Assisting medical annotation in Swiss-Prot using statistical classifiers

Bio-medical knowledge bases are valuable resources for the research community. Original scientific publications are the main source used to annotate them. Medical annotation in Swiss-Prot is specifically targeted at finding and extracting data about human genetic diseases and polymorphisms. Curators have to scan through hundreds of publications to select the relevant ones. This workload can be greatly reduced by using bio-text mining techniques. Using a combination of natural language processing (NLP) techniques and statistical classifiers, we achieve recall points of up to 84% on the potentially interesting documents and a precision of more than 96% in detecting irrelevant documents. Careful analysis of the document pre-processing chain allows us to measure the impact of some steps on the overall result, as well as test different classifier configurations. The best combination was used to create a prototype of a search and classification tool that is currently tested by the database curators.     

Rapid and accurate identification of human isolates of Pasteurella and related species by sequencing the sodA gene

The identification of Pasteurella and related bacteria remains a challenge. Here, a 449- to 473-bp fragment (sodA(int)) internal to the sodA gene, encoding the manganese-dependent superoxide dismutase, was amplified and sequenced with a single pair of degenerate primers from the type strains of Pasteurella (18 strains), Gallibacterium (1 strain), and Mannheimia (5 strains) species. The sodA(int)-based phylogenetic tree was in general agreement with that inferred from the analysis of the corresponding 16S rRNA gene sequences, with members of the Pasteurella sensu stricto cluster (Pasteurella multocida, Pasteurella canis, Pasteurella dagmatis, and Pasteurella stomatis) forming a monophyletic group and Gallibacterium and Mannheimia being independent monophyletic genera. However, the sodA(int) sequences showed a markedly higher divergence than the corresponding 16S rRNA genes, confirming that sodA is a potent target to differentiate related species. Thirty-three independent human clinical isolates phenotypically assigned to 13 Pasteurella species by a reference laboratory were successfully identified by comparing their sodA(int) sequences to those of the type species. In the course of this work, we identified the first Gallibacterium anatis isolate ever reported from a human clinical specimen. The sodA(int) sequences of the clinical isolates displayed less than 2.5% divergence from those of the corresponding type strains, except for the Pasteurella pneumotropica isolates, which were closely related to each other (> 98% sodA(int) sequence identity) but shared only 92% sodA(int) identity with the type strain. The method described here provides a rapid and accurate tool for species identification of Pasteurella isolates when access to a sequencing facility is available.     

A novel tyrosine kinase inhibitor restores chondrocyte differentiation and promotes bone growth in a gain-of-function Fgfr3 mouse model

Activating germline fibroblast growth factor receptor 3 (FGFR3) mutations cause achondroplasia (ACH), the most common form of human dwarfism and a spectrum of skeletal dysplasias. FGFR3 is a tyrosine kinase receptor and constitutive FGFR3 activation impairs endochondral ossification and triggers severe disorganization of the cartilage with shortening of long bones. To decipher the role of FGFR3 in endochondral ossification, we analyzed the impact of a novel tyrosine kinase inhibitor (TKI), A31, on both human and mouse mutant FGFR3-expressing cells and on the skeleton of Fgfr3(Y367C/+) dwarf mice. We found that A31 inhibited constitutive FGFR3 phosphorylation and restored the size of embryonic dwarf femurs using an ex vivo culture system. The increase in length of the treated mutant femurs was 2.6 times more than for the wild-type. Premature cell cycle exit and defective chondrocyte differentiation were observed in the Fgfr3(Y367C/+) growth plate. A31 restored normal expression of cell cycle regulators (proliferating cell nuclear antigen, KI67, cyclin D1 and p57) and allowed pre-hypertrophic chondrocytes to properly differentiate into hypertrophic chondocytes. Our data reveal a specific role for FGFR3 in the cell cycle and chondrocyte differentiation and support the development of TKIs for the treatment of FGFR3-related chondrodysplasias.     

Comparative inflammatory properties of staphylococcal superantigenic enterotoxins SEA and SEG: implications for septic shock

The severity of Staphylococcus aureus sepsis is positively associated with staphylococcal enterotoxin A (SEA) and negatively associated with the enterotoxin gene cluster (egc), which encodes five staphylococcal enterotoxins. We postulated that the variable, clinical severity of S. aureus sepsis might be a result of differences in the inflammatory properties of staphylococcal superantigens. We therefore compared the inflammatory properties of SEA with those of staphylococcal entérotoxin G (SEG), a member of the five egc superantigens. We found that SEA and SEG had similar superantigenic properties, as they induced CD69 expression on T lymphocytes and selective expansion of Vbeta subpopulations. Contrary to SEG, however, SEA induced a strong proinflammatory/Th1 response, including TNF-alpha and MIP-1alpha production. These results suggest that the association of SEA with the severity of S. aureus septic shock, characterized by a deleterious, inflammatory cascade, may be explained partly by the specific proinflammatory properties of this superantigen.     

