Search

Background: The quality of predictive modeling in biomedicine depends on the amount of data available for model building., Objective: To study the effect of combining microarray data sets on feature selection and predictive modeling performance., Methods: Empirical evaluation of stability of feature selection and discriminatory power of classifiers using three previously published gene expression data sets, analyzed both individually and in combination., Results: Feature selection was not robust for the individual as well as for the combined data sets. The classification performance of models built on individual and combined data sets was heavily dependent on the data set from which the features were extracted., Conclusion: We identified volatility of feature selection as contributing factor to some of the problems faced by predictive modeling using microarray data.

Motivation: The discovery of new and unexpected biomarkers in cardiovascular disease is a highly data-driven process that requires the complementary power of modern metabolite profiling technologies, bioinformatics and biostatistics. Clinical biomarkers of early myocardial injury are lacking. A prospective biomarker cohort study was carried out to identify, categorize and profile kinetic patterns of early metabolic biomarkers of planned myocardial infarction (PMI) and spontaneous (SMI) myocardial infarction. We applied a targeted mass spectrometry (MS)-based metabolite profiling platform to serial blood samples drawn from carefully phenotyped patients undergoing alcohol septal ablation for hypertrophic obstructive cardiomyopathy serving as a human model of PMI. Patients with SMI and patients undergoing catheterization without induction of myocardial infarction served as positive and negative controls to assess generalizability of markers identified in PMI., Results: To identify metabolites of high predictive value in tandem mass spectrometry data, we introduced a new feature selection method for the categorization of metabolic signatures into three classes of weak, moderate and strong predictors, which can be easily applied to both paired and unpaired samples. Our paradigm outperformed standard null-hypothesis significance testing and other popular methods for feature selection in terms of the area under the receiver operating curve and the product of sensitivity and specificity. Our results emphasize that this new method was able to identify, classify and validate alterations of levels in multiple metabolites participating in pathways associated with myocardial injury as early as 10 min after PMI., Availability: The algorithm as well as supplementary material is available for download at: www.umit.at/page.cfm?vpath=departments/technik/iebe/tools/bi

Motivation: Prostate cancer is the most prevalent tumor in males and its incidence is expected to increase as the population ages. Prostate cancer is treatable by excision if detected at an early enough stage. The challenges of early diagnosis require the discovery of novel biomarkers and tools for prostate cancer management., Results: We developed a novel feature selection algorithm termed as associative voting (AV) for identifying biomarker candidates in prostate cancer data measured via targeted metabolite profiling MS/MS analysis. We benchmarked our algorithm against two standard entropy-based and correlation-based feature selection methods [Information Gain (IG) and ReliefF (RF)] and observed that, on a variety of classification tasks in prostate cancer diagnosis, our algorithm identified subsets of biomarker candidates that are both smaller and show higher discriminatory power than the subsets identified by IG and RF. A literature study confirms that the highest ranked biomarker candidates identified by AV have independently been identified as important factors in prostate cancer development., Availability: The algorithm can be downloaded from the following http://biomed.umit.at/page.cfm?pageid=516.

Objective: To improve the calibration of logistic regression (LR) estimates using local information., Background: Individualized risk assessment tools are increasingly being utilized. External validation of these tools often reveals poor model calibration., Methods: We combine a clustering algorithm with an LR model to produce probability estimates that are close to the true probabilities for a particular case. The new method is compared to a standard LR model in terms of calibration, as measured by the sum of absolute differences (SAD) between model estimates and true probabilities, and discrimination, as measured by area under the ROC curve (AUC)., Results: We evaluate the new method on two synthetic data sets. SADs are significantly lower (p < 0.0001) in both data sets, and AUCs are significantly higher in one data set (p < 0.01)., Conclusion: The results suggest that the proposed method may be useful to improve the calibration of LR models.

A novel computational approach, termed Search for Modified Peptides (SeMoP), for the unrestricted discovery and verification of peptide modifications in shotgun proteomic experiments using low resolution ion trap MS/MS spectra is presented. Various peptide modifications, including post-translational modifications, sequence polymorphisms, as well as sample handling-induced changes, can be identified using this approach. SeMoP utilizes a three-step strategy: (1) a standard database search to identify proteins in a sample; (2) an unrestricted search for modifications using a newly developed algorithm; and (3) a second standard database search targeted to specific modifications found using the unrestricted search. This targeted approach provides verification of discovered modifications and, due to increased sensitivity, a general increase in the number of peptides with the specific modification. The feasibility of the overall strategy has been first demonstrated in the analysis of 65 plasma proteins. Various sample handling induced modifications, such as beta-elimination of disulfide bridges and pyrocarbamidomethylation, as well as biologically induced modifications, such as phosphorylation and methylation, have been detected. A subsequent targeted Sequest search has been used to verify selected modifications, and a 4-fold increase in the number of modified peptides was obtained. In a second application, 1367 proteins of a cervical cancer cell line were processed, leading to detection of several novel amino acid substitutions. By conducting the search against a database of peptides derived from proteins with decoy sequences, a false discovery rate of less than 5% for the unrestricted search resulted. SeMoP is shown to be an effective and easily implemented approach for the discovery and verification of peptide modifications.

