Biplots in biomedical research

Improved approximation and visualization of the correlation matrix

The graphical representation of the correlation matrix by means of different multivariate statistical methods is reviewed, a comparison of the different procedures is presented with the use of an example data set, and an improved representation with better fit is proposed. Principal component analysis is widely used for making pictures of correlation structure, though as shown a weighted alternating least squares approach that avoids the fitting of the diagonal of the correlation matrix outperforms both principal component analysis and principal factor analysis in approximating a correlation matrix. Weighted alternating least squares is a very strong competitor for principal component analysis, in particular if the correlation matrix is the focus of the study, because it improves the representation of the correlation matrix, often at the expense of only a minor percentage of explained variance for the original data matrix, if the latter is mapped onto the correlation biplot by regression. In this article, we propose to combine weighted alternating least squares with an additive adjustment of the correlation matrix, and this is seen to lead to further improved approximation of the correlation matrix.

“Automatic” interpretation of multiple correspondence analysis (MCA) results for nonexpert users, using R programming

Purpose

The purpose of this paper is to create an automatic interpretation of the results of the method of multiple correspondence analysis (MCA) for categorical variables, so that the nonexpert user can immediately and safely interpret the results, which concern, as the authors know, the categories of variables that strongly interact and determine the trends of the subject under investigation.

Design/methodology/approach

This study is a novel theoretical approach to interpreting the results of the MCA method. The classical interpretation of MCA results is based on three indicators: the projection (F) of the category points of the variables in factorial axes, the point contribution to axis creation (CTR) and the correlation (COR) of a point with an axis. The synthetic use of the aforementioned indicators is arduous, particularly for nonexpert users, and frequently results in misinterpretations. The current study has achieved a synthesis of the aforementioned indicators, so that the interpretation of the results is based on a new indicator, as correspondingly on an index, the well-known method principal component analysis (PCA) for continuous variables is based.

Findings

Two (2) concepts were proposed in the new theoretical approach. The interpretative axis corresponding to the classical factorial axis and the interpretative plane corresponding to the factorial plane that as it will be seen offer clear and safe interpretative results in MCA.

Research limitations/implications

It is obvious that in the development of the proposed automatic interpretation of the MCA results, the authors do not have in the interpretative axes the actual projections of the points as is the case in the original factorial axes, but this is not of interest to the simple user who is only interested in being able to distinguish the categories of variables that determine the interpretation of the most pronounced trends of the phenomenon being examined.

Practical implications

The results of this research can have positive implications for the dissemination of MCA as a method and its use as an integrated exploratory data analysis approach.

Originality/value

Interpreting the MCA results presents difficulties for the nonexpert user and sometimes lead to misinterpretations. The interpretative difficulty persists in the MCA's other interpretative proposals. The proposed method of interpreting the MCA results clearly and accurately allows for the interpretation of its results and thus contributes to the dissemination of the MCA as an integrated method of categorical data analysis and exploration.

Using HJ-Biplot and External Logistic Biplot as Machine Learning Methods for Corporate Social Responsibility Practices for Sustainable Development

In recent years, social responsibility has been revolutionizing sustainable development. After the development of new mathematical techniques, the improvement of computers’ processing capacity and the greater availability of possible explanatory variables, the analysis of these topics is moving towards the use of different machine learning techniques. However, within the field of machine learning, the use of Biplot techniques is little known for these analyses. For this reason, in this paper we explore the performance of two of the most popular techniques in multivariate statistics: External Logistic Biplot and the HJ-Biplot, to analyse the data structure in social responsibility studies. The results obtained from the sample of companies representing the Fortune Global 500 list indicate that the most frequently reported indicators are related to the social aspects are labour practices and decent work and society. On the contrary, the disclosure of indicators is less frequently related to human rights and product responsibility. Additionally, we have identified the countries and sectors with the highest CSR in social matters. We discovered that both machine learning algorithms are extremely competitive and practical to apply in CSR since they are simple to implement and work well with relatively big datasets.

Detection of Bulbar Involvement in Patients With Amyotrophic Lateral Sclerosis by Machine Learning Voice Analysis: Diagnostic Decision Support Development Study

Background

Bulbar involvement is a term used in amyotrophic lateral sclerosis (ALS) that refers to motor neuron impairment in the corticobulbar area of the brainstem, which produces a dysfunction of speech and swallowing. One of the earliest symptoms of bulbar involvement is voice deterioration characterized by grossly defective articulation; extremely slow, laborious speech; marked hypernasality; and severe harshness. Bulbar involvement requires well-timed and carefully coordinated interventions. Therefore, early detection is crucial to improving the quality of life and lengthening the life expectancy of patients with ALS who present with this dysfunction. Recent research efforts have focused on voice analysis to capture bulbar involvement.

Objective

The main objective of this paper was (1) to design a methodology for diagnosing bulbar involvement efficiently through the acoustic parameters of uttered vowels in Spanish, and (2) to demonstrate that the performance of the automated diagnosis of bulbar involvement is superior to human diagnosis.

Methods

The study focused on the extraction of features from the phonatory subsystem—jitter, shimmer, harmonics-to-noise ratio, and pitch—from the utterance of the five Spanish vowels. Then, we used various supervised classification algorithms, preceded by principal component analysis of the features obtained.

Results

To date, support vector machines have performed better (accuracy 95.8%) than the models analyzed in the related work. We also show how the model can improve human diagnosis, which can often misdiagnose bulbar involvement.

Conclusions

The results obtained are very encouraging and demonstrate the efficiency and applicability of the automated model presented in this paper. It may be an appropriate tool to help in the diagnosis of ALS by multidisciplinary clinical teams, in particular to improve the diagnosis of bulbar involvement.

Analysis of Madrid Metro Network: From Structural to HJ-Biplot Perspective

With the growth of cities, urban traffic has increased and traffic congestion has become a serious problem. Due to their characteristics, metro systems are one of the most used public transportation networks in big cities. So, optimization and planning of metro networks are challenges which governments must focus on. The objective of this study was to analyze Madrid metro network using graph theory. Through complex network theory, the main structural and topological properties of the network as well as robustness characteristics were obtained. Furthermore, to inspect these results, multivariate analysis techniques were employed, specifically HJ-Biplot. This analysis tool allowed us to explore relationships between centrality measures and to classify stations according to their centrality. Therefore, it is a multidisciplinary study that includes network analysis and multivariate analysis. The study found that closeness and eccentricity were strongly negatively correlated. In addition, the most central stations were those located in the city center, that is, there is a relationship between centrality and geographic location. In terms of robustness, a highly agglomerated community structure was found.

