Assessment of Optimal Selected Prognostic Factors

The identification and assessment of prognostic factors is one of the major tasks in clinical research. The assessment of one single prognostic factor can be done by recently established methods for using optimal cutpoints. Here, we suggest a method to consider an optimal selected prognostic factor from a set of prognostic factors of interest. This can be viewed as a variable selection method and is the underlying decision problem at each node of various tree building algorithms. We propose to use maximally selected statistics where the selection is defined over the set of prognostic factors and over all cutpoints in each prognostic factor. We demonstrate that it is feasible to compute the approximate null distribution. We illustrate the new variable selection test with data of the German Breast Cancer Study Group and of a small study on patients with diffuse large B‐cell lymphoma. Using the null distribution for a p‐value adjusted regression trees algorithm, we adjust for the number of variables analysed at each node as well. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Evidence for top predator control of a grazing ecosystem

The importance of top predators in controlling ecological processes in large, intact ecosystems is unclear. In grasslands that support abundant ungulates, top–down control by predators may be particularly important, because of the tight biogeochemical linkages of ungulate prey with plants and soil microbes. Here, I examined the effects of the recent reintroduction of the gray wolf Canis lupus on ecosystem processes in Yellowstone National Park, where herds of grazing ungulates previously have been shown to stimulate several processes, including soil net nitrogen (N) mineralization. Rates of ungulate grazing intensity and soil net N mineralization were compared before and after wolf reintroduction in grasslands ranging five‐fold in aboveground production. Grazing intensity and grassland net N mineralization declined after wolf reintroduction, a likely partial function of fewer ungulates; wolf predation has been one of several factors implicated in causing the decline in Yellowstone ungulates. In addition, the spatial pattern of grazing and net N mineralization changed after reintroduction. A shift in the spatial patterns of grazer‐associated processes is consistent with a growing body of work indicating that wolves have changed habitat use patterns of ungulates in Yellowstone National Park. These findings suggest widespread wolf effects on ungulate prey, plants, and microbial activity that have spatially reorganized grassland energy and nutrient dynamics in Yellowstone Park.     

Elevated levels of plasma lipofuscins in patients with chronic renal failure

The fluorescence excitation and emission spectra observed in plasma from patients with chronic renal failure were reproduced by the generation of soluble lipofuscins in normal plasma samples by incubation with mixtures of L-dopa, dopamine, L-norepinephrine, L-epinephrine, 3-hydroxy-DL-kynurenine and 3-hydroxy-anthranilic acid for 24 h at 37 degrees C. Relative fluorescence intensity measurements consistently showed elevated plasma levels of the soluble lipofuscins in chronic renal failure: the means (n = 27) were 73.9 +/- 33.4 (SD) and 71.1 +/- 14.8 at emissions 413 nm and 445 nm respectively, in contrast to those of normal plasma samples (n = 11), 18.2 +/- 5.3 and 23.1 +/- 5.6. The maximum or shoulder at approximately 413 nm represents soluble lipofuscin that can be generated from 3-hydroxyanthranilic acid and the maximum or shoulder at approximately 445 nm represents soluble lipofuscins derived from the precursors listed above and probably from other related precursors. Gravimetric measurements also showed elevated levels of melanins in the plasma samples of patients with chronic renal failure: 2.72 +/- 0.38 mg/ml (n = 16), as compared to normal values: 1.70 +/- 0.10 mg/ml (n = 6). In individual patients haemodialysis reduced the fluorescence intensities to a range of 65-99% and the melanin levels to a range of 86-99% of the pre-dialysis values.     

Proceedings of the fifty-fourth Annual Meeting of the American Association of physical anthropologists to be held in Knoxville,Tennessee—Holiday Inn World's fair April 10–13, 1985

Empirical analyses and models of the lineal effects of fission indicate that considerable genetic differentiation may occur at the time of group formation, thus confusing the usual positive relationship between historical affiliation and genetic differentiation. We analyze the effects of fission pattern on variation in highly heritable morphological traits among eight social groups on Cayo Santiago. The analysis is performed using general network autocorrelation methods that quantitatively and directly measure the amount of variation in social group mean morphology that can be explained by fission. All of the fission autocorrelation coefficients are strongly negative, indicating that groups most recently formed by fission are most dissimilar. Also, most of the variation between groups can be explained by the fission pattern, indicating that lineal fission is the most important process generating between-group variation on Cayo Santiago.     

Spin probe clustering in human erythrocyte ghosts

Summary A model has been developed for 5-nitroxide stearate, I(12,3), distribution in human erythrocyte ghosts which accurately predicts ESR spectral alterations observed with increased probe/total lipid (P/L) at 37°C. This spin probe occupies a class of high-affinity, noninteracting sites at low loading. Saturation occurs with increasing probe concentration, and, at higher loading, the probe inserts itself at initially dilute sites to form membranebound clusters of variable size. No low probe remains at high P/L where all I(12,3) clusters in a concentrated phase. This model allows determination of the dilute/clustered probe ratio, and shows that I(12,3) segregates in erythrocytes at what might otherwise be considered low P/L (e.g., 1/359). These findings validate the earlier use of empirical parameters to estimate probe sequestration in biological membranes.