Modeling strategies in longitudinal data analysis: Covariate, variance function and correlation structure selection

A modeling paradigm is proposed for covariate, variance and working correlation structure selection for longitudinal data analysis. Appropriate selection of covariates is pertinent to correct variance modeling and selecting the appropriate covariates and variance function is vital to correlation structure selection. This leads to a stepwise model selection procedure that deploys a combination of different model selection criteria. Although these criteria find a common theoretical root based on approximating the Kullback-Leibler distance, they are designed to address different aspects of model selection and have different merits and limitations. For example, the extended quasi-likelihood information criterion (EQIC) with a covariance penalty performs well for covariate selection even when the working variance function is misspecified, but EQIC contains little information on correlation structures. The proposed model selection strategies are outlined and a Monte Carlo assessment of their finite sample properties is reported. Two longitudinal studies are used for illustration.

Structural basis for the function of anti-idiotypic antibody in immune memory

We had earlier proposed a hypothesis to explain the mechanism of perpetuation of immunological memory based on the operation of idiotypic network in the complete absence of antigen. Experimental evidences were provided for memory maintenance through anti-idiotypic antibody (Ab(2)) carrying the internal image of the antigen. In the present work, we describe a structural basis for such memory perpetuation by molecular modeling and structural analysis studies. A three-dimensional model of Ab(2) was generated and the structure of the antigenic site on the hemagglutinin protein H of Rinderpest virus was modeled using the structural template of hemagglutinin protein of Measles virus. Our results show that a large portion of heavy chain containing the CDR regions of Ab(2) resembles the domain of the hemagglutinin housing the epitope regions. The similarity demonstrates that an internal image of the H antigen is formed in Ab(2), which provides a structural basis for functional mimicry demonstrated earlier. This work brings out the importance of the structural similarity between a domain of hemagglutinin protein to that of its corresponding Ab(2). It provides evidence that Ab(2) is indeed capable of functioning as surrogate antigen and provides support to earlier proposed relay hypothesis which has provided a mechanism for the maintenance of immunological memory.

Robust estimation using the Huber function with a data-dependent tuning constant

Robust estimation often relies on a dispersion function that is more slowly varying at large values than the square function. However, the choice of tuning constant in dispersion functions may impact the estimation efficiency to a great extent. For a given family of dispersion functions such as the Huber family, we suggest obtaining the "best" tuning constant from the data so that the asymptotic efficiency is maximized. This data-driven approach can automatically adjust the value of the tuning constant to provide the necessary resistance against outliers. Simulation studies show that substantial efficiency can be gained by this data-dependent approach compared with the traditional approach in which the tuning constant is fixed. We briefly illustrate the proposed method using two datasets.

Effects of variance-function misspecification in analysis of longitudinal data

The approach of generalized estimating equations (GEE) is based on the framework of generalized linear models but allows for specification of a working matrix for modeling within-subject correlations. The variance is often assumed to be a known function of the mean. This article investigates the impacts of misspecifying the variance function on estimators of the mean parameters for quantitative responses. Our numerical studies indicate that (1) correct specification of the variance function can improve the estimation efficiency even if the correlation structure is misspecified; (2) misspecification of the variance function impacts much more on estimators for within-cluster covariates than for cluster-level covariates; and (3) if the variance function is misspecified, correct choice of the correlation structure may not necessarily improve estimation efficiency. We illustrate impacts of different variance functions using a real data set from cow growth.