Where Do Adolescents Eat Less-Healthy Foods? Correspondence Analysis and Logistic Regression Results from the UK National Diet and Nutrition Survey

This study investigates the relationship between the consumption of foods and eating locations (home, school/work and others) in British adolescents, using data from the UK National Diet and Nutrition Survey Rolling Program (2008-2012 and 2013-2016). A cross-sectional analysis of 62,523 food diary entries from this nationally representative sample was carried out for foods contributing up to 80% total energy to the daily adolescent's diet. Correspondence analysis (CA) was used to generate food-location relationship hypotheses followed by logistic regression (LR) to quantify the evidence in terms of odds ratios and formally test those hypotheses. The less-healthy foods that emerged from CA were chips, soft drinks, chocolate and meat pies. Adjusted odds ratios (99% CI) for consuming specific foods at a location "other" than home (H) or school/work (S) in the 2008-2012 survey sample were: for soft drinks, 2.8 (2.1 to 3.8) vs. H and 2.0 (1.4 to 2.8) vs. S; for chips, 2.8 (2.2 to 3.7) vs. H and 3.4 (2.1 to 5.5) vs. S; for chocolates, 2.6 (1.9 to 3.5) vs. H and 1.9 (1.2 to 2.9) vs. S; and for meat pies, 2.7 (1.5 to 5.1) vs. H and 1.3 (0.5 to 3.1) vs. S. These trends were confirmed in the 2013-2016 survey sample. Interactions between location and BMI were not significant in either sample. In conclusion, public health policies to discourage less-healthy food choices in locations away from home and school/work are warranted for adolescents, irrespective of their BMI.

Platelets from HIV-infected individuals on antiretroviral drug therapy with poor CD4+ T cell recovery can harbor replication-competent HIV despite viral suppression

In addition to hemostasis, human platelets have several immune functions and interact with infectious pathogens including HIV in vitro. Here, we report that platelets from HIV-infected individuals on combined antiretroviral drug therapy (ART) with low blood CD4+ T cell counts (<350 cells/μl) contained replication-competent HIV despite viral suppression. In vitro, human platelets harboring HIV propagated the virus to macrophages, a process that could be prevented with the biologic abciximab, an anti–integrin αIIb/β3 Fab. Furthermore, in our cohort, 88% of HIV-infected individuals on ART with viral suppression and with platelets containing HIV were poor immunological responders with CD4+ T cell counts remaining below <350 cells/μl for more than one year. Our study suggests that platelets may be transient carriers of HIV and may provide an alternative pathway for HIV dissemination in HIV-infected individuals on ART with viral suppression and poor CD4+ T cell recovery.

Cultures under stress: A cross-national meta-analysis of cortisol responses to the Trier Social Stress Test and their association with anxiety-related value orientations and internalizing mental disorders

Acute cortisol changes are physiological indicators (i.e., biomarkers) of psychosocial stress that have been repeatedly assessed using standardized laboratory protocols like the Trier Social Stress Test (TSST). Despite this methodological standardization, however, cortisol stress responses vary considerably across different studies. Based on multilevel meta-analyses of 237 TSST studies (n = 8487 individuals), we demonstrate that ∼25% of this variability is actually attributable to systematic differences between countries. Particularly North American and European studies show a remarkable effect disparity (d = 0.45 vs. d = 0.73, respectively). Specifically, in-depth analyses suggested that the cultural orientation towards anxiety-related values in English-speaking countries (incl. the UK and Ireland) was associated with decreased cortisol stress responses. This finding was further supported by associations with the regionally varying prevalence of internalizing mental disorders (e.g., major depression or post-traumatic stress disorder) as reported by the WHO World Mental Health Surveys, a larger inequality of family incomes, larger population growth, larger expenses for health and the military, and lower tax rates. The size of these effects is comparable to the most important moderators of cortisol stress responses known to date, that is, male sex and higher age. Based on these observations, we argue that the cortisol stress response may reflect the persistent threats in the sociocultural environment an individual is accustomed too. Highly competitive cultures emphasize the individuals' responsibility for socioeconomic prosperity, but simultaneously increase the collective population stress and thus lower sustainable ontogeny and resilience towards unexpected environmental adversity.

Advantages of Using HJ-Biplot Analysis in Executive Functions Studies

Abstract Executive functions (EFs) are considered a multiple system of processing, associated with different components, such as inhibition, working memory, planning, among others. The study of EFs requires the assessment of all its components, having in mind the socio-demographic and cognitive characteristics of the target population. Nowadays, analysis of variance is used to achieve this goal; nevertheless, HJ-Biplot analysis overcome its limitations by allowing simultaneous examination of multiple data, such as those generated in EFs studies. This study evaluates possible differences in the EFs of 80 8-year-old Colombian children by their sex, socio-economic status and type of school they attend, to exemplify the advantages of using HJ-Biplot analysis in neuropsychological studies.

Childhood asthma clusters reveal neutrophil-predominant phenotype with distinct gene expression

BACKGROUND

Childhood asthma comprises different phenotypes with complex pathophysiology. Different asthma phenotypes evoke various clinical symptoms and vary in their responses to treatments.

METHODS

We applied k-means clustering algorithm of twelve objective laboratory tests among 351 asthmatic children enrolled in the Taiwanese Consortium of Childhood Asthma Study (TCCAS). We constructed gene expression profiles of peripheral blood mononuclear cells (PBMC) from children with different asthma phenotypes.

RESULTS

Five distinct phenotypes of childhood asthma were identified and can be characterized by either eosinophil-predominant or neutrophil-predominant inflammatory characteristics. In the gene expression profile analysis, significant differences were noted for neutrophil-predominant asthma, compared with samples from all the other asthma phenotypes. The vast majority of the differentially expressed genes in neutrophil-predominant asthma was associated with corticosteroid response. From an independent inhaled corticosteroid (ICS) response cohort, we also found neutrophils could be activated in this severe asthma phenotype and neutrophil-predominant asthma may be associated with corticosteroid nonresponsiveness.