Background and Objectives: In a majority of all patients with systemic mastocytosis (SM) including those with mast cell leukemia (MCL), neoplastic mast cells (MC) display the D816V-mutated variant of KIT. The respective oncoprotein, KIT D816V, exhibits constitutive tyrosine kinase (TK) activity and has been implicated in malignant cell growth. Therefore, several attempts have been made to identify KIT D816V-targeting drugs., Design and Methods: We examined the effects of the novel TK-inhibitor dasatinib alone and in combination with other targeted drugs on growth of neoplastic MC., Results: Confirming previous studies, dasatinib was found to inhibit the TK activity of wild type (wt) KIT and KIT-D816V as well as growth and survival of neoplastic MC and of the MCL cell line, HMC-1. The growth-inhibitory effects of dasatinib in HMC-1 cells were found to be associated with a decrease in expression of CD2 and CD63. In addition, we found that dasatinib blocks KIT D816V-induced cluster-formation and viability in Ba/F3 cells. In drug combination experiments, dasatinib was found to co-operate with PKC412, AMN107, imatinib, and 2CdA in producing growth-inhibition and apoptosis in neoplastic MC. In HMC-1.1 cells lacking KIT D816V, all drug interactions were found to be synergistic in nature. By contrast, in HMC-1.2 cells exhibiting KIT D816V, only the combinations dasatinib+PKC412 and dasatinib+2CdA were found to produce synergistic effects., Interpretation and Conclusions: Combinations of targeted drugs may represent an interesting pharmacologic approach for the treatment of aggressive SM or MCL.

Objective: Aortic complications determine the life expectancy of most patients with Marfan syndrome. To find out whether there is heterogenous aortic involvement among patients and, if there is, to characterize aortic patterns and response to long-term beta-blocker therapy, we investigated aortic elastic properties before and during beta-blocker treatment., Methods: In 46 patients with Marfan syndrome (age, 17.4 +/- 11.1 years) and 46 healthy control subjects, ascending and descending aortic elastic parameters were determined noninvasively before and after 39 +/- 16 months of beta-blocker treatment with atenolol., Results: Aortic diameters and distensibility distinguished Marfan patients and controls with a sensitivity of 85% and a specificity of 87%. Cluster analysis revealed 4 patterns of aortic phenotypic expression: (1) reduced ascending aortic elasticity (46% of patients), (2) diminished ascending and descending aortic elasticity (17%), (3) minimal alterations of ascending and descending aortic elasticity (20%), and (4) reduced descending aortic elasticity (17%). During follow-up, aortic elastic properties improved in 21 (70%) of 30 patients and deteriorated in 9 (30%) irrespective of beta-blocker dosage. Improvement was observed in 100% of patients (n = 7; age, 5.3 +/- 4.2 years) with end-diastolic aortic root diameters between 20 and 30 mm and in 61% of patients (14/23; age, 20.5 +/- 10.0 years) with root diameters between 30 and 52 mm., Conclusions: Aortic elastic parameters distinguish between patients with Marfan syndrome and healthy controls and show the pattern of regional aortic involvement. Improvement or deterioration during follow-up can influence therapeutic decisions to prevent aortic dissection and rupture. Young age, small root diameter, and high distensibility are favorable prognostic factors.

Background: In patients with Marfan syndrome, progressive aortic dilation implicates a still-unpredictable risk of life-threatening aortic dissection and rupture. We sought to quantify aortic wall dysfunction noninvasively, determine the diagnostic power of various aortic parameters, and establish a diagnostic model for the early detection of aortic abnormalities associated with Marfan syndrome., Methods: In 19 patients with Marfan syndrome (age, 17.7 +/- 9.5 years) and 19 age- and sex-matched healthy control subjects, computerized ascending and abdominal aortic wall contour analysis with continuous determination of aortic diameters was performed out of transthoracic M-mode echocardiographic tracings. After simultaneous oscillometric blood pressure measurement, aortic elastic properties were determined automatically., Results: The following ascending aortic elastic parameters showed statistically significant differences between the Marfan group and the control group: (1) decreased aortic distensibility ( P < .001), (2) increased wall stiffness index ( P < .01), (3) decreased systolic diameter increase ( P < .01), and (4) decreased maximum systolic area increase ( P < .001). The diagnostic power of all investigated parameters was tested by single logistic regression models. A multiple logistic regression model including solely aortic parameters yielded a sensitivity of 95% and a specificity of 100%., Conclusions: In young patients with Marfan syndrome, a computerized image-analyzing technique revealed decreased aortic elastic properties expressed by parameters showing high diagnostic power. A multiple logistic regression model including merely aortic parameters can serve as useful predictor for Marfan syndrome.