Kallikrein-related peptidases in prostate cancer: From molecular function to clinical application

Prostate cancer is a leading contributor to male cancer-related deaths worldwide. Kallikrein-related peptidases (KLKs) are serine proteases that exhibit deregulated expression in prostate cancer, with KLK3, or prostate specific antigen (PSA), being the widely-employed clinical biomarker for prostate cancer. Other KLKs, such as KLK2, show promise as prostate cancer biomarkers and, additionally, their altered expression has been utilised for the design of KLK-targeted therapies. There is also a large body of in vitro and in vivo evidence supporting their role in cancer-related processes. Here, we review the literature on studies to date investigating the potential of other KLKs, in addition to PSA, as biomarkers and in therapeutic options, as well as their current known functional roles in cancer progression. Increased knowledge of these KLK-mediated functions, including degradation of the extracellular matrix, local invasion, cancer cell proliferation, interactions with fibroblasts, angiogenesis, migration, bone metastasis and tumour growth in vivo, may help define new roles as prognostic biomarkers and novel therapeutic targets for this cancer.

Effects of the sensorimotor training volume on sensorimotor function in patients following lower limb arthroplasty

Background Sensorimotor function is degraded in patients after lower limb arthroplasty. Sensorimotor training is thought to improve sensorimotor skills, however, the optimal training stimulus with regard to volume, frequency, duration, and intensity is still unknown. The aim of this study, therefore, was to firstly quantify the progression of sensorimotor function after total hip (THA) or knee (TKA) arthroplasty and, as second step, to evaluate effects of different sensorimotor training volumes. Methods 58 in-patients during their rehabilitation after THA or TKA participated in this prospective cohort study. Sensorimotor function was assessed using a test battery including measures of stabilization capacity, static balance, proprioception, and gait, along with a self-reported pain and function. All participants were randomly assigned to one of three intervention groups performing sensorimotor training two, four, or six times per week. Outcome measures were taken at three instances, at baseline (pre), after 1.5 weeks (mid) and at the conclusion of the 3 week program (post). Results All measurements showed significant improvements over time, with the exception of proprioception and static balance during quiet bipedal stance which showed no significant main effects for time or intervention. There was no significant effect of sensorimotor training volume on any of the outcome measures. Conclusion We were able to quantify improvements in measures of dynamic, but not static, sensorimotor function during the initial three weeks of rehabilitation following TKA/THA. Although sensorimotor improvements were independent of the training volume applied in the current study, long-term effects of sensorimotor training volume need to be investigated to optimize training stimulus recommendations.

Coordinated Verb Pairs in Texts

This dissertation discusses the relation between lexis, grammar and textual organisation. The major premise adopted here is that grammatical structures are motivated both by semantic potential of words and by text-pragmatic demands. In other words, it is argued that grammatical structures form the interface between lexis and textual organisation, and that linguistic analysis should not concentrate on analysing grammatical structures in isolation, independent of context. From this point of view, grammatical structures are said to be 'well-formed' only in relation to the context they occur in. This study is based on a corpus of three million words of recent Finnish fiction from which all the occurrences of the coordinated verb pairs ([V ja V] -pairs]) containing one of the intransitive motion verbs 'lähteä' (to go), 'mennä' (to go), 'päästä' (to get into), 'nousta' (to get up), and 'laskea' (to go down), were extracted. This set of verbs was established using methods described in earlier work by Lagus & Airola (2001, and 2005). The quantitative analysis of the [V ja V] -pairs was used to carry out a qualitative analysis of individual texts. In analysing the texts, an analogy was made between musical and textual structure. The results show among others that individual verbs specialise in different functions when occurring in coordinated verb pairs. One aspect was that those verb pairs including the verb 'nousta' tend to function as markers of textual boundaries and thus reflect the organisation of narrative substance. The verb 'mennä' has weakened literal meanings, but strengthened modal meanings when occurring in [V ja V] -pairs, and, in many cases, the verb 'lähteä' in [V ja V] -pairs function as an aspectual marker rather than a pure verb of motion. That there is a gradient from the concrete sense of motion into more differentiated senses of a verb in [V ja V] -pairs alongside the structure-creating potential of the [V ja V] -pairs themselves suggest an ongoing grammaticalisation process of the patterns discussed.