CONCLUSION

Phenotype clustering of childhood asthma can be helpful to identify clinically relevant patients and reveal different inflammatory characteristics in asthmatic children. Neutrophil-predominant asthma is the most severe asthma phenotype with poor corticosteroid response. Gene expression profile of different asthma phenotypes not only improve our knowledge of childhood asthma, but also can guide asthma precision medicine.

Estimating correlations among cardiovascular patients' psychiatric and physical symptom indicators: The biplot in correspondence analysis approach

OBJECTIVES

We employed the correspondence analysis (CA) biplot to estimate correlations between gender-age levels of cardiovascular disease patients and their psychiatric and physical symptoms. Utilization of this correlation estimation can inform clinical practice by elucidating associations between certain psychiatric or physical symptoms and specific gender-age levels.

METHOD

The CA biplot utilized here was designed to visually inspect row-column category associations in a 2-dimensional plane and then to numerically estimate the category associations with correlations. To do so, we (a) estimated dimensions from row and column categories with CA; (b) verified statistical significance of dimensions with a permutation test; (c) projected row and column categories in a plan constructed with the first 2 dimensions that were statistically significant; (d) visually inspected category associations in the plane; and (e) numerically estimated category associations with correlations.

RESULTS

Consistent with the previous results, female cardiovascular disease patients were more likely to experience psychiatric symptoms than the male patients. However, when examining the results by gender and age, both female and male patients in their 50s and 60s tended to experience elevated rates of the psychiatric symptoms.

CONCLUSIONS

The CA biplot can be useful for isolating key clinical concerns among any medical populations.

Analysis of nutrient deficiencies affecting in vitro growth and development of Eucalyptus dunnii Maiden

Although basal medium optimization is a key factor in the success of tissue culture, its mineral composition is frequently disregarded when optimizing in vitro propagation protocols. A previous work on Eucalyptus dunnii micropropagation suggests that excessive callus formation and leaf chlorosis are related to specific nutritional conditions of the basal media. Recently, a novel basal medium based on the mineral nutrient analysis of E. dunnii young stump shoots was developed and successfully tested in plant regeneration and micropropagation of E. dunnii, avoiding all these issues. Considering this basal medium as an ideal growth condition, a mild deprivation of each macro and micronutrient and of the total organic fraction was imposed to E. dunnii in vitro cultures for 30 d. As a result, K, Mg, Mn, Cl, Zn, Mo, Ni or Co deprivation quantitatively affected growth and development of axillary shoots. Moreover, leaf chlorosis and the development of organogenic callus under Fe deficiency, and leaf drop along with shoot tip necrosis under N deficiency were observed. These symptoms suggest that nutrient content in E. dunnii tissues needs to be above 420.3 mg kg^-1 for Fe and 27.7 g kg^-1 for N to avoid the symptoms of leaf chlorosis and shoot tip necrosis. Additionally, the main role of Mn in quantitative responses and the antagonism between ions, especially for Mg/K and Mg/Zn, were denoted by the multivariate analysis. Overall, these results make a relevant contribution to the optimization of in vitro propagation of E. dunnii and other hard-to-propagate related species.

Genomic characterization of liver metastases from colorectal cancer patients

Metastatic dissemination is the most frequent cause of death of sporadic colorectal cancer (sCRC) patients. Genomic abnormalities which are potentially characteristic of such advanced stages of the disease are complex and so far, they have been poorly described and only partially understood. We evaluated the molecular heterogeneity of sCRC tumors based on simultaneous assessment of the overall GEP of both coding mRNA and non-coding RNA genes in primary sCRC tumor samples from 23 consecutive patients and their paired liver metastases. Liver metastases from the sCRC patients analyzed, systematically showed deregulated transcripts of those genes identified as also deregulated in their paired primary colorectal carcinomas. However, some transcripts were found to be specifically deregulated in liver metastases (vs. non-tumoral colorectal tissues) while expressed at normal levels in their primary tumors, reflecting either an increased genomic instability of metastatic cells or theiradaption to the liver microenvironment. Newly deregulated metastatic transcripts included overexpression of APOA1, HRG, UGT2B4, RBP4 and ADH4 mRNAS and the miR-3180-3p, miR-3197, miR-3178, miR-4793 and miR-4440 miRNAs, together with decreased expression of the IGKV1-39, IGKC, IGKV1-27, FABP4 and MYLK mRNAS and the miR-363, miR-1, miR-143, miR-27b and miR-28-5p miRNAs. Canonical pathways found to be specifically deregulated in liver metastatic samples included multiple genes related with intercellular adhesion and the metastatic processes (e.g., IGF1R, PIK3CA, PTEN and EGFR), endocytosis (e.g., the PDGFRA, SMAD2, ERBB3, PML and FGFR2), and the cell cycle (e.g., SMAD2, CCND2, E2F5 and MYC). Our results also highlighted the activation of genes associated with the TGFβ signaling pathway, -e.g. RHOA, SMAD2, SMAD4, SMAD5, SMAD6, BMPR1A, SMAD7 and MYC-, which thereby emerge as candidate genes to play an important role in CRC tumor metastasis.

RT-qPCR for Fecal Mature MicroRNA Quantification and Validation

By routinely and systematically being able to perform quantitative stem-loop reverse transcriptase (RT) followed by TaqMan® minor-groove binding (MGB) probe, real-time quantitative PCR analysis on exfoliated enriched colonocytes in stool, using human (Homo sapiens, hsa) micro(mi)RNAs to monitor changes of their expression at various stages of colorectal (CRC) progression, this method allows for the reliable and quantitative diagnostic screening of colon cancer (CC). Although the expression of some miRNA genes tested in tissue shows less variability in normal or cancerous patients than in stool, the noninvasive stool by itself is well suited for CC screening. An miRNA approach using stool promises to offer more sensitivity and specificity than currently used genomic, methylomic, or proteomic methods for CC screening.To present an application of employing miRNAs as diagnostic markers for CC screening, we carried out global microarray expression studies on stool colonocytes isolated by paramagnetic beads, using Affymetrix GeneChip miRNA 3.0 Array, to select a panel of miRNAs for subsequent focused semiquantitative PCR analysis studies. We then conducted a stem-loop RT-TaqMan® MGB probes, followed by a modified real-time qPCR expression study on 20 selected miRNAs for subsequent validation of the extracted immunocaptured total small RNA isolated from stool colonocytes. Results showed 12 miRNAs (miR-7, miR-17, miR-20a, miR-21, miR-92a, miR-96, miR-106a, miR-134, miR-183, miR-196a, miR-199a-3p, and miR214) to have an increased expression in stool of CC patients, and that later TNM stages exhibited more increased expressions than adenomas, while 8 miRNAs (miR-9, miR-29b, miR-127-5p, miR-138, miR-143, miR-146a, miR-222, and miR-938) showed decreased expressions in stool of CC patients, which becomes more pronounced as the cancer progresses from early to late TNM stages (0-IV).

