MARML: Motif-Aware Deep Representation Learning in Multilayer Networks.

The rapid increase in high-throughput, complex, and heterogeneous data has led to the adoption of network-structured models and analyses for interpretation. However, these data are inherently complex and challenging to understand, prompting researchers to turn to graph embedding methods to facilitate analysis. While general network embedding techniques have shown promise in improving downstream prediction and classification tasks, real-world data are complicated due to cross-domain interactions between different types of entities. Multilayered networks have been successful in integrating biological data to represent biological systems' hierarchy, but embedding nodes based on different types of interactions remains an unsolved problem. To address this challenge, we propose the Motif-aware deep representation learning in multilayer (MARML) networks for learning network representations. Our method considers recurring motif patterns, topological information, and attributive information from other sources as node features. We validated the MARML method using various multilayer network datasets. In addition, by incorporating motif information, MARML considers higher order connections across different hierarchies. The learned features exhibited excellent accuracy in tasks related to link prediction and link differentiation, enabling us to distinguish between existing and disconnected triplets. Through the integration of both intrinsic node attributes and topological network structures, we enhance our understanding of complex biological systems.

Patient Graph Deep Learning to Predict Breast Cancer Molecular Subtype.

Breast cancer is a heterogeneous disease consisting of a diverse set of genomic mutations and clinical characteristics. The molecular subtypes of breast cancer are closely tied to prognosis and therapeutic treatment options. We investigate using deep graph learning on a collection of patient factors from multiple diagnostic disciplines to better represent breast cancer patient information and predict molecular subtype. Our method models breast cancer patient data into a multi-relational directed graph with extracted feature embeddings to directly represent patient information and diagnostic test results. We develop a radiographic image feature extraction pipeline to produce vector representation of breast cancer tumors in DCE-MRI and an autoencoder-based genomic variant embedding method to map variant assay results to a low-dimensional latent space. We leverage related-domain transfer learning to train and evaluate a Relational Graph Convolutional Network to predict the probabilities of molecular subtypes for individual breast cancer patient graphs. Our work found that utilizing information from multiple multimodal diagnostic disciplines improved the model's prediction results and produced more distinct learned feature representations for breast cancer patients. This research demonstrates the capabilities of graph neural networks and deep learning feature representation to perform multimodal data fusion and representation in the breast cancer domain.

Protein Family Classification from Scratch: A CNN Based Deep Learning Approach.

Next-generation sequencing techniques provide us with an opportunity for generating sequenced proteins and identifying the biological families and functions of these proteins. However, compared with identified proteins, uncharacterized proteins consist of a notable percentage of the overall proteins in the bioinformatics research field. Traditional family classification methods often devote themselves to extracting N-Gram features from sequences while ignoring motif information as well as affinity information between motifs and adjacent amino acids. Previous clustering-based algorithms have typically been used to define protein features with domain knowledge and annotate protein families based on extensive data samples. In this paper, we apply CNN based amino acid representation learning with limited characterized proteins to explore the performances of annotated protein families by taking into account the amino acid location information. Additionally, we apply the method to all reviewed protein sequences with their families retrieved from the UniProt database to evaluate our approach. Last but not least, we verify our model using those unreviewed protein records, which is typically ignored by other methods.

Ontology-based metabolomics data integration with quality control.

Aim: The complications that arise when performing meta-analysis of datasets from multiple metabolomics studies are addressed with computational methods that ensure data quality, completeness of metadata and accurate interpretation across studies. Results & methodology: This paper presents an integrated system of quality control (QC) methods to assess metabolomics results by evaluating the data acquisition strategies and metabolite identification process when integrating datasets for meta-analysis. An ontology knowledge base and a rule-based system representing the experiment and chemical background information direct the processes involved in data integration and QC verification. A diabetes meta-analysis study using these QC methods finds putative biomarkers that differ between cohorts. Conclusion: The methods presented here ensure the validity of meta-analysis when integrating data from different metabolic profiling studies.

Possible association between social cognition and metabolic dysfunctions in Bipolar Disorder and Schizophrenia: Preliminary results.

