Global Negative Correlation Learning: A Unified Framework for Global Optimization of Ensemble Models

Ensembles are a widely implemented approach in the machine learning community and their success is traditionally attributed to the diversity within the ensemble. Most of these approaches foster diversity in the ensemble by data sampling or by modifying the structure of the constituent models. Despite this, there is a family of ensemble models in which diversity is explicitly promoted in the error function of the individuals. The negative correlation learning (NCL) ensemble framework is probably the most well-known algorithm within this group of methods. This article analyzes NCL and reveals that the framework actually minimizes the combination of errors of the individuals of the ensemble instead of minimizing the residuals of the final ensemble. We propose a novel ensemble framework, named global negative correlation learning (GNCL), which focuses on the optimization of the global ensemble instead of the individual fitness of its components. An analytical solution for the parameters of base regressors based on the NCL framework and the global error function proposed is also provided under the assumption of fixed basis functions (although the general framework could also be instantiated for neural networks with nonfixed basis functions). The proposed ensemble framework is evaluated by extensive experiments with regression and classification data sets. Comparisons with other state-of-the-art ensemble methods confirm that GNCL yields the best overall performance.

Floating Search Methodology for Combining Classification Models for Site Recognition in DNA Sequences

Recognition of the functional sites of genes, such as translation initiation sites, donor and acceptor splice sites and stop codons, is a relevant part of many current problems in bioinformatics. The best approaches use sophisticated classifiers, such as support vector machines. However, with the rapid accumulation of sequence data, methods for combining many sources of evidence are necessary as it is unlikely that a single classifier can solve this problem with the best possible performance. A major issue is that the number of possible models to combine is large and the use of all of these models is impractical. In this paper we present a methodology for combining many sources of information to recognize any functional site using "floating search", a powerful heuristics applicable when the cost of evaluating each solution is high. We present experiments on four functional sites in the human genome, which is used as the target genome, and use another 20 species as sources of evidence. The proposed methodology shows significant improvement over state-of-the-art methods. The results show an advantage of the proposed method and also challenge the standard assumption of using only genomes not very close and not very far from the human to improve the recognition of functional sites.

Now is the Time for a Postracial Medicine: Biomedical Research, the National Institutes of Health, and the Perpetuation of Scientific Racism

The consideration of racial differences in the biology of disease and treatment options is a hallmark of modern medicine. However, this time-honored medical tradition has no scientific basis, and the premise itself, that is, the existence of biological differences between the commonly known races, is false inasmuch as races are only sociocultural constructions. It is time to rid medical research of the highly damaging exercise of searching for supposed racial differences in the biological manifestations of disease. The practice not only condoned but required by the National Institutes of Health (NIH) of utilizing racial identification as a demographic characteristic with assumed biological implications is at best badly flawed, and at worst unintentionally contributes to perpetuating the fallacy of natural differences between persons of different skin color, which has been used in the past to advance the cause of racial discrimination.

Nephrogenic systemic fibrosis: incidence, associations, and effect of risk factor assessment--report of 33 cases

PURPOSE

To describe the presentation and clinical course of patients with nephrogenic systemic fibrosis (NSF) at a large acute-care hospital, to evaluate the overall incidence of NSF, and to assess the effect of a hospital-wide policy regarding gadolinium-based contrast agent (GBCA) use on NSF incidence.

MATERIALS AND METHODS

A review of all cases of NSF observed at an institution from 2003 to 2008 was conducted. This HIPAA-compliant study was approved by the institutional review board. The informed consent requirement was waived. Demographics, medical history, and associated conditions were recorded. Radiologic procedures were evaluated if they were performed within 1 year prior to NSF onset. GBCA use was assessed by checking the electronic database for each procedure. The incidence of NSF was compared before and after implementation of an institutional policy designed to assess risk of NSF prior to GBCA use.

RESULTS

All 33 patients with NSF (mean age, 49 years; age range, 15-78 years) had advanced renal failure (estimated glomerular filtration rate < 15 mL/min/1.73 m(2)) when the GBCA was injected. Twenty-six patients had severe chronic or end-stage renal disease, and seven had acute renal failure. The mean interval between contrast material injection and NSF onset was 29 days +/- 25 (standard deviation) (range, 4-112 days). The overall incidence of NSF was 36.5 cases per 100,000 gadolinium-enhanced magnetic resonance (MR) examinations between 2003 and 2006 and four cases per 100,000 gadolinium-enhanced MR examinations between 2007 and 2008 after screening for NSF risk was instituted (Fisher exact test, P = .001). Five patients developed NSF in the peritransplant period, and four underwent a catheter-based radiographic procedure with administration of a GBCA.