Principal components analysis in clinical studies

In multivariate analysis, independent variables are usually correlated to each other which can introduce multicollinearity in the regression models. One approach to solve this problem is to apply principal components analysis (PCA) over these variables. This method uses orthogonal transformation to represent sets of potentially correlated variables with principal components (PC) that are linearly uncorrelated. PCs are ordered so that the first PC has the largest possible variance and only some components are selected to represent the correlated variables. As a result, the dimension of the variable space is reduced. This tutorial illustrates how to perform PCA in R environment, the example is a simulated dataset in which two PCs are responsible for the majority of the variance in the data. Furthermore, the visualization of PCA is highlighted.

The structure of obsessive compulsive symptoms and beliefs: A correspondence and biplot analysis

Research has suggested that obsessive-compulsive (OC) beliefs are differentially predictive of OC symptom dimensions. One additional way in which beliefs and symptoms may be related is by severity; that is, the role of beliefs may vary as a function of symptom severity. In order to evaluate this possibility, correspondence analysis with biplot was applied to evaluate the association between OC beliefs and OC symptom severity across three subsamples, individuals with obsessive-compulsive disorder (OCD; N=398), individuals with anxiety disorders (N=104), and undergraduate students (N=285). To do so, we generated five row categories of symptom severity and six columns based on the Obsessive Compulsive Beliefs Questionnaire (OBQ) for three subsamples. Unlike factor analyses of inter-variable correlations (or covariances), the CA-biplot paradigm calibrates simultaneously row and column information and estimates dimensional coordinates (analogous to factor loadings) separately for rows and columns. We used the first two dimensions from each subsample because they accounted for most variance (on average 89%) so as to construct a hypothetical plane with them. Then, we visually inspected associations among five severity categories (rows) and six OBQ subscales (columns) in the plane and also calculated their correlations. The visual configurations and numerical correlations were consistent across three subsamples, indicating that inflated responsibility was significantly associated with OC symptoms, but only for those with the least severe symptoms. Importance and control of thoughts were associated with OC symptoms across all levels of symptom severity. The implications of these findings for the cognitive model of OCD are considered.

The “need to know” and the lack of online transparency among political parties

Purpose

– This paper aims to analyse the level of informative transparency among Spanish political parties and political foundations, according to general and descriptive information (e.g. contact, ideas and values, electoral programmes, members, etc.) and economic, governance and financial information.

Design/methodology/approach

– The situation of the parties in each year is represented by a biplot, which is a graphical representation of a multivariate sample. The data for this analysis were obtained from the reports published by Fundación Compromiso y Transparencia (Foundation for Integrity and Transparency) for 2011 and 2012.

Findings

– This paper evidences the existence of serious problems of opacity, especially in relation to financial information (balance sheet, income statement, annual accounts and audit report) and information on the fulfilment of goals and programmes (management report and compliance report).

Originality/value

– This is the first attempt to analyse statistically the level of transparency of political parties and foundations, showing the need for a robust control system and for mechanisms to penalise conduct that limits citizens’ access to public information.

Predicting genome-wide DNA methylation using methylation marks, genomic position, and DNA regulatory elements

BACKGROUND

Recent assays for individual-specific genome-wide DNA methylation profiles have enabled epigenome-wide association studies to identify specific CpG sites associated with a phenotype. Computational prediction of CpG site-specific methylation levels is critical to enable genome-wide analyses, but current approaches tackle average methylation within a locus and are often limited to specific genomic regions.

RESULTS

We characterize genome-wide DNA methylation patterns, and show that correlation among CpG sites decays rapidly, making predictions solely based on neighboring sites challenging. We built a random forest classifier to predict methylation levels at CpG site resolution using features including neighboring CpG site methylation levels and genomic distance, co-localization with coding regions, CpG islands (CGIs), and regulatory elements from the ENCODE project. Our approach achieves 92% prediction accuracy of genome-wide methylation levels at single-CpG-site precision. The accuracy increases to 98% when restricted to CpG sites within CGIs and is robust across platform and cell-type heterogeneity. Our classifier outperforms other types of classifiers and identifies features that contribute to prediction accuracy: neighboring CpG site methylation, CGIs, co-localized DNase I hypersensitive sites, transcription factor binding sites, and histone modifications were found to be most predictive of methylation levels.

CONCLUSIONS

Our observations of DNA methylation patterns led us to develop a classifier to predict DNA methylation levels at CpG site resolution with high accuracy. Furthermore, our method identified genomic features that interact with DNA methylation, suggesting mechanisms involved in DNA methylation modification and regulation, and linking diverse epigenetic processes.

Is there an association between symptoms of anxiety and depression and quality of life in patients with chronic obstructive pulmonary disease?

BACKGROUND

In addition to symptoms, such as dyspnea and fatigue, patients with chronic obstructive pulmonary disease (COPD) also experience mood disturbances.

OBJECTIVE

To explore the relationships between health-related quality of life measures collected from patients with stable COPD and a commonly used measure of depression and anxiety.

METHODS

The present analysis was a retrospective study of patients with COPD enrolled in a pulmonary rehabilitation program. Hospital Anxiety and Depression Scale (HADS), Chronic Respiratory Disease Questionnaire (CRQ), Medical Research Council dyspnea scale and 6 min walk test data were collected. Statistical analyses were performed using Spearman's correlations, and categorical regression and categorical principal component analysis were interpreted using the biplot methodology.

RESULTS

HADS anxiety scores retrieved from 80 patients were grouped as 'no anxiety' (n=43 [54%]), 'probable anxiety' (n=21 [26%]) and 'presence of anxiety' (n=16 [20%]). HADS depression scores were similarly grouped. There was a moderate relationship between the anxiety subscale of the HADS and both the emotional function (r=-0.519; P<0.01) and mastery (r=-0.553; P<0.01) domains of the CRQ. Categorical regression showed that the CRQ-mastery domain explained 40% of the total variation in anxiety. A principal component analysis biplot showed that the highest distance between the groups was along the mastery domain, which separated patients without feelings of anxiety from those with anxiety. However, none of the CRQ domains were able to discriminate the three depression groups.

