Large-scale clustering through functional embedding

We present a new framework for large-scale data clustering. The main idea is to modify functional dimensionality reduction techniques to directly optimize over discrete labels using stochastic gradient descent. Compared to methods like spectral clustering our approach solves a single optimization problem, rather than an ad-hoc two-stage optimization approach, does not require a matrix inversion, can easily encode prior knowledge in the set of implementable functions, and does not have an ?out-of-sample? problem. Experimental results on both artificial and real-world datasets show the usefulness of our approach.

Unlicensed and off-label drug use in a Swiss paediatric university hospital.

BACKGROUND: Many medicines used in newborns, infants, children and adolescents are not licensed ("unlicensed") or are prescribed outside the terms of the marketing authorization ("off-label"). Several studies have shown that this is a common practice in various healthcare settings in the USA, Europe and Australia, but data are scarce in Switzerland. OBJECTIVES: The aim of our prospective study was to determine the proportion of unlicensed or off-label prescriptions in paediatric patients. METHODS: This pilot study was conducted prospectively over a six month period in the department of paediatrics of a university hospital. RESULTS: Sixty patients aged from three days to 14 years were included in the study. A total of 483 prescriptions were written for the patients. More than half of all prescriptions (247; 51%) followed the terms of the marketing authorization. 114 (24%) were unlicensed and 122 (25%) off-label. All patients received at least one unlicensed or offlabel medicine. CONCLUSION: The use of unlicensed or off-label medicines to treat children was found to be common. Co-operation between the pharmaceutical industry, national regulatory authorities, clinical researchers, healthcare professionals and parents is required in order to ensure that children do not remain "therapeutic orphans".

A Note on Interviewer Performance Measures in centralised CATI Surveys

Interviewer performance with respect to convincing sample members to participate in surveys is an important dimension of survey quality. However, unlike in CAPI surveys where each sample case 'belongs' to one interviewer, there are hardly any good measures of interview performance for centralised CATI surveys, where even single contacts are assigned to interviewers at random. If more than one interviewer works one sample case, it is not clear how to attribute success or failure to the interviewers involved. In this article, we propose two correlated methods to measure interviewer contact performance in centralised CATI surveys. Their modelling must take complex multilevel clustering effects, which need not be hierarchical, into account. Results are consistent with findings from CAPI data modelling, and we find that when comparing effects with a direct ('naive') measure of interviewer contact results, interviewer random effects are largely underestimated using the naive measure.

Influence of label information on dark chocolate acceptability

The aim of the present work was to study how the information on product labels influences consumer expectations and their acceptance and purchase intention of dark chocolate. Six samples of dark choco- late, varying in brand (premium and store brand) and in type of product (regular dark chocolate, single cocoa origin dark chocolate and high percentage of cocoa dark chocolate), were evaluated by 109 con- sumers who scored their liking and purchase intention under three conditions: blind (only tasting the products), expected (observing product label information) and informed (tasting the products together with provision of the label information). In the expected condition, consumer liking was mainly affected by the brand. In the blind condition, differences in liking were due to the type of product; the samples with a high percentage of cocoa were those less preferred by consumers. Under the informed condition, liking of dark chocolates varied depending on both brand and type of product. Premium brand chocolates generated high consumer expectations of chocolate acceptability, which were fulfilled by the sensory characteristics of the products. Store brand chocolates created lower expectations, but when they were tasted they were as acceptable as premium chocolates. Claims of a high percentage of cocoa and single cocoa origin on labels did not generate higher expectations than regular dark chocolates.

Full activation of the T cell receptor requires both clustering and conformational changes at CD3.

T cell receptor (TCR-CD3) triggering involves both receptor clustering and conformational changes at the cytoplasmic tails of the CD3 subunits. The mechanism by which TCRalphabeta ligand binding confers conformational changes to CD3 is unknown. By using well-defined ligands, we showed that induction of the conformational change requires both multivalent engagement and the mobility restriction of the TCR-CD3 imposed by the plasma membrane. The conformational change is elicited by cooperative rearrangements of two TCR-CD3 complexes and does not require accompanying changes in the structure of the TCRalphabeta ectodomains. This conformational change at CD3 reverts upon ligand dissociation and is required for T cell activation. Thus, our permissive geometry model provides a molecular mechanism that rationalizes how the information of ligand binding to TCRalphabeta is transmitted to the CD3 subunits and to the intracellular signaling machinery.