Distinct patterns of DNA copy number alteration are associated with different clinicopathological features and gene-expression subtypes of breast cancer

Breast cancer is a leading cause of cancer-death among women, where the clinicopathological features of tumors are used to prognosticate and guide therapy. DNA copy number alterations (CNAs), which occur frequently in breast cancer and define key pathogenetic events, are also potentially useful prognostic or predictive factors. Here, we report a genome-wide array-based comparative genomic hybridization (array CGH) survey of CNAs in 89 breast tumors from a patient cohort with locally advanced disease. Statistical analysis links distinct cytoband loci harboring CNAs to specific clinicopathological parameters, including tumor grade, estrogen receptor status, presence of TP53 mutation, and overall survival. Notably, distinct spectra of CNAs also underlie the different subtypes of breast cancer recently defined by expression-profiling, implying these subtypes develop along distinct genetic pathways. In addition, higher numbers of gains/losses are associated with the "basal-like" tumor subtype, while high-level DNA amplification is more frequent in "luminal-B" subtype tumors, suggesting also that distinct mechanisms of genomic instability might underlie their pathogenesis. The identified CNAs may provide a basis for improved patient prognostication, as well as a starting point to define important genes to further our understanding of the pathobiology of breast cancer. This article contains Supplementary Material available at http://www.interscience.wiley.com/jpages/1045-2257/suppmat     

Pilot study of a strategy combining coronary angioplasty with valvular and/or coronary surgery on the same day

A strategy combining percutaneous coronary angioplasty followed by valvular and/or coronary surgery was recently proposed as an alternative to the classical surgical only approach. The aim of this study was to assess the feasibility and the results of such a combined strategy with the two procedures performed the same day. The population comprised 34 patients including 17 with valvular disease and revascularisable coronary lesions (15 symptomatic severe aortic stenoses and two acute mitral insufficiencies) plus 17 multitrunk coronary patients without valvular disease but with an indication for revascularisation. Angioplasty was performed several hours prior to surgery and a loading dose of 300mg clopidogrel was administered immediately postoperatively; all patients were on aspirin before the procedure. The average age was 67 +/- 11 years, NYHA class 2.3 +/- 0.7, angina 73%, LVEF 58 +/- 10%. Single coronary artery disease was present in 26%, two vessel disease in 35% and three vessel disease in 39%. The success rate for angioplasty plus stent was 98%. 60 stents were active. Bypasses were exclusively arterial (left or right internal mammary arteries). We observed 4 in-hospital deaths, one of which was due to an infarct and three due to extra-cardiac causes (1 non-cardiogenic acute respiratory distress syndrome, 1 respiratory tract infection and 1 pyelonephritis). Further surgery was necessary in 4 cases: for haemorrhage and one episode of digestive tract haemorrhage. There were no additional deaths, coronary events nor haemorrhage at the end of an average follow-up of 15 +/- 6 months. The results of this combined strategy are encouraging in this population and merit further evaluation in a prospective study.     

Persisting low monocyte human leukocyte antigen-DR expression predicts mortality in septic shock

Objective The immediate overwhelming release of inflammatory mediators in septic shock is rapidly followed by strong anti-inflammatory responses inducing a state of immunosuppression. The patients who survive the initial hyper-inflammatory step of septic shock but subsequently die may be those who do not recover from immunosuppression. We assessed whether a low monocyte human leukocyte antigen-DR (mHLA-DR) expression, proposed as a marker of immunosuppression, is an independent predictor of mortality in patients who survived the initial 48 h of septic shock. Design and setting Prospective observational study performed in two adult intensive care units at a university hospital. Patients 93 consecutive patients with septic shock.Measurements and results At days 1–2, mHLA-DR values (determined by flow cytometry) were not significantly different between survivors and non-survivors. A sharp difference became highly significant at days 3–4 when survivors had increased their values, while non-survivors had not (43% vs. 18%, percentage of HLA-DR positive monocyte, p < 0.001). Multivariate logistic regression analysis revealed that low mHLA-DR (< 30%) at days 3–4 remained independently associated with mortality after adjustment for usual clinical confounders, adjusted odds ratio (CI): 6.48 (95% CI: 1.62–25.93). Conclusion The present preliminary results show that mHLA-DR is an independent predictor of mortality in septic shock patients. Being a marker of immune failure, low mHLA-DR may provide a rationale for initiating therapy to reverse immunosuppression. After validation of the current results in multicenter studies, mHLA-DR may help to stratify patients when designing a mediator-directed therapy in a time-dependent manner. This article is discussed in the editorial available at:     

Significance of TP53 mutations as predictive markers of adjuvant cisplatin‐based chemotherapy in completely resected non‐small‐cell lung cancer

Adjuvant cisplatin‐based chemotherapy only marginally improves survival in patients with completely resected non‐small‐cell lung cancer (NSCLC). We have evaluated the predictive value of mutations in TP53, encoding the tumour suppressor p53, in the International Adjuvant Lung Cancer Trial (IALT), a randomized trial of adjuvant cisplatin‐based chemotherapy against observation. TP53 (exons 4–8) was sequenced in 524 archived specimens of IALT patients with a median follow‐up of 7.5 years. Predictive analyses were based on Cox models adjusted for clinical and pathological variables. P‐values ≤0.01 were considered as significant. Mutations were detected in 221 patients (42%) and had no predictive value for the effect of chemotherapy (interaction between TP53 and treatment: p = 0.17 for Overall Survival (OS); p = 0.06 for Disease‐Free Interval, (DFS)). However, among patients with mutations, outcome appeared worse in treatment compared to observation arms (HR for OS = 1.36 (95% CI [0.97–1.31), p = 0.08; DFS = 1.40 (95% CI [1.01–1.95]), p = 0.04). When grouping mutations into classes according to predicted effects on protein structure, the tendency towards worse outcomes was restricted to “structure” mutations affecting residues of the hydrophobic core that are not located at the p53 protein‐DNA interface (HR for death in this class vs wild‐type T53 = 1.66; 95% CI [1.10–2.52], p = 0.02). Overall, TP53 mutations are not significant predictors of outcome in this trial of cisplatin‐based chemotherapy, although a specific class of structural mutations may be associated with a tendency towards worse outcomes upon treatment.