CONCLUSIONS

The CRQ-mastery domain may identify symptoms of anxiety in patients with COPD; however, the relationship is not strong enough to use the CRQ-mastery domain as a surrogate measure. None of the CRQ domains were able to discriminate the three depression groups (no depression, probable and presence); therefore, specific, validated tools to identify symptoms of depression should be used.

Fixed effects modelling for provider mortality outcomes: Analysis of the Australia and New Zealand Intensive Care Society (ANZICS) Adult Patient Data-base

BACKGROUND

Risk adjusted mortality for intensive care units (ICU) is usually estimated via logistic regression. Random effects (RE) or hierarchical models have been advocated to estimate provider risk-adjusted mortality on the basis that standard estimators increase false outlier classification. The utility of fixed effects (FE) estimators (separate ICU-specific intercepts) has not been fully explored.

METHODS

Using a cohort from the Australian and New Zealand Intensive Care Society Adult Patient Database, 2009-2010, the model fit of different logistic estimators (FE, random-intercept and random-coefficient) was characterised: Bayesian Information Criterion (BIC; lower values better), receiver-operator characteristic curve area (AUC) and Hosmer-Lemeshow (H-L) statistic. ICU standardised hospital mortality ratios (SMR) and 95%CI were compared between models. ICU site performance (FE), relative to the grand observation-weighted mean (GO-WM) on odds ratio (OR), risk ratio (RR) and probability scales were assessed using model-based average marginal effects (AME).

RESULTS

The data set consisted of 145355 patients in 128 ICUs, years 2009 (47.5%) & 2010 (52.5%), with mean(SD) age 60.9(18.8) years, 56% male and ICU and hospital mortalities of 7.0% and 10.9% respectively. The FE model had a BIC = 64058, AUC = 0.90 and an H-L statistic P-value = 0.22. The best-fitting random-intercept model had a BIC = 64457, AUC = 0.90 and H-L statistic P-value = 0.32 and random-coefficient model, BIC = 64556, AUC = 0.90 and H-L statistic P-value = 0.28. Across ICUs and over years no outliers (SMR 95% CI excluding null-value = 1) were identified and no model difference in SMR spread or 95%CI span was demonstrated. Using AME (OR and RR scale), ICU site-specific estimates diverged from the GO-WM, and the effect spread decreased over calendar years. On the probability scale, a majority of ICUs demonstrated calendar year decrease, but in the for-profit sector, this trend was reversed.

CONCLUSIONS

The FE estimator had model advantage compared with conventional RE models. Using AME, between and over-year ICU site-effects were easily characterised.

miRNA as markers for the diagnostic screening of colon cancer

Early screening for colon cancer (CC) allows for early stage diagnosis of the malignancy and potentially reduces disease mortality as the cancer is most likely curable at its earliest stages. Early detection would be desirable if accurate, practical and cost-effective diagnostic measures for this cancer were available. Mortality and morbidity from CC represent a major health problem involving a malignant disease that is theoretically preventable through screening. Current screening methods (e.g., the convenient and inexpensive immunological fecal occult blood test, FOBTi, obtained from patients' medical records) either lack sensitivity and require dietary restriction, which impedes compliance and use; are costly (e.g., colonoscopy), which decreases compliance; or could result in mortality. In comparison with the FOBT test, a non-invasive sensitive screen for which there is no requirement for dietary restriction would be a more convenient test. Colorectal cancer is the only cancer for which colonoscopy is recommended as a screening method. Although colonoscopy is a reliable screening tool, the invasive nature, abdominal pain, potential complications and high cost have hampered the application of this procedure worldwide. A screening approach using the stable miRNA molecules, which are relatively non-degradable when extracted from non-invasive stool and semi-invasive blood samples by commercially available kits and manipulated thereafter, would be preferable to a transcriptomic mRNA-, a mutation DNA-, an epigenetic- or a proteomic-based test. The approach uses reverse transcriptase, modified real-time quantitative PCR. Although exosomal RNA would be missed, using a restricted extraction of total RNA from stool or blood, a parallel test could also be carried out on RNA obtained from stool or plasma samples, and appropriate corrections for exsosomal loss can be made for accurate and quantitative test result. Eventually, a chip can be developed to facilitate diagnosis, as has been done for the quantification of genetically modified organisms in foods. The gold standard to which the molecular miRNA test is compared is colonoscopy, which can be obtained from patients' medical records. If performance criteria are met, as detailed herein, a miRNA test in human stool or blood samples based on high-throughput automated technologies and quantitative expression measurements commonly used in the diagnostic clinical laboratory should be advanced to the clinical setting, which will make a significant impact on CC prevention.

Analysis of the airway microbiota of healthy individuals and patients with chronic obstructive pulmonary disease by T-RFLP and clone sequencing

Chronic obstructive pulmonary disease (COPD) is a progressive, inflammatory lung disease that affects a large number of patients and has significant impact. One hallmark of the disease is the presence of bacteria in the lower airways. Objective: The aim of this study was to analyze the detailed structure of microbial communities found in the lungs of healthy individuals and patients with COPD. Nine COPD patients as compared and 9 healthy individuals underwent flexible bronchoscopy and BAL was performed. Bacterial nucleic acids were subjected to terminal restriction fragment (TRF) length polymorphism and clone library analysis. Overall, we identified 326 T-RFLP band, 159 in patients and 167 in healthy controls. The results of the TRF analysis correlated partly with the data obtained from clone sequencing. Although the results of the sequencing showed high diversity, the genera Prevotella, Sphingomonas, Pseudomonas, Acinetobacter, Fusobacterium, Megasphaera, Veillonella, Staphylococcus, and Streptococcus constituted the major part of the core microbiome found in both groups. A TRF band possibly representing Pseudomonas sp. monoinfection was associated with a reduction of the microbial diversity. Non-cultural methods reveal the complexity of the pulmonary microbiome in healthy individuals and in patients with COPD. Alterations of the microbiome in pulmonary diseases are correlated with disease.

Assessing the applicability of trial evidence to a target sample in the presence of heterogeneity of treatment effect

PURPOSE

To propose methods for the quantitative assessment of the applicability of evidence from a trial to a target sample using individual data.

