphology, natural and experimental host specificity, and genetic characterisation were investigated. The morphology and diagnostic characteristics are typical of other intestinal species of Cryptosporidium, albeit with slightly larger oocysts (5.55-6.40 × 5.02-5.92 μm; mean 5.98 × 5.38 μm; length:width = 1.1; n= 50). Natural hosts appear to be European rabbits (Oryctolagus cuniculus) and humans (Homo sapiens). Experimental infections have been established in weanling rabbits (O. cuniculus), immunosuppressed Mongolian gerbils (Meriones unguiculatus) and immunosuppressed adult Porton strain mice (Mus musculus), but not in neonatal mice. Patterns of infection measured by oocyst shedding are significantly different compared with C. hominis, particularly in rabbits. Histological examination reveals endogenous stages in the brush border of the epithelium of the small intestinal villi, but clinical signs are absent. Inoculation of human HCT-8 cells results in discrete clusters of endogenous stages. A close relationship with C. hominis is inferred from molecular analyses at the ssrRNA, 70. kDa heat shock protein (HSP70), actin, Cryptosporidium oocyst wall protein (COWP), 60. kDa glycoprotein (GP60) genes and a region encoding a product of unknown function (LIB13). Sequences contained limited, consistent polymorphisms at the ssrRNA, HSP70 and actin genes, were identical at the COWP and LIB13 genes and demonstrated two unique families at the GP60 gene. Although genetically closely related, there are significant biological differences between C. cuniculus and C. hominis that support these protozoa being separate species. This is based on the current understanding of these organisms and relies on the assumption that mating between these species would not normally occur. If this is subsequently demonstrated their categorisation may need to be re-addressed. © 2010 Australian Society for Parasitology Inc.

Grzegorczyk M.,TU Dortmund | Husmeier D.,Biomathematics and Statistics Scotland BioSS
Bioinformatics | Year: 2011

Method: Dynamic Bayesian networks (DBNs) have been applied widely to reconstruct the structure of regulatory processes from time series data, and they have established themselves as a standard modelling tool in computational systems biology. The conventional approach is based on the assumption of a homogeneous Markov chain, and many recent research efforts have focused on relaxing this restriction. An approach that enjoys particular popularity is based on a combination of a DBN with a multiple changepoint process, and the application of a Bayesian inference scheme via reversible jump Markov chain Monte Carlo (RJMCMC). In the present article, we expand this approach in two ways. First, we show that a dynamic programming scheme allows the changepoints to be sampled from the correct conditional distribution, which results in improved convergence over RJMCMC. Second, we introduce a novel Bayesian clustering and information sharing scheme among nodes, which provides a mechanism for automatic model complexity tuning. Results: We evaluate the dynamic programming scheme on expression time series for Arabidopsis thaliana genes involved in circadian regulation. In a simulation study we demonstrate that the regularization scheme improves the network reconstruction accuracy over that obtained with recently proposed inhomogeneous DBNs. For gene expression profiles from a synthetically designed Saccharomyces cerevisiae strain under switching carbon metabolism we show that the combination of both: dynamic programming and regularization yields an inference procedure that outperforms two alternative established network reconstruction methods from the biology literature. © The Author 2010. Published by Oxford University Press. All rights reserved.

Grzegorczyk M.,TU Dortmund | Husmeier D.,Biomathematics and Statistics Scotland BioSS
Machine Learning | Year: 2011

Classical dynamic Bayesian networks (DBNs) are based on the homogeneous Markov assumption and cannot deal with non-homogeneous temporal processes. Various approaches to relax the homogeneity assumption have recently been proposed. The present paper presents a combination of a Bayesian network with conditional probabilities in the linear Gaussian family, and a Bayesian multiple changepoint process, where the number and location of the changepoints are sampled from the posterior distribution with MCMC. Our work improves four aspects of an earlier conference paper: it contains a comprehensive and self-contained exposition of the methodology; it discusses the problem of spurious feedback loops in network reconstruction; it contains a comprehensive comparative evaluation of the network reconstruction accuracy on a set of synthetic and real-world benchmark problems, based on a novel discrete changepoint process; and it suggests new and improved MCMC schemes for sampling both the network structures and the changepoint configurations from the posterior distribution. The latter study compares RJMCMC, based on changepoint birth and death moves, with two dynamic programming schemes that were originally devised for Bayesian mixture models. We demonstrate the modifications that have to be made to allow for changing network structures, and the critical impact that the prior distribution on changepoint configurations has on the overall computational complexity. © 2011 The Author(s).

