Enabling semantic similarity estimation across multiple ontologies: an evaluation in the biomedical domain

The estimation of the semantic similarity between terms provides a valuable tool to enable the understanding of textual resources. Many semantic similarity computation paradigms have been proposed both as general-purpose solutions or framed in concrete fields such as biomedicine. In particular, ontology-based approaches have been very successful due to their efficiency, scalability, lack of constraints and thanks to the availability of large and consensus ontologies (like WordNet or those in the UMLS). These measures, however, are hampered by the fact that only one ontology is exploited and, hence, their recall depends on the ontological detail and coverage. In recent years, some authors have extended some of the existing methodologies to support multiple ontologies. The problem of integrating heterogeneous knowledge sources is tackled by means of simple terminological matchings between ontological concepts. In this paper, we aim to improve these methods by analysing the similarity between the modelled taxonomical knowledge and the structure of different ontologies. As a result, we are able to better discover the commonalities between different ontologies and hence, improve the accuracy of the similarity estimation. Two methods are proposed to tackle this task. They have been evaluated and compared with related works by means of several widely-used benchmarks of biomedical terms using two standard ontologies (WordNet and MeSH). Results show that our methods correlate better, compared to related works, with the similarity assessments provided by experts in biomedicine.     

Capture of cell-derived microvesicles (exosomes and apoptotic bodies) by human plasmacytoid dendritic cells

cDCs and pDCs differ in multiple aspects. Among those, antigen capture is a recognized feature of cDCs, whereas pDCs display poor capacity to capture cell-derived antigens. However, animal models of organ transplantation suggested a role for pDCs in tolerance induction via phagocytosis of donor antigens. In a transplantation setting, microvesicles, such as apoptotic bodies and exosomes secreted by the graft, may be potential sources of alloantigen. Here, we tested the capacity of human pDCs to capture exosomes and apoptotic bodies from Jurkat T cells. Exosomes and apoptotic bodies were indeed captured by pDCs, although required longer times of incubation when compared with the highly endocytic cDCs. In cDCs and pDCs, exosome capture was more efficient than apoptotic bodies. Endocytosis inhibitors clearly impaired exosome capture by cDCs, although this could not be verified in pDCs as a result of cellular toxicity. Functionally, capture of Jurkat-derived exosomes did not induce nor prevent pDC maturation, and exosome-loaded pDCs induced T cell proliferation, suggesting a link between capture and presentation. Thus, exosomes and apoptotic bodies may be sources of antigen for human pDCs.     

Oxidative stress triggers the amyloidogenic pathway in human vascular smooth muscle cells

Cerebral amyloid angiopathy, associated to most cases of Alzheimer's disease (AD), is characterized by the deposition of amyloid ss-peptide (Ass) in brain vessels, although the origin of the vascular amyloid deposits is still controversial: neuronal versus vascular. In the present work, we demonstrate that primary cultures of human cerebral vascular smooth muscle cells (HC-VSMCs) have all the secretases involved in amyloid ss-protein precursor (APP) cleavage and produce Ass(1-40) and Ass(1-42). Oxidative stress, a key factor in the etiology and pathophysiology of AD, up-regulates ss-site APP cleaving enzyme 1 (BACE1) expression, as well as Ass(1-40) and Ass(1-42) secretion in HC-VSMCs. This process is mediated by c-Jun N-terminal Kinase and p38 MAPK signaling and appears restricted to BACE1 regulation as no changes in the other secretases were observed. In conclusion, oxidative stress-mediated up-regulation of the amyloidogenic pathway in human cerebral vascular smooth muscle cells may contribute to the overall cerebrovascular amyloid angiopathy observed in AD patients.     

Gender differences in a cohort study of 604 bipolar patients: the role of predominant polarity

Background

Some clinical differences between gender regarding the course and outcome of bipolar disorders have already been described and some others remain still controversial.