BACKGROUND: Social cognition (SC) and Theory of Mind (ToM) are compromised in patients with Schizophrenia (SKZ) and Bipolar Disorder (BD) and an increased frequency of metabolic abnormalities is reported in both disorders. Obesity seems associated with cognitive impairments The aim of our study is thus to assess the relationship between obesity and ToM in SKZ and BD. METHODS: 36 stabilized outpatients (18 SKZ and 18 BD) were recruited and completed Reading the Mind in the Eyes Test, Italian version and Faux Pas Recognition Test, adult version. BMI was calculated from self-reported height and weight. Two different Generalized Linear Models were created including performance in Eyes test and in Faux Pas test as outcomes and BMI as covariate. RESULTS: After stratifying for sex, we found a significant relationship between BMI and Faux Pas performance for male patients (p = 0.017), without significant interactions between sex and diagnosis. These results suggest a BMI effect on both affective and cognitive ToM in male patients. LIMITATIONS: Major confounders need to be considered: the greater number of subjects with SKZ in male subsample, a possible influence of neurocognitive performance, small sample size and self-reported BMI. CONCLUSIONS: There could be a relationship between ToM and metabolic dysfunctions, at least in male patients. The exact nature of this relationship has yet to be determined; an interesting theoretical framework is based on a combination of increased brain energy request and inefficient peripheral compensatory mechanisms, resulting in inefficient energy allocation to the brain.

BDNF and BMI effects on brain structures of bipolar offspring: results from the global mood and brain science initiative.

OBJECTIVE: To compare brain-derived neurotrophic factor (BDNF) levels between offspring of individuals with bipolar disorders (BD) and healthy controls (HCs) and investigate the effects of BDNF levels and body mass index (BMI) on brain structures. METHOD: Sixty-seven bipolar offspring and 45 HCs were included (ages 8-28). Structural images were acquired using 3.0 Tesla magnetic resonance imaging. Serum BDNF levels were measured using enzyme-linked immunosorbent assay. Multivariate and univariate analyses of covariance were conducted. RESULTS: Significantly higher BDNF levels were observed among bipolar offspring, relative to HCs (P > 0.025). Offspring status moderated the association between BDNF and BMI (F1 =4.636, P = 0.034). After adjustment for relevant covariates, there was a trend for a significant interaction of group and BDNF on neuroimaging parameters (Wilks'λ F56,94 =1.463, P = 0.052), with significant effects on cerebellar white matter and superior and middle frontal regions. Brain volume and BDNF were positively correlated among HCs and negatively correlated among bipolar offspring. Interactions between BDNF and BMI on brain volumes were non-significant among HCs (Wilks'λ F28,2 =2.229, P = 0.357), but significant among bipolar offspring (Wilks'λ F28,12 =2.899, P = 0.028). CONCLUSION: Offspring status and BMI moderate the association between BDNF levels and brain structures among bipolar offspring, underscoring BDNF regulation and overweight/obesity as key moderators of BD pathogenesis.

U-shaped relationship between depression and body mass index in the Korean adults.

BACKGROUND: Although a number of studies have examined the relationship between depression and obesity, it is still insufficient to establish the specific pattern of relationship between depression and body mass index (BMI) categories. Thus, this study was aimed to investigate the relationship between depression and BMI categories. METHODS: A cross-sectional study was conducted for a cohort of 159,390 Korean based on Kangbuk Samsung Health Study (KSHS). Study participants were classified into 5 groups by Asian-specific cut-off of BMI (18.5, 23, 25 and 30kg/m2). The presence of depression was determined by Center for Epidemiologic Studies-Depression scales (CES-D)≥16 and≥25. The adjusted odd ratios (ORs) for depression were evaluated by multiple logistic regression analysis, in which independent variable was 5 categories of BMI and dependent variable was depression. Subgroup analysis was conducted by gender and age. RESULTS: When normal group was set as a reference, the adjusted ORs for depression formed U-shaped pattern of relationship with BMI categories [underweight: 1.31 (1.14-1.50), overweight: 0.94 (0.85-1.04), obese group: 1.01 (0.91-1.12), severe obese group: 1.28 (1.05-1.54)]. This pattern of relationship was more prominent in female and young age group than male and elderly subgroup. BMI level with the lowest likelihood of depression was 18.5kg/m2 to 25kg/m2 in women and 23kg/m2 to 25kg/m2 in men. CONCLUSIONS: There was a U-shaped relationship between depression and BMI categories. This finding suggests that both underweight and severe obesity are associated with the increased risk for depression.

Repurposing of Anti-Diabetic Agents for the Treatment of Cognitive Impairment and Mood Disorders.