METHODS

Demonstration was with a trial of drug therapy to prevent mortality and an accompanying registry of people with heart failure. Principal components analysis with biplots did not identify measurement discrepancies. Multiple imputation with chained equations addressed missing predictor values. A proportional hazards model with interaction term, including graphical interpretation and a multivariate interaction test, identified heterogeneity of treatment effect. An interval of homogeneity of treatment effect was the interval of the baseline risk of outcome in which no two treatment effects were statistically significantly different. Absolute risk reduction for individuals was estimated for both benefit and harm outcomes and presented in a bivariate treatment effects scatterplot.

RESULTS

Overall, the trial evidence applied to most of the registry according to overlapping distributions of estimated benefit and harm. However, 52% of trial and 33% of registry participants were estimated to have net benefit, and 14% of trial and 36% of registry participants were estimated to have strong net harmful treatment effect, that is, the individual estimate of harm was more than twice the estimate of benefit.

CONCLUSIONS

The proposed methods provide quantitative assessment of the applicability of trial evidence to a target sample. They combine the strengths of different study designs, namely, unbiased effects estimation from trials and representation in observational studies, while addressing the practical challenges of combining information, namely, measurement discrepancies and missing data.

Global quantitative indices reflecting provider process-of-care: data-base derivation

Background

Controversy has attended the relationship between risk-adjusted mortality and process-of-care. There would be advantage in the establishment, at the data-base level, of global quantitative indices subsuming the diversity of process-of-care.

Methods

A retrospective, cohort study of patients identified in the Australian and New Zealand Intensive Care Society Adult Patient Database, 1993-2003, at the level of geographic and ICU-level descriptors (n = 35), for both hospital survivors and non-survivors. Process-of-care indices were established by analysis of: (i) the smoothed time-hazard curve of individual patient discharge and determined by pharmaco-kinetic methods as area under the hazard-curve (AUC), reflecting the integrated experience of the discharge process, and time-to-peak-hazard (TMAX, in days), reflecting the time to maximum rate of hospital discharge; and (ii) individual patient ability to optimize output (as length-of-stay) for recorded data-base physiological inputs; estimated as a technical production-efficiency (TE, scaled [0,(maximum)1]), via the econometric technique of stochastic frontier analysis. For each descriptor, multivariate correlation-relationships between indices and summed mortality probability were determined.

Results

The data-set consisted of 223129 patients from 99 ICUs with mean (SD) age and APACHE III score of 59.2(18.9) years and 52.7(30.6) respectively; 41.7% were female and 45.7% were mechanically ventilated within the first 24 hours post-admission. For survivors, AUC was maximal in rural and for-profit ICUs, whereas TMAX (≥ 7.8 days) and TE (≥ 0.74) were maximal in tertiary-ICUs. For non-survivors, AUC was maximal in tertiary-ICUs, but TMAX (≥ 4.2 days) and TE (≥ 0.69) were maximal in for-profit ICUs. Across descriptors, significant differences in indices were demonstrated (analysis-of-variance, P ≤ 0.0001). Total explained variance, for survivors (0.89) and non-survivors (0.89), was maximized by combinations of indices demonstrating a low correlation with mortality probability.

Conclusions

Global indices reflecting process of care may be formally established at the level of national patient data-bases. These indices appear orthogonal to mortality outcome.

Bioinformatic characterizations and prediction of K+ and Na+ ion channels effector toxins

BACKGROUND

K+ and Na+ channel toxins constitute a large set of polypeptides, which interact with their ion channel targets. These polypeptides are classified in two different structural groups. Recently a new structural group called birtoxin-like appeared to contain both types of toxins has been described. We hypothesized that peptides of this group may contain two conserved structural motifs in K+ and/or Na+ channels scorpion toxins, allowing these birtoxin-like peptides to be active on K+ and/or Na+ channels.

RESULTS

Four multilevel motifs, overrepresented and specific to each group of K+ and/or Na+ ion channel toxins have been identified, using GIBBS and MEME and based on a training dataset of 79 sequences judged as representative of K+ and Na+ toxins.Unexpectedly birtoxin-like peptides appeared to present a new structural motif distinct from those present in K+ and Na+ channels Toxins. This result, supported by previous experimental data, suggests that birtoxin-like peptides may exert their activity on different sites than those targeted by classic K+ or Na+ toxins.Searching, the nr database with these newly identified motifs using MAST, retrieved several sequences (116 with e-value < 1) from various scorpion species (test dataset). The filtering process left 30 new and highly likely ion channel effectors.Phylogenetic analysis was used to classify the newly found sequences. Alternatively, classification tree analysis, using CART algorithm adjusted with the training dataset, using the motifs and their 2D structure as explanatory variables, provided a model for prediction of the activity of the new sequences.

CONCLUSION

The phylogenetic results were in perfect agreement with those obtained by the CART algorithm.Our results may be used as criteria for a new classification of scorpion toxins based on functional motifs.

Genome adaptation to chemical stress: clues from comparative transcriptomics in Saccharomyces cerevisiae and Candida glabrata

Comparative transcriptomics of Saccharomyces cerevisiae and Candida glabrata revealed a remarkable conservation of response to drug-induced stress, despite underlying differences in the regulatory networks.

Background

Recent technical and methodological advances have placed microbial models at the forefront of evolutionary and environmental genomics. To better understand the logic of genetic network evolution, we combined comparative transcriptomics, a differential clustering algorithm and promoter analyses in a study of the evolution of transcriptional networks responding to an antifungal agent in two yeast species: the free-living model organism Saccharomyces cerevisiae and the human pathogen Candida glabrata.

Results

We found that although the gene expression patterns characterizing the response to drugs were remarkably conserved between the two species, part of the underlying regulatory networks differed. In particular, the roles of the oxidative stress response transcription factors ScYap1p (in S. cerevisiae) and Cgap1p (in C. glabrata) had diverged. The sets of genes whose benomyl response depends on these factors are significantly different. Also, the DNA motifs targeted by ScYap1p and Cgap1p are differently represented in the promoters of these genes, suggesting that the DNA binding properties of the two proteins are slightly different. Experimental assays of ScYap1p and Cgap1p activities in vivo were in accordance with this last observation.

Conclusions

Based on these results and recently published data, we suggest that the robustness of environmental stress responses among related species contrasts with the rapid evolution of regulatory sequences, and depends on both the coevolution of transcription factor binding properties and the versatility of regulatory associations within transcriptional networks.