Hierarchical modeling gave plausible estimates of associations between metabolic syndrome and components of antiretroviral therapy.

OBJECTIVE: Hierarchical modeling has been proposed as a solution to the multiple exposure problem. We estimate associations between metabolic syndrome and different components of antiretroviral therapy using both conventional and hierarchical models. STUDY DESIGN AND SETTING: We use discrete time survival analysis to estimate the association between metabolic syndrome and cumulative exposure to 16 antiretrovirals from four drug classes. We fit a hierarchical model where the drug class provides a prior model of the association between metabolic syndrome and exposure to each antiretroviral. RESULTS: One thousand two hundred and eighteen patients were followed for a median of 27 months, with 242 cases of metabolic syndrome (20%) at a rate of 7.5 cases per 100 patient years. Metabolic syndrome was more likely to develop in patients exposed to stavudine, but was less likely to develop in those exposed to atazanavir. The estimate for exposure to atazanavir increased from hazard ratio of 0.06 per 6 months' use in the conventional model to 0.37 in the hierarchical model (or from 0.57 to 0.81 when using spline-based covariate adjustment). CONCLUSION: These results are consistent with trials that show the disadvantage of stavudine and advantage of atazanavir relative to other drugs in their respective classes. The hierarchical model gave more plausible results than the equivalent conventional model.

Multicentric carpotarsal osteolysis is caused by mutations clustering in the amino-terminal transcriptional activation domain of MAFB.

Multicentric carpotarsal osteolysis (MCTO) is a rare skeletal dysplasia characterized by aggressive osteolysis, particularly affecting the carpal and tarsal bones, and is frequently associated with progressive renal failure. Using exome capture and next-generation sequencing in five unrelated simplex cases of MCTO, we identified previously unreported missense mutations clustering within a 51 base pair region of the single exon of MAFB, validated by Sanger sequencing. A further six unrelated simplex cases with MCTO were also heterozygous for previously unreported mutations within this same region, as were affected members of two families with autosomal-dominant MCTO. MAFB encodes a transcription factor that negatively regulates RANKL-induced osteoclastogenesis and is essential for normal renal development. Identification of this gene paves the way for development of novel therapeutic approaches for this crippling disease and provides insight into normal bone and kidney development.

Clustering of HCV coinfections on HIV phylogeny indicates domestic and sexual transmission of HCV.

BACKGROUND: HCV coinfection remains a major cause of morbidity and mortality among HIV-infected individuals and its incidence has increased dramatically in HIV-infected men who have sex with men(MSM). METHODS: Hepatitis C virus (HCV) coinfection in the Swiss HIV Cohort Study(SHCS) was studied by combining clinical data with HIV-1 pol-sequences from the SHCS Drug Resistance Database(DRDB). We inferred maximum-likelihood phylogenetic trees, determined Swiss HIV-transmission pairs as monophyletic patient pairs, and then considered the distribution of HCV on those pairs. RESULTS: Among the 9748 patients in the SHCS-DRDB with known HCV status, 2768(28%) were HCV-positive. Focusing on subtype B(7644 patients), we identified 1555 potential HIV-1 transmission pairs. There, we found that, even after controlling for transmission group, calendar year, age and sex, the odds for an HCV coinfection were increased by an odds ratio (OR) of 3.2 [95% confidence interval (CI) 2.2, 4.7) if a patient clustered with another HCV-positive case. This strong association persisted if transmission groups of intravenous drug users (IDUs), MSMs and heterosexuals (HETs) were considered separately(in all cases OR>2). Finally we found that HCV incidence was increased by a hazard ratio of 2.1 (1.1, 3.8) for individuals paired with an HCV-positive partner. CONCLUSIONS: Patients whose HIV virus is closely related to the HIV virus of HIV/HCV-coinfected patients have a higher risk for carrying or acquiring HCV themselves. This indicates the occurrence of domestic and sexual HCV transmission and allows the identification of patients with a high HCV-infection risk.

