Use of posterior predictive assessments to evaluate model fit in multilevel logistic regression

Assessing the fit of a model is an important final step in any statistical analysis, but this is not straightforward when complex discrete response models are used. Cross validation and posterior predictions have been suggested as methods to aid model criticism. In this paper a comparison is made between four methods of model predictive assessment in the context of a three level logistic regression model for clinical mastitis in dairy cattle; cross validation, a prediction using the full posterior predictive distribution and two “mixed” predictive methods that incorporate higher level random effects simulated from the underlying model distribution. Cross validation is considered a gold standard method but is computationally intensive and thus a comparison is made between posterior predictive assessments and cross validation. The analyses revealed that mixed prediction methods produced results close to cross validation whilst the full posterior predictive assessment gave predictions that were over-optimistic (closer to the observed disease rates) compared with cross validation. A mixed prediction method that simulated random effects from both higher levels was best at identifying the outlying level two (farm-year) units of interest. It is concluded that this mixed prediction method, simulating random effects from both higher levels, is straightforward and may be of value in model criticism of multilevel logistic regression, a technique commonly used for animal health data with a hierarchical structure.

Anopheles darlingi polytene chromosomes: revised maps including newly described inversions and evidence for population structure in Manaus

Salivary gland polytene chromosomes of 4th instar Anopheles darlingi Root were examined from multiple locations in the Brazilian Amazon. Minor modifications were made to existing polytene photomaps. These included changes to the breakpoint positions of several previously described paracentric inversions and descriptions of four new paracentric inversions, two on the right arm of chromosome 3 and two on the left arm of chromosome 3 that were found in multiple locations. A total of 18 inversions on the X (n = 1) chromosome, chromosome 2 (n = 7) and 3 (n = 11) were scored for 83 individuals from Manaus, Macapá and Porto Velho municipalities. The frequency of 2Ra inversion karyotypes in Manaus shows significant deficiency of heterozygotes (p < 0.0009). No significant linkage disequilibrium was found between inversions on chromosome 2 and 3. We hypothesize that at least two sympatric subpopulations exist within the An. darlingi population at Manaus based on inversion frequencies.

Design of bicycling suitability maps for hilly cities

Post-print version. Pictures and tables separated from main text and presented at the end.

Studies of MHC class I antigen presentation & the origins of the immunopeptidome

La présentation d'antigène par les molécules d'histocompatibilité majeure de classe I (CMHI) permet au système immunitaire adaptatif de détecter et éliminer les agents pathogènes intracellulaires et des cellules anormales. La surveillance immunitaire est effectuée par les lymphocytes T CD8 qui interagissent avec le répertoire de peptides associés au CMHI présentés à la surface de toutes cellules nucléées. Les principaux gènes humains de CMHI, HLA-A et HLA-B, sont très polymorphes et par conséquent montrent des différences dans la présentation des antigènes. Nous avons étudié les différences qualitatives et quantitatives dans l'expression et la liaison peptidique de plusieurs allotypes HLA. Utilisant la technique de cytométrie de flux quantitative nous avons établi une hiérarchie d'expression pour les quatre HLA-A, B allotypes enquête. Nos résultats sont compatibles avec une corrélation inverse entre l'expression allotypique et la diversité des peptides bien que d'autres études soient nécessaires pour consolider cette hypothèse. Les origines mondiales du répertoire de peptides associés au CMHI restent une question centrale à la fois fondamentalement et dans la recherche de cibles immunothérapeutiques. Utilisant des techniques protéogénomiques, nous avons identifié et analysé 25,172 peptides CMHI isolées à partir des lymphocytes B de 18 personnes qui exprime collectivement 27 allotypes HLA-A,B. Alors que 58% des gènes ont été la source de 1-64 peptides CMHI par gène, 42% des gènes ne sont pas représentés dans l'immunopeptidome. Dans l'ensemble, l’immunopeptidome présenté par 27 allotypes HLA-A,B ne couvrent que 17% des séquences exomiques exprimées dans les cellules des sujets. Nous avons identifié plusieurs caractéristiques des transcrits et des protéines qui améliorent la production des peptides CMHI. Avec ces données, nous avons construit un modèle de régression logistique qui prédit avec une grande précision si un gène de notre ensemble de données ou à partir d'ensembles de données indépendants génèrerait des peptides CMHI. Nos résultats montrent la sélection préférentielle des peptides CMHI à partir d'un répertoire limité de produits de gènes avec des caractéristiques distinctes. L'idée que le système immunitaire peut surveiller des peptides CMHI couvrant seulement une fraction du génome codant des protéines a des implications profondes dans l'auto-immunité et l'immunologie du cancer.