Impairments in cognitive function represent a consistent, non-specific, and clinically significant feature in metabolic, mood, and dementing disorders. The foregoing observation is instantiated by evidence demonstrating that these disorders share pathophysiological mechanisms including, but not limited to, aberrant insulin signaling, inflammation, and glucocorticoid activity. Moreover, these mechanisms have been consistently reported to increase vulnerability to and/or exacerbate impairments in cognitive function. Notwithstanding evidence suggesting a bidirectional relationship between disturbances in the metabolic milieu, mood, and increased risk for dementia, efficacious treatments that target cognitive impairments in these populations do not presently exist. Taken together, it is proposed that anti-diabetic agents may aid the management of mood disorders and future risk for dementia through disease modification by targeting underlying pathophysiological mechanisms (e.g., aberrant metabolic function) rather than focusing solely on symptom mitigation. The current aim is to provide a brief narrative review of extant studies that report on the potential neurotherapeutic effects of anti-diabetic agents on disturbances in mood and impairments in cognitive function.

Inflammation, neurotrophism and oxidative stress and childhood psychopathology in a large community sample.

OBJECTIVE: To investigate the association between peripheral biomarkers and child psychopathology in a large community sample. METHOD: A total of 625 aged 6- to 13-year old subjects were recruited from a community school-based study. Psychopathology was assessed using the Child Behaviour Checklist (CBCL). Psychiatric diagnosis was evaluated using the Development and Well-Being Assessment. The following biomarkers were examined in peripheral blood: brain-derived neurotrophic factor, cytokines (IL-2, IL-4, IL-6, IL-10, IL-17, IFN-g, and TNF-α), chemokines (eotaxin/CCL11, IP-10, MCP-1), cytokine receptors (sTNFR1 and sTNFR2), and the oxidative stress marker TBARS. RESULTS: We found significant associations between sTNFR2, eotaxin/CCL11 and CBCL total score, as well as with specific dimensions of psychopathology. There were different patterns of association between these biomarkers and psychological and behavioural symptoms in children with and without a mental disorder. TBARS, IL-6 and MCP-1 were more specific to some clusters of symptoms in children with a psychiatric diagnosis. CONCLUSION: Our data support the potential use of biomarkers, especially those involved in immune-inflammatory pathways, in investigating neurodevelopmental psychopathology. Their association with different dimensions of symptoms might be of useful when analyzing illness severity and clusters of symptoms within specific disorders.

New results with 100 Excluder cases.

AIM: The purpose of this study was to report the outcome of EVAR using EXCLUDER endoprosthesis (W. L. Gore & Associates, Flagstaff, Ariz.) with low-porosity polytetrafluoroethylene (PTFE) in the medium term. We reviewed a 10-year-experience with this device to document the outcome of 100 consecutive elective EVARs with Excluder-Device performed at a single centre. METHODS: From 01.2006 to 01.2009 all elective abdominal aortic aneurysm (AAA) cases (N=100) with an abdominal aortic aneurysm (AAA; mean diameter 5.61 cm; range 4.2-7.3 cm) that were treated electively with the EXCLUDER Bifurcated Endoprosthesis were entered in an index. Anatomical and clinical evaluations and radiological results have been analyzed. Mean aortic neck length was 12.24 mm, mean proximal aortic diameter was 24.39 mm. Primary outcome that has been examined, include operative mortality, aneurysm rupture, aneurysm-related mortality, open surgical conversion, and late survival rates. The incidence of endoleak, migration, aneurysm enlargement, and graft patency have also been determined. Finally, the need for reinterventions and success of such secondary procedures were evaluated. Endoleaks were diagnosed from contrast-enhanced computed tomography, the rate of type II endoleaks was analyzed after 1, 3 and 12 months. RESULTS: Mean patient age was 74.1 years (range 44-91 years); 91% were male. Mean follow-up was 20 months; 78 % of the patients had 2 or more major comorbidities, and 32 % were categorized as inappropriate for open repair. On an intent-to-treat basis, device deployment was successful in 100%. Thirty-day mortality was 0%; freedom from AAA rupture was 100%. Type II endoleak appeared directly after the procedure in 24%, after 3 months in 15% and after 12 months in 7%. The initial technical success defined as endovascular aneurysm exclusion and absence of type I endoleak was 99%. Proximal type I early endoleak occurred (detected by intraoperative DSA) in 3 patients, in 2 cases intraoperative proximal aortic cuffs were needed to seal proximal type I endoleaks. In one case we performed a proximal bending via a retroperitoneal access and sealed the endoprosthesis. Postoperatively, the size of the AAA decreased or remained unchanged in 93% after 12 months. Freedom from reinterventions was 94% after 2 years. CONCLUSION: EVAR using the EXCLUDER-Device is a safe, effective, and durable method to prevent AAA rupture and aneurysm-related death. Assuming suitable AAA anatomy, these data justify a broad application within a wide spectrum of patients.

