Sparse computation for large-scale binary classification

Well-known data mining algorithms rely on inputs in the form of pairwise similarities between objects. For large datasets it is computationally impossible to perform all pairwise comparisons. We therefore propose a novel approach that uses approximate Principal Component Analysis to efficiently identify groups of similar objects. The effectiveness of the approach is demonstrated in the context of binary classification using the supervised normalized cut as a classifier. For large datasets from the UCI repository, the approach significantly improves run times with minimal loss in accuracy.

Index Tracking Using Data-Mining Techniques and Mixed-Binary Linear Programming

Index tracking has become one of the most common strategies in asset management. The index-tracking problem consists of constructing a portfolio that replicates the future performance of an index by including only a subset of the index constituents in the portfolio. Finding the most representative subset is challenging when the number of stocks in the index is large. We introduce a new three-stage approach that at first identifies promising subsets by employing data-mining techniques, then determines the stock weights in the subsets using mixed-binary linear programming, and finally evaluates the subsets based on cross validation. The best subset is returned as the tracking portfolio. Our approach outperforms state-of-the-art methods in terms of out-of-sample performance and running times.

Panels of cytokines and other secretory proteins as potential biomarkers of ovarian endometriosis.

Endometriosis is a gynecologic disease that is characterized by nonspecific symptoms and invasive diagnostics. To date, there is no adequate noninvasive method for the diagnosis of endometriosis. Although more than 100 potential biomarkers have been investigated in blood and/or peritoneal fluid, none of these has proven useful in clinical practice. The aim to find a suitable panel of biomarkers that would allow noninvasive diagnosis thus remains of interest. We evaluated the concentrations of 16 cytokines and other secretory proteins in serum and peritoneal fluid of 58 women with ovarian endometriosis (cases) and 40 healthy women undergoing sterilization or patients with benign ovarian cysts (controls) using multiplexed double fluorescence-based immunometric assay platform and enzyme-linked immunosorbent assay. Significantly higher concentrations of glycodelin-A were shown in serum, and significantly higher levels of glycodelin-A, IL-6, and IL-8, and lower levels of leptin were measured in the peritoneal fluid of cases versus controls. In serum, the best performance was shown by models that included the ratio of leptin/glycodelin-A and the ratio of ficolin 2/glycodelin-A, whereas in the peritoneal fluid the best models included the ratio of biglycan/leptin, regulated on activation normal T-cell expressed and secreted/IL-6 and ficolin-2/glycodelin-A, and IL-8 per milligram of total protein, all in combination with age. The models using serum and peritoneal fluid distinguished between ovarian endometriosis patients and controls regardless of the menstrual cycle phase with relatively high sensitivity (72.5% to 84.2%), specificity (78.4% to 91.2%), and area under the curve (0.85 to 0.90).

Binary recombinase systems for high-resolution conditional mutagenesis.

Conditional mutagenesis using Cre recombinase expressed from tissue specific promoters facilitates analyses of gene function and cell lineage tracing. Here, we describe two novel dual-promoter-driven conditional mutagenesis systems designed for greater accuracy and optimal efficiency of recombination. Co-Driver employs a recombinase cascade of Dre and Dre-respondent Cre, which processes loxP-flanked alleles only when both recombinases are expressed in a predetermined temporal sequence. This unique property makes Co-Driver ideal for sequential lineage tracing studies aimed at unraveling the relationships between cellular precursors and mature cell types. Co-InCre was designed for highly efficient intersectional conditional transgenesis. It relies on highly active trans-splicing inteins and promoters with simultaneous transcriptional activity to reconstitute Cre recombinase from two inactive precursor fragments. By generating native Cre, Co-InCre attains recombination rates that exceed all other binary SSR systems evaluated in this study. Both Co-Driver and Co-InCre significantly extend the utility of existing Cre-responsive alleles.

PANELS: Stata module to count panels and apply a command to panel units

panels provides a quick way to count the number of panels (groups) in a dataset and display some basic information about the sizes of the panels. Furthermore, -panels- can be used as a prefix command to other Stata commands to apply them to panel units instead of individual observations. This is useful, for example, if you want to compute frequency distributions or summary statistics for panel characteristics.

FAIRLIE: Stata module to generate nonlinear decomposition of binary outcome differentials

fairlie computes the nonlinear decomposition of binary outcome differentials proposed by Fairlie (1999, 2003).