AGWAN: A Generative Model for Labelled, Weighted Graphs

Real-world graphs or networks tend to exhibit a well-known set of properties, such as heavy-tailed degree distributions, clustering and community formation. Much effort has been directed into creating realistic and tractable models for unlabelled graphs, which has yielded insights into graph structure and evolution. Recently, attention has moved to creating models for labelled graphs: many real-world graphs are labelled with both discrete and numeric attributes. In this paper, we present AGWAN (Attribute Graphs: Weighted and Numeric), a generative model for random graphs with discrete labels and weighted edges. The model is easily generalised to edges labelled with an arbitrary number of numeric attributes. We include algorithms for fitting the parameters of the AGWAN model to real-world graphs and for generating random graphs from the model. Using the Enron “who communicates with whom” social graph, we compare our approach to state-of-the-art random labelled graph generators and draw conclusions about the contribution of discrete vertex labels and edge weights to the structure of real-world graphs.

Statin Use and Risk of Delirium in the Critically Ill

Rationale: Delirium is common in intensive care unit (ICU) patients and is a predictor of worse outcomes and neuroinflammation is a possible mechanism. The antiinflammatory actions of statins may reduce delirium.

Objectives: To determine whether critically ill patients receiving statin therapy had a reduced risk of delirium than those not on statins.

Methods: A prospective cohort analysis of data from consecutive ICU patients admitted to a UK mixed medical and surgical critical care unit between August 2011 and February 2012; the Confusion Assessment Method for ICU was used to determine the days each patient was assessed as being free of delirium during ICU admission.

Measurements and Main Results: Delirium-free days, daily administration of statins, and serum C-reactive protein (CRP) were recorded. Four hundred and seventy consecutive critical care patients were followed, of whom 151 patients received statins. Using randomeffects multivariable logistic regression, statin administration the previous evening was associated with the patient being assessed as free of delirium (odds ratio, 2.28; confidence interval, 1.01-5.13; P , 0.05) and with lower CRP (b = 20.52; P , 0.01) the following day. When the association between statin and being assessed as free of delirium was controlled for CRP, the effect size became nonsignificant (odds ratio, 1.56; confidence interval, 0.64-3.79; P = 0.32).

Conclusions: Ongoing statin therapy is associated with a lower daily risk of delirium in critically ill patients. An ongoing clinical trial, informed by this study, is investigating if statins are a potential therapy for delirium in the critically ill.Copyright © 2014 by the American Thoracic Society.

Human Parechovirus Infection in Neonatal Intensive Care

Background: Approximately 5-6% of all infective episodes in NICU are of viral origin. Previous studies suggest that human parechovirus (HPeV) infection presents most commonly in term infants, as a sepsis-like syndrome in which meningoencephalitis is prominent. Our aim was to study the infection rate and associated features of HPeV.

Methods: Blood samples were taken from NICU babies greater than 48 hours old, who were being investigated for late onset sepsis. Clinical and laboratory data were collected at the time of the suspected sepsis episode. Samples were tested using universal primers and probe directed at the 5'-untranslated region of the HPeV genome by reverse transcriptase PCR. Results were confirmed by electrophoresis and DNA sequencing.

Results: HPeV was detected in 11 of 84 samples (13%). These infants had a mean (interquartile range, IQR) gestational age of 28.9 (26.9 - 30.6) weeks and mean birth weight of 1.26 (SD = 0.72) kg. The median day of presentation was 16 (IQR: 11-27). These characteristics were similar to the infants without positive viral detection. Six infants presented with respiratory signs. One infant presented with signs of meningitis. Six of the 11 episodes of HPeV infection occurred during the winter months (December - February). No HPeV positive infants had abnormal findings on their 28-day cranial ultrasound examination.

Conclusions: We found a HPeV infection rate of 13% in infants being tested for late onset sepsis. HPeV should be considered as a possible cause of sepsis-like symptoms in preterm infants.

Finding the Most Descriptive Substructures in Graphs with Discrete and Numeric Labels

Many graph datasets are labelled with discrete and numeric attributes. Most frequent substructure discovery algorithms ignore numeric attributes; in this paper we show how they can be used to improve search performance and discrimination. Our thesis is that the most descriptive substructures are those which are normative both in terms of their structure and in terms of their numeric values. We explore the relationship between graph structure and the distribution of attribute values and propose an outlier-detection step, which is used as a constraint during substructure discovery. By pruning anomalous vertices and edges, more weight is given to the most descriptive substructures. Our method is applicable to multi-dimensional numeric attributes; we outline how it can be extended for high-dimensional data. We support our findings with experiments on transaction graphs and single large graphs from the domains of physical building security and digital forensics, measuring the effect on runtime, memory requirements and coverage of discovered patterns, relative to the unconstrained approach.