[Zollinger-Ellison syndrome: a case report.]. / Síndrome de Zollinger-Ellison: informe de un caso.

Forty-two year-old woman, with no relevant medical history, presented with 3-year burning epigastric pain radiating to right upper quadrant, accompanied by nausea and vomiting, food intolerance and estheatorrea. She was treated by 2 years with H2 blockers. Cholelithiasis was found and laparoscopic cholecystectomy was performed without changes in clinical picture. She presented upper gastrointestinal bleeding secondary to severe esophagitis and gastritis. Proton pump inhibitors at high doses were added to the treatment with partial improvement. Abdominal ultrasound showed tumor in pancreatic head but it was not confirmed by computed tomography. High levels of serum gastrin (17,251 µg/L) were detected. Patient underwent laparotomy and with the aid of ultrasound the mass in the head of the pancreas was localized and resected. Recovery was uneventful. The pathology report was a gastrinoma. At one year follow up, patient remains asymptomatic.

semQA: SPARQL with Idempotent Disjunction.

The SPARQL LeftJoin abstract operator is not distributive over Union; this limits the algebraic manipulation of graph patterns, which in turn restricts the ability to create query plans for distributed processing or query optimization. In this paper, we present semQA, an algebraic extension for the SPARQL query language for RDF, which overcomes this issue by transforming graph patterns through the use of an idempotent disjunction operator Or as a substitute for Union. This permits the application of a set of equivalences that transform a query into distinct forms. We further present an algorithm to derive the solution set of the original query from the solution set of a query where Union has been substituted by Or. We also analyze the combined complexity of SPARQL, proving it to be NP-complete. It is also shown that the SPARQL query language is not, in the general case, fixed-parameter tractable. Experimental results are presented to validate the query evaluation methodology presented in this paper against the SPARQL standard to corroborate the complexity analysis and to illustrate the gains in processing cost reduction that can be obtained through the application of semQA.

Ontology Matching with Semantic Verification.

ASMOV (Automated Semantic Matching of Ontologies with Verification) is a novel algorithm that uses lexical and structural characteristics of two ontologies to iteratively calculate a similarity measure between them, derives an alignment, and then verifies it to ensure that it does not contain semantic inconsistencies. In this paper, we describe the ASMOV algorithm, and then present experimental results that measure its accuracy using the OAEI 2008 tests, and that evaluate its use with two different thesauri: WordNet, and the Unified Medical Language System (UMLS). These results show the increased accuracy obtained by combining lexical, structural and extensional matchers with semantic verification, and demonstrate the advantage of using a domain-specific thesaurus for the alignment of specialized ontologies.

semCDI: a query formulation for semantic data integration in caBIG.

OBJECTIVES: To develop mechanisms to formulate queries over the semantic representation of cancer-related data services available through the cancer Biomedical Informatics Grid (caBIG). DESIGN: The semCDI query formulation uses a view of caBIG semantic concepts, metadata, and data as an ontology, and defines a methodology to specify queries using the SPARQL query language, extended with Horn rules. semCDI enables the joining of data that represent different concepts through associations modeled as object properties, and the merging of data representing the same concept in different sources through Common Data Elements (CDE) modeled as datatype properties, using Horn rules to specify additional semantics indicating conditions for merging data. Validation In order to validate this formulation, a prototype has been constructed, and two queries have been executed against currently available caBIG data services. DISCUSSION: The semCDI query formulation uses the rich semantic metadata available in caBIG to build queries and integrate data from multiple sources. Its promise will be further enhanced as more data services are registered in caBIG, and as more linkages can be achieved between the knowledge contained within caBIG's NCI Thesaurus and the data contained in the Data Services. CONCLUSION: semCDI provides a formulation for the creation of queries on the semantic representation of caBIG. This constitutes the foundation to build a semantic data integration system for more efficient and effective querying and exploratory searching of cancer-related data.

Prevalence and magnitude of trachoma in a local government area of Sokoto State, north western Nigeria.