Aims

To explore gender differences regarding clinical and socio-demographic characteristics amongst bipolar patients with particular attention to predominant polarity and depressive symptoms.

Method

Data were collected from DSM-IV type I and II bipolar patients (n = 604), resulting from the systematic follow-up of the Bipolar Disorders Program, Hospital Clinic of Barcelona, over an average follow-up of 10 years. Socio-demographic and clinical variables were collected in order to detect gender-related differences.

Results

Bipolar women are more likely than men to show a predominance of depressive polarity as well as a depressive onset whilst men would be more likely to suffer from comorbid substance use disorders. Women significantly have a higher lifetime prevalence of psychotic depression and a higher prevalence of axis II comorbid disorders. Bipolar women are also more likely to have a family history of suicide and a lifetime history of attempted suicide. Suicide attempts are more often violent amongst bipolar men. In a backward logistic regression model, two variables were responsible for most gender-related clinical differences: type of predominant polarity - more likely to be depressive amongst women - (B = − 0.794, p = 0.027, Exp(B) = 0.452; CI =  0.223-0.915), alcohol abuse (B = − 1.095, p = 0.000, Exp(B) = 2990; CI =  1.817-4.919) and cocaine abuse (B = 0.784, p = 0.033, Exp(B) = 2.189; CI =  1.066-4.496) - more prevalent amongst men.

Conclusion

The main characteristic featuring bipolar women is depression, both at illness onset and as a predominant polarity all along the illness course. This may have important diagnostic and therapeutic implications.     

Gain of multiple copies of the CBFB gene: a new genetic aberration in a case of granulocytic sarcoma

Granulocytic sarcomas (GS) are tumor masses of immature myeloid cells presenting at an extramedullary site, mainly the skin, bone, and lymph node. They are often associated with acute myeloid leukemia (AML) with monoblastic or myelomonocytic differentiation, including either AML M2 with t(8;21)(q22;q22) or AML M4Eo with inv(16)(p13q22). We present a case diagnosed with GS associated with AML M4 that presented a normal karyotype with conventional cytogenetic analysis. Although the myeloblasts did not show the inv(16)(p13q22) (CBFB/MYH11), a gain of multiple copies of the CBFB gene was detected with fluorescence in situ hybridization analysis. To our knowledge, no cases with this rare genetic anomaly have been previously described.     

Oxidative stress time course in the rat diaphragm after freezing-thawing cycles

Hyperthermia was shown to induce oxidative stress by uncoupling mitochondrial respiratory chain and to reduce superoxide dismutase (SOD) activity in muscles. Reactive carbonyl groups, malondialdehyde (MDA)-protein adducts, 3-nitrotyrosine immunoreactivity, Mn-SOD, and catalase were detected using immunoblotting in rat diaphragm specimens and homogenates thawed at room temperature (after previous storage at -80 degrees C) for 5, 15, 30, and 60 min, and 3, 6, and 24h to be subsequently and immediately stored at -80 degrees C. Mn-SOD activity was also measured in all muscles. Both total protein carbonylation (reactive carbonyl groups and MDA-protein adducts) and nitration were significantly increased over time, reaching their peaks in the diaphragms of the 60- and 15-min groups, respectively. Mn-SOD expression and activity were significantly reduced over time, while catalase expression showed no significant variation. Protein oxidation was significantly increased in the rat diaphragms exposed to freezing-thawing cycles of different time lengths, while Mn-SOD was substantially reduced in all muscles.     

Using mental tasks transitions detection to improve spontaneous mental activity classification

This paper presents an algorithm based on canonical variates transformation (CVT) and distance based discriminant analysis (DBDA) combined with a mental tasks transitions detector (MTTD) to classify spontaneous mental activities in order to operate a brain-computer interface working under an asynchronous protocol. The algorithm won the BCI Competition III--Data Set V: Multiclass Problem, Continuous EEG--achieving an averaged classification accuracy over three subjects of 68.65% (79.60, 70.31 and 56.02%, respectively) in a three-class problem.     