A genome-wide association study of anorexia nervosa

Anorexia nervosa (AN) is a complex and heritable eating disorder characterized by dangerously low body weight. Neither candidate gene studies nor an initial genome-wide association study (GWAS) have yielded significant and replicated results. We performed a GWAS in 2907 cases with AN from 14 countries (15 sites) and 14 860 ancestrally matched controls as part of the Genetic Consortium for AN (GCAN) and the Wellcome Trust Case Control Consortium 3 (WTCCC3). Individual association analyses were conducted in each stratum and meta-analyzed across all 15 discovery data sets. Seventy-six (72 independent) single nucleotide polymorphisms were taken forward for in silico (two data sets) or de novo (13 data sets) replication genotyping in 2677 independent AN cases and 8629 European ancestry controls along with 458 AN cases and 421 controls from Japan. The final global meta-analysis across discovery and replication data sets comprised 5551 AN cases and 21 080 controls. AN subtype analyses (1606 AN restricting; 1445 AN binge-purge) were performed. No findings reached genome-wide significance. Two intronic variants were suggestively associated: rs9839776 (P=3.01 × 10(-7)) in SOX2OT and rs17030795 (P=5.84 × 10(-6)) in PPP3CA. Two additional signals were specific to Europeans: rs1523921 (P=5.76 × 10(-)(6)) between CUL3 and FAM124B and rs1886797 (P=8.05 × 10(-)(6)) near SPATA13. Comparing discovery with replication results, 76% of the effects were in the same direction, an observation highly unlikely to be due to chance (P=4 × 10(^-6)), strongly suggesting that true findings exist but our sample, the largest yet reported, was underpowered for their detection. The accrual of large genotyped AN case-control samples should be an immediate priority for the field.

AGWAN: A Generative Model for Labelled, Weighted Graphs

Real-world graphs or networks tend to exhibit a well-known set of properties, such as heavy-tailed degree distributions, clustering and community formation. Much effort has been directed into creating realistic and tractable models for unlabelled graphs, which has yielded insights into graph structure and evolution. Recently, attention has moved to creating models for labelled graphs: many real-world graphs are labelled with both discrete and numeric attributes. In this paper, we presentAgwan (Attribute Graphs: Weighted and Numeric), a generative model for random graphs with discrete labels and weighted edges. The model is easily generalised to edges labelled with an arbitrary number of numeric attributes. We include algorithms for fitting the parameters of the Agwanmodel to real-world graphs and for generating random graphs from the model. Using real-world directed and undirected graphs as input, we compare our approach to state-of-the-art random labelled graph generators and draw conclusions about the contribution of discrete vertex labels and edge weights to graph structure.

"Hidden" Social Networks in Behavior Change Interventions

We investigated whether “hidden” (or unobserved) social networks were evident in a 2011 physical activity behavior change intervention in Belfast, Northern Ireland. Results showed evidence of unobserved social networks in the intervention and illustrated how the network evolved over short periods and affected behavior. Behavior change interventions should account for the interaction among participants (i.e., social networks) and how such interactions affect intervention outcome.

The Politics of Observation:Radical Psychiatry and Social Documentary

How did the counter-cultural aims of Radical Psychiatry coincide with those of documentary filmmaking in the 1960s? Where the forms and structures of new approaches to the documentary necessarily complicit in promoting the clinical and anti-clinical practices, and wider political agenda, of Radical Psychiatry? How did the documentary deal with the ethical, aesthetic, and audience-related issues associated with filming personalities and environments associated with Radical Psychiatry? How did Radical Psychiatry and the documentary shape postwar discourses on trauma, especially within conflict and post-conflict (PTSD) contexts? What is the legacy of Radical Pschiatry today, and how has it been explored by contemporary documentray film?

This article addresses these question by examining a range of documentaries dealing with the radical and 'anti-psychiatric' ideas and methods of figures such as R.D.Laing, David Cooper, Jan Bastiaans, Timothy Leary, and Franco Basaglia. Films analysed include Peter Robinson's Asylum (1972) and Psychiatry and Violence (1973); Ah, Sunflower (Klinkert and Sinclair, 1967); Anatomy of Violence (Davis, 1967); Turn On, Tune In, Drop Out (Robin Clarke, 1967), W. R. - Mysteries of the Organism (Makavejev, 1971); Raymond Depardon's San Clemente (1980) and Urgences (1988); and Louis van Gasteren's trilogy Now Do You Get it Why I am Crying (1969), The Price of Survival (2003), and There is No Plane to Zagreb (2012). 
The article concludes with a discussion of Nicolas Philibert's Every Little Thing (1997) within the context of the French documentary tradition and the film's more immediate subject - the famous clinic at La Borde established by Jean Oury, and associated with the methods and theories of figures such as Jacques
 Lacan, Francesc Tosquelles, Franz Fanon, and Félix Guattari.