Wilson K.,Moredun Research Institute | Sammin D.,DAFF Laboratories | Harmeyer S.,Pfizer | Nath M.,Biomathematics and Statistics Scotland BioSS | And 2 more authors.
Veterinary Journal | Year: 2012

Although few studies have investigated the prevalence of chlamydial infections in cattle, reported prevalence rates vary hugely. In order to assess the prevalence of this infection in cattle in Ireland, serum samples (100 herds, 20 samples/herd) collected for statutory screening for brucellosis were examined by soluble chlamydial antigen indirect ELISA. The assay detects antibodies to the two most common Chlamydiaceae spp. affecting cattle, namely Chlamydia abortus and Chlamydia pecorum. A total of 95 samples from 57 herds were seropositive, representing an observed prevalence rate of 4.75%. The parametric bootstrap estimate of the mean disease prevalence in the population was 6.04% (95%, CI 4.70-7.50). The results suggest the prevalence of chlamydial infection is low in cattle in Ireland. © 2012 Elsevier Ltd.

Liu H.,James Hutton Institute | Liu H.,University of St. Andrews | Bayer M.,James Hutton Institute | Druka A.,James Hutton Institute | And 6 more authors.
BMC Genomics | Year: 2014

Abstract: We explored the use of genotyping by sequencing (GBS) on a recombinant inbred line population (GPMx) derived from a cross between the two-rowed barley cultivar 'Golden Promise' (ari-e.GP/Vrs1) and the six-rowed cultivar 'Morex' (Ari-e/vrs1) to map plant height. We identified three Quantitative Trait Loci (QTL), the first in a region encompassing the spike architecture gene Vrs1 on chromosome 2H, the second in an uncharacterised centromeric region on chromosome 3H, and the third in a region of chromosome 5H coinciding with the previously described dwarfing gene Breviaristatum-e (Ari-e).Background: Barley cultivars in North-western Europe largely contain either of two dwarfing genes; Denso on chromosome 3H, a presumed ortholog of the rice green revolution gene OsSd1, or Breviaristatum-e (ari-e) on chromosome 5H. A recessive mutant allele of the latter gene, ari-e.GP, was introduced into cultivation via the cv. 'Golden Promise' that was a favourite of the Scottish malt whisky industry for many years and is still used in agriculture today.Results: Using GBS mapping data and phenotypic measurements we show that ari-e.GP maps to a small genetic interval on chromosome 5H and that alternative alleles at a region encompassing Vrs1 on 2H along with a region on chromosome 3H also influence plant height. The location of Ari-e is supported by analysis of near-isogenic lines containing different ari-e alleles. We explored use of the GBS to populate the region with sequence contigs from the recently released physically and genetically integrated barley genome sequence assembly as a step towards Ari-e gene identification.Conclusions: GBS was an effective and relatively low-cost approach to rapidly construct a genetic map of the GPMx population that was suitable for genetic analysis of row type and height traits, allowing us to precisely position ari-e.GP on chromosome 5H. Mapping resolution was lower than we anticipated. We found the GBS data more complex to analyse than other data types but it did directly provide linked SNP markers for subsequent higher resolution genetic analysis. © 2014 Liu et al.; licensee BioMed Central Ltd.

Kettle H.,Biomathematics and Statistics Scotland BioSS | Nutter D.,Biomathematics and Statistics Scotland BioSS
Methods in Ecology and Evolution | Year: 2015

We present an r-package, stagePop, which is a tool for predicting the deterministic dynamics and interactions of stage-structured populations (i.e. where the life cycle consists of distinct stages, for example eggs, juveniles and reproductive adults). The continuous-time formulation enables stagePop to easily simulate time-varying stage durations, overlapping generations and density-dependent vital rates. The package can be used to predict predator-prey interactions, host-parasitoid interactions, resource competition, intra-specific competition and the effects of environmental change on stage-structured (and non-stage structured) species. It can be used for predicting the effects of bio-control and climate change, investigating mechanisms for maintaining diversity, predicting the dynamics of complex food webs following perturbation and so on. The code is based on the formulation by Nisbet and Gurney (Theoretical Population Biology, 23, 1983, 114) using delay-differential equations, which are solved using the r-packages deSolve or PBSddesolve. © 2015 British Ecological Society.

Loading Biomathematics and Statistics Scotland BioSS collaborators
Loading Biomathematics and Statistics Scotland BioSS collaborators