Spatial Scale of Clustering of Motor Vehicle Crash Types and Appropriate Countermeasures, 2009

This project analyzes the characteristics and spatial distributions of motor vehicle crash types in order to evaluate the degree and scale of their spatial clustering. Crashes occur as the result of a variety of vehicle, roadway, and human factors and thus vary in their clustering behavior. Clustering can occur at a variety of scales, from the intersection level, to the corridor level, to the area level. Conversely, other crash types are less linked to geographic factors and are more spatially “random.” The degree and scale of clustering have implications for the use of strategies to promote transportation safety. In this project, Iowa's crash database, geographic information systems, and recent advances in spatial statistics methodologies and software tools were used to analyze the degree and spatial scale of clustering for several crash types within the counties of the Iowa Northland Regional Council of Governments. A statistical measure called the K function was used to analyze the clustering behavior of crashes. Several methodological issues, related to the application of this spatial statistical technique in the context of motor vehicle crashes on a road network, were identified and addressed. These methods facilitated the identification of crash clusters at appropriate scales of analysis for each crash type. This clustering information is useful for improving transportation safety through focused countermeasures directly linked to crash causes and the spatial extent of identified problem locations, as well as through the identification of less location-based crash types better suited to non-spatial countermeasures. The results of the K function analysis point to the usefulness of the procedure in identifying the degree and scale at which crashes cluster, or do not cluster, relative to each other. Moreover, for many individual crash types, different patterns and processes and potentially different countermeasures appeared at different scales of analysis. This finding highlights the importance of scale considerations in problem identification and countermeasure formulation.

Aggregation invariance in general clustering approaches

General clustering deals with weighted objects and fuzzy memberships. We investigate the group- or object-aggregation-invariance properties possessed by the relevant functionals (effective number of groups or objects, centroids, dispersion, mutual object-group information, etc.). The classical squared Euclidean case can be generalized to non-Euclidean distances, as well as to non-linear transformations of the memberships, yielding the c-means clustering algorithm as well as two presumably new procedures, the convex and pairwise convex clustering. Cluster stability and aggregation-invariance of the optimal memberships associated to the various clustering schemes are examined as well.

Available evidence and outcome of off-label use of rituximab in clinical practice

Purpose: To analyze the therapeutic indications for off-label use of rituximab, the available evidence for its use, the outcomes, and the cost. Methods: This was a retrospective analysis of patients treated with rituximab for off-label indications from January 2007 to December 2009 in two tertiary hospitals. Information on patient characteristics, medical conditions, and therapeutic responses was collected from medical records. Available evidence for the efficacy of rituximab in each condition was reviewed, and the cost of treatment was calculated. Results: A total of 101 cases of off-label rituximab use were analyzed. The median age of the patients involved was 53 [interquartile range (IQR) 37.568.0] years; 55.4 % were women. The indications for prescribing rituximab were primarily hematological diseases (46 %), systemic connective tissue disorders (27 %), and kidney diseases (20 %). Available evidence supporting rituximab treatment for these indications mainly came from individual cohort studies (53.5 % of cases) and case series (25.7 %). The short-term outcome (median 3 months, IQR 24 months) was a complete response in 38 % of cases and partial response in 32.6 %. The highest short-term responses were observed for systemic lupus erythematosus and membranous glomerulonephritis, and the lowest was for neuromyelitis optica, idiopathic thrombocytopenic purpura, and miscellaneous indications. Some response was maintained in long-term follow-up (median 23 months IQR 1230months) in 69.2%of patients showing a short-term response. Median cost per patient was 5,187.5 (IQR 5,187.57,781.3). Conclusions: In our study, off-label rituximab was mainly used for the treatment of hematological, kidney, and systemic connective tissue disorders, and the response among our patient cohort was variable depending on the specific disease. The level of evidence supporting the use of rituximab for these indications was low and the cost was very high. We conclude that more clinical trials on the off-label use of rituximab are needed, although these may be difficult to conduct in some rare diseases. Data from observational studies may provide useful information to assist prescribing in clinical practice.