AN ASSESSMENT OF FLOW REGIME MAPS AND A NUMERICAL HEAT TRANSFER CORRELATION FOR THE STRATIFIED/ANNULAR REGIME OF SHEAR-DRIVEN INTERNAL CONDENSING FLOWS

Several modern-day cooling applications require the incorporation of mini/micro-channel shear-driven flow condensers. There are several design challenges that need to be overcome in order to meet those requirements. The difficulty in developing effective design tools for shear-driven flow condensers is exacerbated due to the lack of a bridge between the physics-based modelling of condensing flows and the current, popular approach based on semi-empirical heat transfer correlations. One of the primary contributors of this disconnect is a lack of understanding caused by the fact that typical heat transfer correlations eliminate the dependence of the heat transfer coefficient on the method of cooling employed on the condenser surface when it may very well not be the case. This is in direct contrast to direct physics-based modeling approaches where the thermal boundary conditions have a direct and huge impact on the heat transfer coefficient values. Typical heat transfer correlations instead introduce vapor quality as one of the variables on which the value of the heat transfer coefficient depends. This study shows how, under certain conditions, a heat transfer correlation from direct physics-based modeling can be equivalent to typical engineering heat transfer correlations without making the same apriori assumptions. Another huge factor that raises doubts on the validity of the heat-transfer correlations is the opacity associated with the application of flow regime maps for internal condensing flows. It is well known that flow regimes influence heat transfer rates strongly. However, several heat transfer correlations ignore flow regimes entirely and present a single heat transfer correlation for all flow regimes. This is believed to be inaccurate since one would expect significant differences in the heat transfer correlations for different flow regimes. Several other studies present a heat transfer correlation for a particular flow regime - however, they ignore the method by which extents of the flow regime is established. This thesis provides a definitive answer (in the context of stratified/annular flows) to: (i) whether a heat transfer correlation can always be independent of the thermal boundary condition and represented as a function of vapor quality, and (ii) whether a heat transfer correlation can be independently obtained for a flow regime without knowing the flow regime boundary (even if the flow regime boundary is represented through a separate and independent correlation). To obtain the results required to arrive at an answer to these questions, this study uses two numerical simulation tools - the approximate but highly efficient Quasi-1D simulation tool and the exact but more expensive 2D Steady Simulation tool. Using these tools and the approximate values of flow regime transitions, a deeper understanding of the current state of knowledge in flow regime maps and heat transfer correlations in shear-driven internal condensing flows is obtained. The ideas presented here can be extended for other flow regimes of shear-driven flows as well. Analogous correlations can also be obtained for internal condensers in the gravity-driven and mixed-driven configuration.