Molecular tools for studying puberty in the grey mullet, Mugil cephalus: cloning of cDNAs that regulate reproductive function

The cDNAs coding for the brain GnRHs (AY373449-51), pituitary GH, SL and PRL, and liver IGFs (AY427954-5) were isolated. Partial cDNA sequences of the brain (Cyp19b) and gonadal (Cyp19a) aromatases have also been obtained. These tools would be utilized to study the endocrine regulation of puberty in the grey mullet.

Tree species diversity and ecosystem function: Can tropical multi-species plantations generate greater productivity?

Results from the humid tropics of Australia demonstrate that diverse plantations can achieve greater productivity than monocultures. We found that increases in both the observed species number and the effective species richness were significantly related to increased levels of productivity as measured by stand basal area or mean individual tree basal area. Four of five plantation species were more productive in mixtures with other species than in monocultures, offering on average, a 55% increase in mean tree basal area. A general linear model suggests that species richness had a significant effect on mean individual tree basal area when environmental variables were included in the model. As monoculture plantations are currently the preferred reforestation method throughout the tropics these results suggest that significant productivity and ecological gains could be made if multi-species plantations are more broadly pursued.

Serum immunoreactive pancreatic lipase and cationic trypsinogen for the assessment of exocrine pancreatic function in older patients with cystic fibrosis

Indirect and qualitative tests of pancreatic function are commonly used to screen patients with cystic fibrosis for pancreatic insufficiency. In an attempt to develop a more quantitative assessment, we compared the usefulness of measuring serum pancreatic lipase using a newly developed enzyme-linked immunosorbent immunoassay with that of cationic trypsinogen using a radioimmunoassay in the assessment of exocrine pancreatic function in patients with cystic fibrosis. Previously, we have shown neither lipase nor trypsinogen to be of use in assessing pancreatic function prior to 5 years of age because the majority of patients with cystic fibrosis in early infancy have elevated serum levels regardless of pancreatic function. Therefore, we studied 77 patients with cystic fibrosis older than 5 years of age, 41 with steatorrhea and 36 without steatorrhea. In addition, 28 of 77 patients consented to undergo a quantitative pancreatic stimulation test. There was a significant difference between the steatorrheic and nonsteatorrheic patients with the steatorrheic group having lower lipase and trypsinogen values than the nonsteatorrheic group (P < .001). Sensitivities and specificities in detecting steatorrhea were 95% and 86%, respectively, for lipase and 93% and 92%, respectively, for trypsinogen. No correlations were found between the serum levels of lipase and trypsinogen and their respective duodenal concentrations because of abnormally high serum levels of both enzymes found in some nonsteatorrheic patients. We conclude from this study that both serum lipase and trypsinogen levels accurately detect steatorrhea in patients with cystic fibrosis who are older than 5 years but are imprecise indicators of specific pancreatic exocrine function above the level needed for normal fat absorption.

Isolation and characterization of novel microsatellite markers and their application for diversity assessment in cultivated groundnut (Arachis hypogaea)

Background: Cultivated peanut or groundnut (Arachis hypogaea L.) is the fourth most important oilseed crop in the world, grown mainly in tropical, subtropical and warm temperate climates. Due to its origin through a single and recent polyploidization event, followed by successive selection during breeding efforts, cultivated groundnut has a limited genetic background. In such species, microsatellite or simple sequence repeat (SSR) markers are very informative and useful for breeding applications. The low level of polymorphism in cultivated germplasm, however, warrants a need of larger number of polymorphic microsatellite markers for cultivated groundnut. Results: A microsatellite- enriched library was constructed from the genotype TMV2. Sequencing of 720 putative SSR-positive clones from a total of 3,072 provided 490 SSRs. 71.2% of these SSRs were perfect type, 13.1% were imperfect and 15.7% were compound. Among these SSRs, the GT/CA repeat motifs were the most common (37.6%) followed by GA/CT repeat motifs (25.9%). The primer pairs could be designed for a total of 170 SSRs and were optimized initially on two genotypes. 104 (61.2%) primer pairs yielded scorable amplicon and 46 (44.2%) primers showed polymorphism among 32 cultivated groundnut genotypes. The polymorphic SSR markers detected 2 to 5 alleles with an average of 2.44 per locus. The polymorphic information content (PIC) value for these markers varied from 0.12 to 0.75 with an average of 0.46. Based on 112 alleles obtained by 46 markers, a phenogram was constructed to understand the relationships among the 32 genotypes. Majority of the genotypes representing subspecies hypogaea were grouped together in one cluster, while the genotypes belonging to subspecies fastigiata were grouped mainly under two clusters. Conclusion. Newly developed set of 104 markers extends the repertoire of SSR markers for cultivated groundnut. These markers showed a good level of PIC value in cultivated germplasm and therefore would be very useful for germplasm analysis, linkage mapping, diversity studies and phylogenetic relationships in cultivated groundnut as well as related Arachis species.

