Search

Background: A standard task in pharmacogenomics research is identifying genes that may be involved in drug response variability, i.e., pharmacogenes. Because genomic experiments tended to generate many false positives, computational approaches based on the use of background knowledge have been proposed. Until now, only molecular networks or the biomedical literature were used, whereas many other resources are available., Method: We propose here to consume a diverse and larger set of resources using linked data related either to genes, drugs or diseases. One of the advantages of linked data is that they are built on a standard framework that facilitates the joint use of various sources, and thus facilitates considering features of various origins. We propose a selection and linkage of data sources relevant to pharmacogenomics, including for example DisGeNET and Clinvar. We use machine learning to identify and prioritize pharmacogenes that are the most probably valid, considering the selected linked data. This identification relies on the classification of gene-drug pairs as either pharmacogenomically associated or not and was experimented with two machine learning methods -random forest and graph kernel-, which results are compared in this article., Results: We assembled a set of linked data relative to pharmacogenomics, of 2,610,793 triples, coming from six distinct resources. Learning from these data, random forest enables identifying valid pharmacogenes with a F-measure of 0.73, on a 10 folds cross-validation, whereas graph kernel achieves a F-measure of 0.81. A list of top candidates proposed by both approaches is provided and their obtention is discussed.

The postoperative peritonitis (POP) remains formidable conditions due to a high mortality rate of between 20 and 80%. The purpose of this study is to identify risk factors for mortality. This study is a retrospective, descriptive analysis carried out over a period of 09 years (1/1/2003 - 30/11/2011) and interesting 102 patients supported for POP following general surgery. Achieved in department of General Surgery B Charles Nicolle hospital Tunis. The parameters measured included epidemiological data, data related to the Initial Surgical Intervention and reoperation for POP, terms of management and evolution. Bacteriological data were also seized. The incidence of POP was 0.90%. The average age of our patients was 58 +/- 19 years with a sex ratio of 1.08. Forty-seven percent of our patients belonged to the ASAII class. The initial operation was performed urgently in 49 patients (48%) with a majority belonging to the class II Altemeier (49.01%). Colorectal pathology (373%) and hepatobiliay (176%) were the most frequent reasons for the initial intervention. The frequency of clinical signs were fever (75.5%), hypothermia (6.9%), abdominal pain (725%), abdominal distension (46.1%), productive gastric aspiration (30.4%), abdominal defense (25.5%), externalizing the digestive fluid (25.5%), vomiting (19.6%), diarrhea (12.7%), tachycardia (569%), oliguria (42.2%), respiratory failure (40.2%), hypotension (35.3%), neuropsychiatric disorders ( 23.5%) and jaundice (69%). The treatment period was 2.95 +/- 3.16 days. The surgical recovery time was 78 days +/- 5.66. At the time of reoperation, the APACHE II score was 8.43 +/- 6.26 and 25.1 +/- MPI score 8.53. The POP was generalized in 52.9% of cases with purulent peritoneal fluid in 51% of cases. The most common cause was the dropping of the anastomosis (59.8%). Empirical antibiotic therapy was appropriate in 69.44 % of cases. The mortality rate was 39.2%. Multivariate analysis using multiple logistic regression identified the following factors as independent mortality factors: Age > or = 60 years (RR = 6.089), multiple organ failure (RR = 18.019), non-appropriate empiric antibiotic therapy (RR = 6.541), stercoral aspect of peritoneal fluid (RR 6.412). Despite a low frequency, the POP are burdened with a high mortality rate. The improved prognosis requires early diagnosis in order to allow a appropriate load medicosurgical support before the installation of multiorgan failure syndrome. Among the independent factors associated with mortality that we have identifed, the not adapted empiric antibiotic therapy is the main factor on which we can act.

During the postoperative peritonitis (PPO) the main stay of treatment is the choice of probabilistic antibiotictherapy, it is also the main prognostic factor The aim of our study was to identify anappropriate antibiotic protocol to the current ecology of our unit. It was a retrospective study including 102 patients over a period of 09 years from 1 January 2003 to 3O November 2011. All of them are supported for the treatments off postoperative peritonitis in surgical intensive care unit of a service of general surgery a university hospital Charles Nicolle of Tunis. All bacteriological data (germs and sensitivity), and the terms of therapeutic modality for the empirical antibiotic therapy were listed. The incidence of PPO was Q90%.The average age of our patients was 57 +/- 18 years. The sex ratio was 1.08. One hundred and seven (107) microorganisms were isolated from 72 samples (44 microbial mono, 28 multi microbial). The frequency of gram-positive cocci (GPC) was 16.82%, the Gram-negative bacilli (BGN) was 82.2%. The Enterobacteriaceae have proved particularity resistant. Thus, the ampicillin resistance was 87.14%, that the C3G was 33.80%, the Piperacillin to Tazobactam combination, was 36.5% and that the association Ticarcillin-clavulanic acid was 43.6%. For non-fermenting BGN, Pseudomonas aeruginosa was sensitive to ticarcillin in 80% of cases, to ceftazidime in 66.6% of cases, PiperacillinTazobactam--in 71.43% of cases, imipenem in 85 72% of cases, colimycin in 100% of cases and Amiklin in 71.43% of cases. For CGP, enterococci were resistant to ampicillin in 50% of cases and vancomycin in 0% of cases. The majority of patients received triple antibiotic therapy (59.8%) or combination therapy (34.3%). The main associations were: cefotaxime + Gentamycin + Metronidazole (35.2%), Amikacin Imipenem + + Metronidazole (12.7%), Imipenem + amikacin (9.8%), Piperacillin / Tazobactam + amikacin (9.8%) + amikacin and ertapenem (5.88%). Probabilitic antibiotic therapy was addapted in 69.4% of cases. The average duration of the prescribed antibiotic was 11 days +/- 6 days. The mortality rate was 39.2%, was 32.23 days. The isolated microorganisms are those of the intestinal flora which is generally changed and thus the bacteria are selected then are multidrug resistant. Prescribing antibiotics should consider probabilistic. Thus, Imipenem-Amiklin combination seems appropriate to our ecology. This empiric antibiotic therapy is secondarily adapted to the results of susceptibility testing to limit the selection of multi-resistant organisms.