Studies of MHC class I antigen presentation & the origins of the immunopeptidome

La présentation d'antigène par les molécules d'histocompatibilité majeure de classe I (CMHI) permet au système immunitaire adaptatif de détecter et éliminer les agents pathogènes intracellulaires et des cellules anormales. La surveillance immunitaire est effectuée par les lymphocytes T CD8 qui interagissent avec le répertoire de peptides associés au CMHI présentés à la surface de toutes cellules nucléées. Les principaux gènes humains de CMHI, HLA-A et HLA-B, sont très polymorphes et par conséquent montrent des différences dans la présentation des antigènes. Nous avons étudié les différences qualitatives et quantitatives dans l'expression et la liaison peptidique de plusieurs allotypes HLA. Utilisant la technique de cytométrie de flux quantitative nous avons établi une hiérarchie d'expression pour les quatre HLA-A, B allotypes enquête. Nos résultats sont compatibles avec une corrélation inverse entre l'expression allotypique et la diversité des peptides bien que d'autres études soient nécessaires pour consolider cette hypothèse. Les origines mondiales du répertoire de peptides associés au CMHI restent une question centrale à la fois fondamentalement et dans la recherche de cibles immunothérapeutiques. Utilisant des techniques protéogénomiques, nous avons identifié et analysé 25,172 peptides CMHI isolées à partir des lymphocytes B de 18 personnes qui exprime collectivement 27 allotypes HLA-A,B. Alors que 58% des gènes ont été la source de 1-64 peptides CMHI par gène, 42% des gènes ne sont pas représentés dans l'immunopeptidome. Dans l'ensemble, l’immunopeptidome présenté par 27 allotypes HLA-A,B ne couvrent que 17% des séquences exomiques exprimées dans les cellules des sujets. Nous avons identifié plusieurs caractéristiques des transcrits et des protéines qui améliorent la production des peptides CMHI. Avec ces données, nous avons construit un modèle de régression logistique qui prédit avec une grande précision si un gène de notre ensemble de données ou à partir d'ensembles de données indépendants génèrerait des peptides CMHI. Nos résultats montrent la sélection préférentielle des peptides CMHI à partir d'un répertoire limité de produits de gènes avec des caractéristiques distinctes. L'idée que le système immunitaire peut surveiller des peptides CMHI couvrant seulement une fraction du génome codant des protéines a des implications profondes dans l'auto-immunité et l'immunologie du cancer.

Atlas versus range maps: robustness of chorological relationships to distribution data types in European mammals

Aim Chorological relationships describe the patterns of distributional overlap among species. In addition to revealing biogeographical structure, the resulting clusters of species with similar geographical distributions can serve as natural units in conservation planning. Here, we assess the extent to which temporal, methodological and taxonomical differences in the source of species’ distribution data can affect the relationships that are found.

Analytics and Methods for Logistic System Design and Operations Management

Analytics is the technology working with the manipulation of data to produce information able to change the world we live every day. Analytics have been largely used within the last decade to cluster people’s behaviour to predict their preferences of items to buy, music to listen, movies to watch and even electoral preference. The most advanced companies succeded in controlling people’s behaviour using analytics. Despite the evidence of the super-power of analytics, they are rarely applied to the big data collected within supply chain systems (i.e. distribution network, storage systems and production plants). This PhD thesis explores the fourth research paradigm (i.e. the generation of knowledge from data) applied to supply chain system design and operations management. An ontology defining the entities and the metrics of supply chain systems is used to design data structures for data collection in supply chain systems. The consistency of this data is provided by mathematical demonstrations inspired by the factory physics theory. The availability, quantity and quality of the data within these data structures define different decision patterns. Ten decision patterns are identified, and validated on-field, to address ten different class of design and control problems in the field of supply chain systems research.

Adapting The Sniffin' Sticks Olfactory Test To Diagnose Parkinson's Disease In Estonia.

The aim of the study was to develop a culturally adapted translation of the 12-item smell identification test from Sniffin' Sticks (SS-12) for the Estonian population in order to help diagnose Parkinson's disease (PD). A standard translation of the SS-12 was created and 150 healthy Estonians were questioned about the smells used as response options in the test. Unfamiliar smells were replaced by culturally familiar options. The adapted SS-12 was applied to 70 controls in all age groups, and thereafter to 50 PD patients and 50 age- and sex-matched controls. 14 response options from 48 used in the SS-12 were replaced with familiar smells in an adapted version, in which the mean rate of correct response was 87% (range 73-99) compared to 83% with the literal translation (range 50-98). In PD patients, the average adapted SS-12 score (5.4/12) was significantly lower than in controls (average score 8.9/12), p < 0.0001. A multiple linear regression using the score in the SS-12 as the outcome measure showed that diagnosis and age independently influenced the result of the SS-12. A logistic regression using the SS-12 and age as covariates showed that the SS-12 (but not age) correctly classified 79.0% of subjects into the PD and control category, using a cut-off of <7 gave a sensitivity of 76% and specificity of 86% for the diagnosis of PD. The developed SS-12 cultural adaption is appropriate for testing olfaction in Estonia for the purpose of PD diagnosis.

Primary Relapse Site Pattern In Women With Triple-negative Breast Cancer.