INTRODUCTION: In order to have baseline data for monitoring and evaluation of trachoma control activities in Sabon Birni LGA of Sokoto state Nigeria, a population based trachoma survey was conducted. The survey was designed to determine the prevalence of active trachoma and trichiasis as well as assess the prevalence of major risk factors for trachoma in the study area. METHOD: The survey was a population based cross sectional survey of persons of all ages in the study area. A minimum sample size of 2760 was determined and the study population was selected by two stage cluster random sampling technique. In the first selection stage 34 communities were randomly selected by probability proportional to size, while for the second sampling stage 90 persons of all ages were randomly chosen in each selected cluster. All chosen persons were assessed for trachoma using WHO trachoma simplified grading system and their households were assessed for trachoma environmental risk factors. RESULTS: The prevalence of blindness in the study population was 2.0% (95% CI 1.4%-2.6%) with trachoma responsible for l3.8%. The prevalence of active trachoma amongst children (1-9 years) was 37% (CI 34.3%-39.8%), while trichiasis amongst adults was 2.3% (95% CI 1.9%-2.7%). Over sixty percent of households had no functional latrine, while 81% of households had refuse and animal dung littered within their compounds. Sixty percent of children had 'unclean faces' and over 85% of households had access to water within their villages. CONCLUSIONS: Trachoma of public health concern still exists in some districts of northern Nigeria with significant environmental risk factors.

A multicenter, site-independent, blinded study to compare the diagnostic accuracy of contrast-enhanced magnetic resonance angiography using 1.0M gadobutrol (Gadovist) to intraarterial digital subtraction angiography in body arteries.

PURPOSE: Prospective evaluation of diagnostic accuracy of single field-of-view contrast-enhanced MR Angiography (ceMRA) with 1.0M gadobutrol compared to intraarterial DSA in body arteries. MATERIALS AND METHODS: In an European multicenter study 179 patients underwent ceMRA and DSA. For each indication five prospectively defined vessel segments were evaluated by local investigators onsite and by three site-independent blinded readers (BR) independently. RESULTS: The agreement between ceMRA and DSA diagnosis was statistically significant in the onsite (96.6%) and blinded reader (86.6-90.2%) evaluation. Sensitivity, specificity, accuracy, positive (PPV) and negative predictive values (NPV) for detection of relevant stenosis (>50%) were calculated for the right and left internal carotid arteries, and common and external iliac arteries: Sensitivity was 95-98% (onsite) and 76-96% (BR), specificity 94-96% (onsite) and 86-94% (BR), accuracy 96% (onsite) and 87-93% (BR), NPV 98-99% (onsite) and 84-98% (BR), and PPV 79-93% (onsite) and 44-91% (BR), respectively. CONCLUSION: CeMRA of body arteries using 1.0M gadobutrol provides diagnostic information comparable to intraarterial DSA.

Multivariate statistical model for 3D image segmentation with application to medical images.

In this article we describe a statistical model that was developed to segment brain magnetic resonance images. The statistical segmentation algorithm was applied after a pre-processing stage involving the use of a 3D anisotropic filter along with histogram equalization techniques. The segmentation algorithm makes use of prior knowledge and a probability-based multivariate model designed to semi-automate the process of segmentation. The algorithm was applied to images obtained from the Center for Morphometric Analysis at Massachusetts General Hospital as part of the Internet Brain Segmentation Repository (IBSR). The developed algorithm showed improved accuracy over the k-means, adaptive Maximum Apriori Probability (MAP), biased MAP, and other algorithms. Experimental results showing the segmentation and the results of comparisons with other algorithms are provided. Results are based on an overlap criterion against expertly segmented images from the IBSR. The algorithm produced average results of approximately 80% overlap with the expertly segmented images (compared with 85% for manual segmentation and 55% for other algorithms).

Potential biomarkers of exposure and effect among glass craftsmen and braziers exposed to nitrogen oxides.

The purpose of this research was to determine occupational exposure of glass craftsmen and braziers to inhaled nitrogen oxides nitrogen dioxide and nitric oxide and to relate this to urinary nitrate, hydroxyproline and thioethers and to breath pentane. The glass craftsmen were exposed to nitrogen oxides at levels exceeding the occupational exposure standard and higher than braziers or controls. Urinary nitrate excretion was elevated. In these exposed workers hydroxyproline, thioethers and breath pentane were all elevated compared with controls and higher than in braziers who were less exposed. There was, however, only a correlation between individual levels of exposure nitrogen oxides in the breathing zone and breath pentane and there were no individual correlations between markers of effect and excretion of nitrate in the urine at the sample time.