Intensified chemotherapy with ACVBP plus rituximab versus standard CHOP plus rituximab for the treatment of diffuse large B-cell lymphoma (LNH03-2B): an open-label randomised phase 3 trial.

BACKGROUND: The outcome of diffuse large B-cell lymphoma has been substantially improved by the addition of the anti-CD20 monoclonal antibody rituximab to chemotherapy regimens. We aimed to assess, in patients aged 18-59 years, the potential survival benefit provided by a dose-intensive immunochemotherapy regimen plus rituximab compared with standard treatment plus rituximab. METHODS: We did an open-label randomised trial comparing dose-intensive rituximab, doxorubicin, cyclophosphamide, vindesine, bleomycin, and prednisone (R-ACVBP) with subsequent consolidation versus standard rituximab, doxorubicin, cyclophosphamide, vincristine, and prednisone (R-CHOP). Random assignment was done with a computer-assisted randomisation-allocation sequence with a block size of four. Patients were aged 18-59 years with untreated diffuse large B-cell lymphoma and an age-adjusted international prognostic index equal to 1. Our primary endpoint was event-free survival. Our analyses of efficacy and safety were of the intention-to-treat population. This study is registered with ClinicalTrials.gov, number NCT00140595. FINDINGS: One patient withdrew consent before treatment and 54 did not complete treatment. After a median follow-up of 44 months, our 3-year estimate of event-free survival was 81% (95% CI 75-86) in the R-ACVBP group and 67% (59-73) in the R-CHOP group (hazard ratio [HR] 0·56, 95% CI 0·38-0·83; p=0·0035). 3-year estimates of progression-free survival (87% [95% CI, 81-91] vs 73% [66-79]; HR 0·48 [0·30-0·76]; p=0·0015) and overall survival (92% [87-95] vs 84% [77-89]; HR 0·44 [0·28-0·81]; p=0·0071) were also increased in the R-ACVBP group. 82 (42%) of 196 patients in the R-ACVBP group experienced a serious adverse event compared with 28 (15%) of 183 in the R-CHOP group. Grade 3-4 haematological toxic effects were more common in the R-ACVBP group, with a higher proportion of patients experiencing a febrile neutropenic episode (38% [75 of 196] vs 9% [16 of 183]). INTERPRETATION: Compared with standard R-CHOP, intensified immunochemotherapy with R-ACVBP significantly improves survival of patients aged 18-59 years with diffuse large B-cell lymphoma with low-intermediate risk according to the International Prognostic Index. Haematological toxic effects of the intensive regimen were raised but manageable. FUNDING: Groupe d'Etudes des Lymphomes de l'Adulte and Amgen.

A Playful Glance at Hierarchical Questions for Two-Way Alternating Automata

Two-way alternating automata were introduced by Vardi in order to study the satisfiability problem for the modal μ-calculus extended with backwards modalities. In this paper, we present a very simple proof by way of Wadge games of the strictness of the hierarchy of Motowski indices of two-way alternating automata over trees.

A cophenetic correlation coefficient for Tocher's method

The objective of this work was to propose a way of using the Tocher's method of clustering to obtain a matrix similar to the cophenetic one obtained for hierarchical methods, which would allow the calculation of a cophenetic correlation. To illustrate the obtention of the proposed cophenetic matrix, we used two dissimilarity matrices - one obtained with the generalized squared Mahalanobis distance and the other with the Euclidean distance - between 17 garlic cultivars, based on six morphological characters. Basically, the proposal for obtaining the cophenetic matrix was to use the average distances within and between clusters, after performing the clustering. A function in R language was proposed to compute the cophenetic matrix for Tocher's method. The empirical distribution of this correlation coefficient was briefly studied. For both dissimilarity measures, the values of cophenetic correlation obtained for the Tocher's method were higher than those obtained with the hierarchical methods (Ward's algorithm and average linkage - UPGMA). Comparisons between the clustering made with the agglomerative hierarchical methods and with the Tocher's method can be performed using a criterion in common: the correlation between matrices of original and cophenetic distances.