Despite the remarkable improvements in breast cancer (BC) characterization, accurate prediction of BC clinical behavior is often still difficult to achieve. Some studies have investigated the association between the molecular subtype, namely the basal-like BC and the pattern of relapse, however only few investigated the association between relapse pattern and immunohistochemical defined triple-negative breast cancers (TNBCs). The aim of this study was to evaluate the pattern of relapse in patients with TNBC, namely the primary distant relapse site. One-hundred twenty nine (129) invasive breast carcinomas with follow-up information were classified according to the molecular subtype using immunohistochemistry for ER, PgR and Her2. The association between TNBC and distant relapse primary site was analyzed by logistic regression. Using multivariate logistic regression analysis patients with TNBC displayed only 0.09 (95% CI: 0.00-0.74; p=0.02) the odds of the non-TNBC patients of developing bone primary relapse. Regarding visceral and lymph-node relapse, no differences between in this cohort were found. Though classically regarded as aggressive tumors, TNBCs rarely development primary relapse in bone when compared to non-TNBC, a clinical relevant fact when investigating a metastasis of an occult or non-sampled primary BC.

What Is The Best Indication For Single-incision Ophira Mini Sling? Insights From A 2-year Follow-up International Multicentric Study.

The Ophira Mini Sling System involves anchoring a midurethral, low-tension tape to the obturator internus muscles bilaterally at the level of the tendinous arc. Success rates in different subsets of patients are still to be defined. This work aims to identify which factors influence the 2-year outcomes of this treatment. Analysis was based on data from a multicenter study. Endpoints for analysis included objective measurements: 1-h pad-weight (PWT), and cough stress test (CST), and questionnaires: International Consultation on Incontinence Questionnaire-Short Form (ICIQ-SF) and Urinary Distress Inventory (UDI)-6. A logistic regression analysis evaluated possible risk factors for failure. In all, 124 female patients with stress urinary incontinence (SUI) underwent treatment with the Ophira procedure. All patients completed 1 year of follow-up, and 95 complied with the 2-year evaluation. Longitudinal analysis showed no significant differences between results at 1 and 2 years. The 2-year overall objective results were 81 (85.3%) patients dry, six (6.3%) improved, and eight (8.4%) incontinent. A multivariate analysis revealed that previous anti-incontinence surgery was the only factor that significantly influenced surgical outcomes. Two years after treatment, women with previous failed surgeries had an odds ratio (OR) for treatment failure (based on PWT) of 4.0 [95% confidence interval (CI) 1.02-15.57). The Ophira procedure is an effective option for SUI treatment, with durable good results. Previous surgeries were identified as the only significant risk factor, though previously operated patients showed an acceptable success rate.

Iis--integrated Interactome System: A Web-based Platform For The Annotation, Analysis And Visualization Of Protein-metabolite-gene-drug Interactions By Integrating A Variety Of Data Sources And Tools.

High-throughput screening of physical, genetic and chemical-genetic interactions brings important perspectives in the Systems Biology field, as the analysis of these interactions provides new insights into protein/gene function, cellular metabolic variations and the validation of therapeutic targets and drug design. However, such analysis depends on a pipeline connecting different tools that can automatically integrate data from diverse sources and result in a more comprehensive dataset that can be properly interpreted. We describe here the Integrated Interactome System (IIS), an integrative platform with a web-based interface for the annotation, analysis and visualization of the interaction profiles of proteins/genes, metabolites and drugs of interest. IIS works in four connected modules: (i) Submission module, which receives raw data derived from Sanger sequencing (e.g. two-hybrid system); (ii) Search module, which enables the user to search for the processed reads to be assembled into contigs/singlets, or for lists of proteins/genes, metabolites and drugs of interest, and add them to the project; (iii) Annotation module, which assigns annotations from several databases for the contigs/singlets or lists of proteins/genes, generating tables with automatic annotation that can be manually curated; and (iv) Interactome module, which maps the contigs/singlets or the uploaded lists to entries in our integrated database, building networks that gather novel identified interactions, protein and metabolite expression/concentration levels, subcellular localization and computed topological metrics, GO biological processes and KEGG pathways enrichment. This module generates a XGMML file that can be imported into Cytoscape or be visualized directly on the web. We have developed IIS by the integration of diverse databases following the need of appropriate tools for a systematic analysis of physical, genetic and chemical-genetic interactions. IIS was validated with yeast two-hybrid, proteomics and metabolomics datasets, but it is also extendable to other datasets. IIS is freely available online at: http://www.lge.ibi.unicamp.br/lnbio/IIS/.