Composition of the HLA‐DR‐associated human thymus peptidome

Major histocompatibility complex class II (MHC‐II) molecules bind to and display antigenic peptides on the surface of antigen‐presenting cells (APCs). In the absence of infection, MHC‐II molecules on APCs present self‐peptides and interact with CD4⁺ T cells to maintain tolerance and homeostasis. In the thymus, self‐peptides bind to MHC‐II molecules expressed by defined populations of APCs specialised for the different steps of T‐cell selection. Cortical epithelial cells present peptides for positive selection, whereas medullary epithelial cells and dendritic cells are responsible for peptide presentation for negative selection. However, few data are available on the peptides presented by MHC molecules in the thymus. Here, we apply mass spectrometry to analyse and identify MHC‐II‐associated peptides from five fresh human thymus samples. The data show a diverse self‐peptide repertoire, mostly consisting of predicted MHC‐II high binders. Despite technical limitations preventing single cell population analyses of peptides, these data constitute the first direct assessment of the HLA‐II‐bound peptidome and provide insight into how this peptidome is generated and how it drives T‐cell repertoire formation.     

Autosomal genetics and Y-chromosome haplogroup L1b-M317 reveal Mount Lebanon Maronites as a persistently non-emigrating population

Currently, there are 18 different religious communities living in Lebanon. While evolving primarily within Lebanon, these communities show a level of local isolation as demonstrated previously from their Y-haplogroup distributions. In order to trace the origins and migratory patterns that may have led to the genetic isolation and autosomal clustering in some of these communities we analyzed Y-chromosome STR and SNP sample data from 6327 individuals, in addition to whole genome autosomal sample data from 609 individuals, from Mount Lebanon and other surrounding communities. We observed Y chromosome L1b Levantine STR branching that occurred around 5000 years ago. Autosomal DNA analyses suggest that the North Lebanese Mountain Maronite community possesses an ancestral Fertile Crescent genetic component distinct from other populations in the region. We suggest that the Levantine L1b group split from the Caucasus ancestral group around 7300 years ago and migrated to the Levant. This event was distinct from the earlier expansions from the Caucasus region that contributed to the wider Levantine populations. Differential cultural adaption by populations from the North Lebanese Mountains are clearly aligned with the L1b haplotype STR haplogroup clusters, indicating pre-existing and persistent cultural barriers marked by the transmission of L1b lineages. Our findings highlight the value of uniparental haplogroups and STR haplotype data for elucidating biosocial events among these populations.Subject terms: Population genetics, Computational biology and bioinformatics     

Determinants of plasma homocyst(e)ine in patients with nephrotic syndrome

Hyperhomocyst(e)inemia is an independent risk factor for atherothrombosis in several clinical settings in which renal function is impaired, but its prevalence in the nephrotic syndrome has not been investigated in detail, even though this syndrome provides an excellent model in which to study a possible link between albuminuria, proteinuria, and hyperhomocyst(e)inemia. We obtained plasma and urine from 27 patients with biopsy-confirmed membranous glomerulonephritis presenting nephrotic syndrome and 27 matched controls and determined the concentrations of homocyst(e)ine and proteins considered putative markers of glomerular and tubular function. Hyperhomocyst(e)inemia, defined as the mean +SD of the plasma homocyst(e)ine concentration of the controls [plasma homocyst(e)ine concentration >10.8 micromol/l] was present in 26% of the patients with nephrotic syndrome but in only 7.4% of the controls. Furthermore, the degree of hyperhomocyst(e)inemia was more severe in the nephrotic patients than in the controls. The existence of renal failure, tubular damage, and, interestingly, relatively well conserved glomerular function barrier were the main predictors of increased levels of plasma homocyst(e)ine. In conclusion, hyperhomocyst(e)inemia is a frequent cardiovascular risk factor present in patients with nephrotic syndrome and renal failure, but it is not directly associated with proteinuria.