CONCLUSION

Common associations of GBCA MR imaging and NSF were acute and severe chronic renal failure and liver or renal transplantation. Screening procedures performed before MR imaging to determine which patients were at risk of developing NSF appear to reduce the incidence of this complication and further support the belief that NSF is associated with GBCA administration.

Human kinetics of orally and intravenously administered low-dose 1,2-(13)C-dichloroacetate

Dichloroacetate (DCA) is a putative environmental hazard, owing to its ubiquitous presence in the biosphere and its association with animal and human toxicity. We sought to determine the kinetics of environmentally relevant concentrations of 1,2-(13)C-DCA administered to healthy adults. Subjects received an oral or intravenous dose of 2.5 microg/kg of 1,2-(13)C-DCA. Plasma and urine concentrations of 1,2-(13)C-DCA were measured by a modified gas chromatography-tandem mass spectrometry method. 1,2-(13)C-DCA kinetics was determined by modeling using WinNonlin 4.1 software. Plasma concentrations of 1,2-(13)C-DCA peaked 10 minutes and 30 minutes after intravenous or oral administration, respectively. Plasma kinetic parameters varied as a function of dose and duration. Very little unchanged 1,2-(13)C-DCA was excreted in urine. Trace amounts of DCA alter its own kinetics after short-term exposure. These findings have important implications for interpreting the impact of this xenobiotic on human health.

Clinical assay of N-duopropenide alcohol solution on hand application in newborn and pediatric intensive care units: control of an outbreak of multiresistant Klebsiella pneumoniae in a newborn intensive care unit with this measure

Outbreaks of gram-negative colonization (generally by antibiotic-resistant enterobacteria) are common in newborn intensive care units (NICUs), and control methods are not always effective. We studied the effectiveness of an alcohol solution of N-duopropenide (NDP) in vivo (germicidal effect on flora of teams in the NICU and the pediatric intensive care unit vs handwashing with nonantiseptic soap) and its effect on the control of a multiresistant (MR) Klebsiella pneumoniae outbreak in our NICU that had persisted for 13 months, despite the use of classic control measures. For educational purposes, we also performed 4 prevalence studies of microbial hand flora in NICU staff (two before and two after introducing NDP). The alcohol solution of NDP was highly germicidal in vivo, destroying microorganisms better than classic handwashing on the hands of 69 health care staff of our NICU and PICU. The flora in both units was reduced from an average of 63% to an average of 95%. Application of this disinfectant to the hands of health care workers after handling newborns helped to eliminate the MR Klebsiella strain in our NICU, (relative risk compared with the period preceding use of the disinfectant: 8.6, with 95% confidence intervals, 4.8-145.5). Four prevalence studies of hand microbial contamination, before and after NDP introduction in the NICU, showed a significant reduction of enterobacteriaceae, mainly MR K pneumoniae, in health care workers. In conclusion, NDP in alcohol was very effective in vivo. It proved to be a useful complementary measure to handwashing and reduced exogenous microorganism transmission in a unit with a heavy patient-care workload.

Purification of phosphate-dependent glutaminase from isolated mitochondria of Ehrlich ascites-tumour cells

Phosphate-dependent glutaminase was purified to homogeneity from isolated mitochondria of Ehrlich ascites-tumour cells. The enzyme had an Mr of 135,000 as judged by chromatography on Sephacryl S-300. SDS/polyacrylamide-gel electrophoresis displayed two protein bands, with Mr values of 64,000 and 56,000. Two major immunoreactive peptides of Mr values of 65,000 and 57,000 were found by immunoblot analysis using anti-(rat kidney glutaminase) antibodies. The concentration-dependences for both glutamine and phosphate were sigmoidal, with S0.5 values of 7.6 mM and 48 mM, and Hill coefficients of 1.5 and 1.6, respectively. The glutaminase pH optimum was 9. The activation energy of the enzymic reaction was 58 kJ/mol. The enzyme showed a high specificity towards glutamine. A possible explanation for the different kinetic behaviour found for purified enzyme and for isolated mitochondria [Kovacević (1974) Cancer Res. 34, 3403-3407] should be that a conformational change occurs when the enzyme is extracted from the mitochondrial inner membrane.