Impaired physical function associated with childhood obesity: How should we intervene?

- Background This study examined relationships between adiposity, physical functioning and physical activity. - Methods Obese (N=107) and healthy-weight (N=132) children aged 10-13 years underwent assessments of percent body fat (%BF, dual energy X-ray absorptiometry), knee extensor strength (KE, isokinetic dynamometry), cardiorespiratory fitness (CRF, peak oxygen uptake by cycle ergometry), physical health-related quality of life (HRQOL), worst pain intensity and walking capacity [six-minute walk (6MWT)]. Structural equation modelling was used to assess relationships between variables. - Results Moderate relationships were observed between %BF and 6MWT, KE strength corrected for mass and CRF relative to mass (r -.36 to -.69, P≤.007). Weak relationships were found between: %BF and physical HRQOL (r -.27, P=.008); CRF relative to mass and physical HRQOL (r -.24, P=.003); physical activity and 6MWT (r .17, P=.004). Squared multiple correlations showed that 29.6% variance in physical HRQOL was explained by %BF, pain and CRF relative to mass, while 28% variance in 6MWT was explained by %BF and physical activity. - Conclusions It appears that children with a higher body fat percentage have poorer KE strength, CRF and overall physical functioning. Reducing percent fat appears to be the best target to improve functioning. However, a combined approach to intervention, targeting reductions in body fat percentage, pain and improvements in physical activity and CRF may assist physical functioning.

An assessment of the genetic relationship between sweet and grain sorghums, within Sorghum bicolor ssp. bicolor (L.) Moench, using AFLP markers

Compared to grain sorghums, sweet sorghums typically have lower grain yield and thick, tall stalks which accumulate high levels of sugar (sucrose, fructose and glucose). Unlike commercial grain sorghum (S. bicolor ssp. bicolor) cultivars, which are usually F1 hybrids, commercial sweet sorghums were selected as wild accessions or have undergone limited plant breeding. Although all sweet sorghums are classified within S. bicolor ssp. bicolor, their genetic relationship with grain sorghums is yet to be investigated. Ninety-five genotypes, including 31 sweet sorghums and 64 grain sorghums, representing all five races within the subspecies bicolor, were screened with 277 polymorphic amplified fragment length polymorphism (AFLP) markers. Cluster analysis separated older sweet sorghum accessions (collected in mid 1800s) from those developed and released during the early to mid 1900s. These groups were emphasised in a principle component analysis of the results such that sweet sorghum lines were largely distinguished from the others, particularly by a group of markers located on sorghum chromosomes SBI-08 and SBI-10. Other studies have shown that QTL and ESTs for sugar-related traits, as well as for height and anthesis, map to SBI-10. Although the clusters obtained did not group clearly on the basis of racial classification, the sweet sorghum lines often cluster with grain sorghums of similar racial origin thus suggesting that sweet sorghum is of polyphyletic origin within S. bicolor ssp. bicolor.