Feasibility And Reproducibility Of Diffusion-weighted Magnetic Resonance Imaging Of The Fetal Brain In Twin-twin Transfusion Syndrome.

The aim of this study is to test the feasibility and reproducibility of diffusion-weighted magnetic resonance imaging (DW-MRI) evaluations of the fetal brains in cases of twin-twin transfusion syndrome (TTTS). From May 2011 to June 2012, 24 patients with severe TTTS underwent MRI scans for evaluation of the fetal brains. Datasets were analyzed offline on axial DW images and apparent diffusion coefficient (ADC) maps by two radiologists. The subjective evaluation was described as the absence or presence of water diffusion restriction. The objective evaluation was performed by the placement of 20-mm(2) circular regions of interest on the DW image and ADC maps. Subjective interobserver agreement was assessed by the kappa correlation coefficient. Objective intraobserver and interobserver agreements were assessed by proportionate Bland-Altman tests. Seventy-four DW-MRI scans were performed. Sixty of them (81.1%) were considered to be of good quality. Agreement between the radiologists was 100% for the absence or presence of diffusion restriction of water. For both intraobserver and interobserver agreement of ADC measurements, proportionate Bland-Altman tests showed average percentage differences of less than 1.5% and 95% CI of less than 18% for all sites evaluated. Our data demonstrate that DW-MRI evaluation of the fetal brain in TTTS is feasible and reproducible.

Effects Of Therapeutic Approach On The Neonatal Evolution Of Very Low Birth Weight Infants With Patent Ductus Arteriosus.

To analyze the effects of treatment approach on the outcomes of newborns (birth weight [BW] < 1,000 g) with patent ductus arteriosus (PDA), from the Brazilian Neonatal Research Network (BNRN) on: death, bronchopulmonary dysplasia (BPD), severe intraventricular hemorrhage (IVH III/IV), retinopathy of prematurity requiring surgical (ROPsur), necrotizing enterocolitis requiring surgery (NECsur), and death/BPD. This was a multicentric, cohort study, retrospective data collection, including newborns (BW < 1000 g) with gestational age (GA) < 33 weeks and echocardiographic diagnosis of PDA, from 16 neonatal units of the BNRN from January 1, 2010 to Dec 31, 2011. Newborns who died or were transferred until the third day of life, and those with presence of congenital malformation or infection were excluded. Groups: G1 - conservative approach (without treatment), G2 - pharmacologic (indomethacin or ibuprofen), G3 - surgical ligation (independent of previous treatment). Factors analyzed: antenatal corticosteroid, cesarean section, BW, GA, 5 min. Apgar score < 4, male gender, Score for Neonatal Acute Physiology Perinatal Extension (SNAPPE II), respiratory distress syndrome (RDS), late sepsis (LS), mechanical ventilation (MV), surfactant (< 2 h of life), and time of MV. death, O2 dependence at 36 weeks (BPD36wks), IVH III/IV, ROPsur, NECsur, and death/BPD36wks. Student's t-test, chi-squared test, or Fisher's exact test; Odds ratio (95% CI); logistic binary regression and backward stepwise multiple regression. Software: MedCalc (Medical Calculator) software, version 12.1.4.0. p-values < 0.05 were considered statistically significant. 1,097 newborns were selected and 494 newborns were included: G1 - 187 (37.8%), G2 - 205 (41.5%), and G3 - 102 (20.6%). The highest mortality was observed in G1 (51.3%) and the lowest in G3 (14.7%). The highest frequencies of BPD36wks (70.6%) and ROPsur were observed in G3 (23.5%). The lowest occurrence of death/BPD36wks occurred in G2 (58.0%). Pharmacological (OR 0.29; 95% CI: 0.14-0.62) and conservative (OR 0.34; 95% CI: 0.14-0.79) treatments were protective for the outcome death/BPD36wks. The conservative approach of PDA was associated to high mortality, the surgical approach to the occurrence of BPD36wks and ROPsur, and the pharmacological treatment was protective for the outcome death/BPD36wks.