Intrastriatal inhibition of aromatic amino acid decarboxylase prevents l-DOPA-induced dyskinesia: a bilateral reverse in vivo microdialysis study in 6-hydroxydopamine lesioned rats

l-3,4-dihydroxyphenylalanine (l-DOPA)-induced dyskinesia consists of involuntary choreiform and dystonic movements. Here we report whether intrastriatal l-DOPA itself is able to trigger dyskinetic behavior and which role the neurotransmitter dopamine (DA) and its metabolites play. Intrastriatal l-DOPA as well as DA administration at the 6-hydroxydopamine (6-OHDA) lesioned side led to a significant appearance of dyskinetic behavior, whereas DA metabolites were ineffective. Intrastriatal inhibition of the enzyme aromatic amino acid decarboxylase (AADC) by benserazide prevented the appearance of l-DOPA-induced dyskinetic movements at the lesioned side. Principle component analysis of DA and DA metabolite levels with dyskinesia scores after l-DOPA/benserazide (6/15 mg/kg) administration indicated a significant correlation only for DA, whereas DA metabolites did not show any significant correlation with the occurrence of dyskinetic behavior. We conclude that intrastriatal l-DOPA itself is not able to induce dyskinetic movements, whereas the increase of intrastriatal DA levels is instrumental for l-DOPA- and DA-induced dyskinetic behavior.

Virulence of Bacillus cereus: A multivariate analysis

Biological activity and presence of DNA sequences related to virulence genes were studied in 21 strains of the Bacillus cereus group. The activity of spent culture supernatants and the effect of infection by vegetative bacterial cells were assessed on cultured human enterocytes (Caco-2 cells). The effect of extracellular factors on the detachment, necrosis and mitochondrial dehydrogenase activity of cultured human enterocytes was studied. Hemolytic activity on rabbit red blood cells was also evaluated and the effect of direct procaryotic-eucaryotic interactions was assessed in infection assays with vegetative bacterial cells. Concerning virulence genes, presence of the DNA sequences corresponding to the genes entS, entFM, nhe (A, B and C), sph, hbl (A, B, C and D), piplC and bceT was assessed by PCR. Ribopatterns were determined by an automated riboprinting analysis after digestion of the DNA with EcoRI. Principal component analysis and biplots were used to address the relationship between variables. Results showed a wide range of biological activities: decrease in mitochondrial dehydrogenase activity, necrosis, cell detachment and hemolytic activity. These effects were strain-dependent. Concerning the occurrence of the DNA sequences tested, different patterns were found. In addition, ribotyping showed that strains under study grouped into two main clusters. One of these clusters includes all the strains that were positive for all the DNA sequences tested. Positive and negative correlations between variables under study were evidenced. Interestingly, high detaching strains were positively correlated with the presence of the sequences entS, nheC and sph. Within gene complexes, high correlation was found between sequences of the hbl complex. In contrast, sequences of the nhe complex were not correlated. Some strains clustered together in the biplots. These strains were positive for all the DNA sequences tested and they were able to detach enterocytes upon infection. Our results highlight the multifactorial character of the virulence of the B. cereus group and show the correlation between ribopatterns, occurrence of toxin genes and biological activity of the strains under study.

Dental transposition as a disorder of genetic origin

A sample of 85 dental transpositions in 75 subjects (27 male, 48 female; mean age at diagnosis 12.25 years) involving both maxillary and mandibular arches was analysed using dental panoramic radiographs and clinical records. Transposition affected the maxillary dentition (76 per cent) more frequently than the mandibular dentition (24 per cent). Unilateral transposition accounted for 88 per cent of cases, with the maxilla being involved more commonly than the mandible. Overall, the most common transposition involved the maxillary canine and first premolar (58 per cent). Considering the jaws in isolation, the canine and first premolar were the most commonly affected teeth in the maxilla (83 per cent) whilst in the mandible, the canine and lateral incisor teeth were most commonly transposed (73 per cent). No significant difference in symmetrical distribution of the unilateral transposition sample occurred. There was evidence of associated hypodontia in 41 per cent of the sample; however, if third molars were excluded, this figure decreased to 25 per cent. Peg-shaped maxillary lateral incisors were judged to be present in 27 per cent of subjects, whilst 41 per cent had retained primary teeth; all of these, except one, were primary canines. Overall, the majority of the sample (76 per cent) demonstrated at least one of the dental anomalies under investigation. Multivariate analysis showed associations between unilateral transposition, gender, and the presence of peg-shaped maxillary lateral incisors; whilst bilateral transposition was more closely associated with gender and the presence of retained primary teeth. There was a poor association between both unilateral and bilateral transposition and hypodontia. Together, these results suggest a mutifactorial aetiology to this disorder, with both genetic and environmental factors playing an important role.

Calibration of Multivariate Scatter plots for Exploratory Analysis of Relations Within and Between Sets of Variables in Genomic Research

The scatter plot is a well known and easily applicable graphical tool to explore relationships between two quantitative variables. For the exploration of relations between multiple variables, generalisations of the scatter plot are useful. We present an overview of multivariate scatter plots focussing on the following situations. Firstly, we look at a scatter plot for portraying relations between quantitative variables within one data matrix. Secondly, we discuss a similar plot for the case of qualitative variables. Thirdly, we describe scatter plots for the relationships between two sets of variables where we focus on correlations. Finally, we treat plots of the relationships between multiple response and predictor variables, focussing on the matrix of regression coefficients. We will present both known and new results, where an important original contribution concerns a procedure for the inclusion of scales for the variables in multivariate scatter plots. We provide software for drawing such scales. We illustrate the construction and interpretation of the plots by means of examples on data collected in a genomic research program on taste in tomato.

Early expression of yeast genes affected by chemical stress

The variety of environmental stresses is probably the major challenge imposed on transcription activators and the transcriptional machinery. To precisely describe the very early genomic response developed by yeast to accommodate a chemical stress, we performed time course analyses of the modifications of the yeast gene expression program which immediately follows the addition of the antimitotic drug benomyl. Similar analyses were conducted with different isogenic yeast strains in which genes coding for relevant transcription factors were deleted and coupled with efficient bioinformatics tools. Yap1 and Pdr1, two well-known key mediators of stress tolerance, appeared to be responsible for the very rapid establishment of a transient transcriptional response encompassing 119 genes. Yap1, which plays a predominant role in this response, binds, in vivo, promoters of genes which are not automatically up-regulated. We proposed that Yap1 nuclear localization and DNA binding are necessary but not sufficient to elicit the specificity of the chemical stress response.