Genus i östra Nyland : från dialektutjämning till dialektmarkör

Gender in eastern Nyland – from dialect levelling to identity marking The study of dialect leveling in eastern Nyland focuses on variation and change in the Swedish dialects of Nyland (Fi. Uusimaa) on the south coast of Finland. During the last century the grammatical gender system of the dialects in the area has been reduced from a three-gender system to a two-gender system (cf. Corbett 1991). The present study is based on five linguistic variables in the gender system: the anaphoric pronouns (han, hon, den) when used for inanimates; the neuter pronouns he(t) and de(t) – when used anaphorically or as expletives; and three different types of morphological postposed definite articles. For all these variables, both dialect variants and standard variants are used in the dialects. Within the study of processes of variation and change, the work focuses on the mechanisms of leveling, simplification and reallocation; cf. Trudgill (1986) and Hinskens, Auer Kerswill (2005). With regard to the reductions of the gender system, the possibility that some of these variables might have turned into becoming dialect markers (Labov 1972) in the modern varieties of eastern Nyland is given special attention. The primary data consist of tape recordings with 25 informants done in the 1960s and 1970s. The informants were born in 1881–1913. In addition, recent changes were investigated in detail in tape recordings from 2005–2008 with 15 informants, who were born in the period 1927–1947 or 1976–1988. The study combines quantitative and qualitative methods in the systematic analysis of the data. Theoretically and methodologically the study relies on methods and results from variation studies and socio-dialectology, as well as on methods and results from traditional dialectology; cf. Ahlbäck (1946) and the dictionary of Swedish dialects, Ordbok över Finlands svenska folkmål, (1976–). The results show that there are different strategies among the informants in their use of the features studied. In the modern varieties of the dialects, most of the informants use only two genders, uter and neuter. Of the variables, the masculine pronoun for inanimates, the traditional neuter pronoun he(t) and some variants of the traditional definite articles have received a new function as dialect markers in my data. These changes first affect the gender distinctions, and the function of marking gender is lost; gradually the features then get new functions as dialect markers through processes of dialect leveling and reallocation. These processes are connected to changes taking place in the communities in eastern Nyland because of urbanization. When the dialect speakers experience that the traditional values of both the dialects and the culture are threatened, they begin to mark their dialectal identity by using dialect markers in their speech.

Prolonged hyperinsulinemia affects metabolic signal transduction markers in a tissue specific manner

Insulin dysregulation is common in horses although the mechanisms of metabolic dysfunction are poorly understood. We hypothesized that insulin signaling in striated (cardiac and skeletal) muscle and lamellae may be mediated through different receptors as a result of receptor content, and that transcriptional regulation of downstream signal transduction and glucose transport may also differ between tissues sites during hyperinsulinemia. Archived samples from horses treated with a prolonged insulin infusion or a balanced electrolyte solution were used. All treated horses developed marked hyperinsulinemia and clinical laminitis. Protein expression was compared across tissues for the insulin receptor and insulin-like growth factor 1 receptor (IGF-1R) by immunoblotting. Gene expression of metabolic insulin-signaling markers (insulin receptor substrate 1, Akt2, and glycogen synthase kinase 3 beta [GSK-3β]) and glucose transport (basal glucose transporter 1 and insulin-sensitive glucose transporter 4) was evaluated using real-time reverse transcription polymerase chain reaction. Lamellar tissue contained significantly more IGF-1R protein than skeletal muscle, indicating the potential significance of IGF-1R signaling for this tissue. Gene expression of the selected markers of insulin signaling and glucose transport in skeletal muscle and lamellar tissues was unaffected by prolonged hyperinsulinemia. In contrast, the significant upregulation of Akt2, GSK-3β, GLUT1, and GLUT4 gene expression in cardiac tissue suggested that the prolonged hyperinsulinemia induced an increase in insulin sensitivity and a transcriptional activation of glucose transport. Responses to insulin are tissue-specific, and extrapolation of data across tissue sites is inappropriate.