Comparative analysis of antibiotic and antimicrobial biocide susceptibility data in clinical isolates of methicillin-sensitive Staphylococcus aureus, methicillin-resistant Staphylococcus aureus and Pseudomonas aeruginosa between 1989 and 2000*

AIMS

To analyse population minimum inhibitory concentrations (MICs) data from clinical strains of Staphylococcus aureus and Pseudomonas aeruginosa for changes over a 10-year period and to look for correlations between the antimicrobials tested.

METHODS AND RESULTS

Data from the MIC study of 256 clinical isolates of Staph. aureus [169 methicillin-sensitive Staph. aureus (MSSA), 87 methicillin-resistant Staph. aureus (MRSA)] and 111 clinical isolates of Ps. aeruginosa against eight antimicrobial biocides and several clinically relevant antibiotics was analysed using anova, Spearman-Rho correlation and principal component analysis. Comparisons suggest that alterations in the mean susceptibility of Staph. aureus to antimicrobial biocides have occurred between 1989 and 2000, but that these changes were mirrored in MSSA and MRSA suggests that methicillin resistance has little to do with these changes. Between 1989 and 2000 a sub-population of MRSA has acquired a higher resistance to biocides, but this has not altered the antibiotic susceptibility of that group. In both Staph. aureus and Ps. aeruginosa several correlations (both positive and negative) between antibiotics and antimicrobial biocides were found.

CONCLUSIONS

From the analyses of these clinical isolates it is very difficult to support a hypothesis that increased biocide resistance is a cause of increased antibiotic resistance either in Staph. aureus or in Ps. aeruginosa.

SIGNIFICANCE AND IMPACT OF THE STUDY

The observation of negative correlations between antibiotics and biocides may be a useful reason for the continued use of biocides promoting hygiene in the hospital environment.

The effect of hereditary disorders on tooth components: a radiographic morphometric study of two syndromes

OBJECTIVE

The purpose of this study was to compare tooth components (enamel and dentin) in Familial Dysautonomia (FD) and Down syndrome (DS) in order to assess the extent to which each was affected.

DESIGN

The design was cross-sectional. The sample consisted of 20 FD patients and 45 DS patients. The control group comprised 250 healthy subjects. Mesio-distal crown width (CW), enamel and dentin thickness and pulp chamber dimensions were measured on standardized bitewing radiographs of mandibular second primary and first permanent molars. Statistical analyses were performed between groups using SAS programs.

RESULTS

CW was reduced in both hereditary disorders. In the DS group enamel height (EH) and dentin thickness were reduced. In FD enamel thickness in the primary and permanent molars as well as dentin height (DH) in permanent molars was increased.

CONCLUSIONS

In both syndromes the reduction in CW suggests reduced proliferation during tooth germ formation. However, the differences in enamel and dentin thickness suggest that ameloblasts and odontoblasts were affected differently in the later phases of cell function. In FD cell function is stimulated resulting in thicker enamel and dentin. In DS cell function is reduced resulting in thin enamel and dentin.

Added value of three-way methods for the analysis of mortality trends illustrated with worldwide female cancer mortality (1968-1985)

Trends in mortality rates are usually presented per tumour site or per country without an overall analysis of the complete data encompassing all three aspects (tumour sites, countries, trends). This paper presents a methodology for such an overall analysis using three-way methods applied to a data set on female mortality rates for 17 tumour sites of 43 countries for the years 1968-1985. Multivariate techniques like biplots and three-mode principal component analysis within an overall three-way analysis-of-variance framework were used. We confirmed the known patterns of comparatively high mortality for women due to cancer of the bladder, intestines, pancreas, rectum, breast, ovary, skin and leukaemia and the relatively low mortality rates for liver cancer in Western and Northern Europe, the USA, Australia and New Zealand. Also, the reverse pattern was observed for Middle and Southern Europe, Hong Kong, Singapore, and in Japan, and in some but not all Latin American countries. The relatively mortality due to cancer was high in the lungs, mouth, larynx and oesophagus in the British Isles, but was much less in other European countries. Mortality due to cancer of the thyroid, uterus, gall bladder and stomach was high in Middle European countries, as was the case in Japan, Chile and Costa Rica. Rates were low for Southern European countries, North America, Australia and New Zealand. Specific deviating patterns in the data were the more rapidly decreasing mortality rates for stomach cancer in Chile and Japan and the more rapidly increasing mortality rates for lung cancer in the USA, Scotland and Denmark. In conclusion, using three-way methods, it was feasible to analyse the cancer mortality data in their entirety. This enabled the simultaneous comparison of trends in relative mortality rates between all countries due to all tumour sites, as well as the identification of specific deviating trends for specific tumour sites in specific countries.

A new way for surgical education--development and evaluation of a computer-based training module

Computer-based training (CBT) programs teach the material of a specific field and at the same time offer various ways of objectively assessing the knowledge gained. The interactive use of multi-media components such as text, graphics, animation, sound, digital slide shows, and videos as well as quizzes can theoretically facilitate the learning process. The aim of this study was the development and evaluation of a CBT-program by surgeons for student training. Using SuperCard, a teaching module for Distal Radius Fracture (DRF) was developed, which contains detailed clinical information. Video clips and vivid animations combine theoretical knowledge with practical experience. Fourth-year medical students (n = 103) were tested after using the module for 90 min. Other students (n = 47) served as the control group. In a 90 min lecture, DRF was discussed. CBT gained in all evaluated criteria (distinctiveness, detailed description, presentation of materials, structure, motivation for learning, time saved learning and memory retention) 15-20% better scores than the lecture. Although 82% of the students stated that their experience with computers was limited or insufficient, 100% found the use of CBT systems useful in student teaching. Most of them suggested the use of such programs as a method of exam preparation/self study (90%) or as a supplement to a lecture (40%). Based on these evaluations, CBT modules are an appropriate future teaching and learning system that is well accepted. In conclusion, the results of this study show that CBT-programs could be a valuable supplement to medical education. In addition, further development of CBT-programs and their use as information systems for surgical residency programs at universities can be suggested.