Nearest Neighbor Classification Rule for Mixtures of Discrete and Continuous Random Variables

In this paper very simple nonparametric classification rule for mixtures of discrete and continuous random variables is described. It is based on the method of nearest neighbor proposed by Cover and Hart (1967). The bounds on the limit of the nearest neighbor rule risks are given. Both lower and upper bound depend on the Bayes risk and the loss function. Finally the method is compared with other existing methods on some practical data set.     

Specificity of sterol-glucosylating enzymes from Sinapis alba and Physarum polycephalum.

1. Sinapis alba L. seedlings contain glycosyltransferase catalyzing the synthesis of sterol glucosides in the presence of UDPglucose as sugar donor. The major activity occurs in the membranous fraction sedimenting at 300--9000 x g. Successive treatment of the particulate enzyme fraction with acetone and Triton X-100 affords a soluble glucosyltransferase preparation which can be partly purified by gel filtration on Sephadex G-150. Molecular weight of the glucosyltransferase is 1.4 . 10(5). Apparent Km values for UDPglucose and sitosterol are 8.0 . 10(-5) M and 5.0 . 10(-6) M, respectively. 2. Comparison was made of the S. alba glucosyltransferase with a similar sterol-glucosylating enzyme isolated from non-photosynthesizing organism Physarum polycephalum (Myxomycetes). UDPglucose was the most efficient glucose donor in both cases but the enzyme from Ph. polycephalum can also utilize CDPglucose and TDPglucose. Glucose acceptors are, in case of both enzymes, sterols containing a beta-OH group at C-3 and a planar ring system (5 alpha-H or double bond at C-5). The number and position of double bonds in the ring system and in the side chain, as well as the presence of additional alkyl groups in the side chain at C-24 are of secondary importance. 3. The present results indicate that both enzymes can be regarded as specific UDPglucose:sterol glucosyltransferases. Certain differences in their specificity towards donors and acceptors of the glucosyl moiety suggest, however, a different structure of the active sites in both enzymes.     

An Autoradiographic Study of the Developing Parietal Cell Population In Neonatal Pigs

Abstract. Autoradiographic labelling using tritiated thymidine ([³H]TdR) was used to examine the pattern of development of gastric parietal cells in newborn pigs. Specific objectives were to establish sites in the gland where cells with a characteristic parietal cell morphology first appear, the extent of their migration or displacement, and the kinetics of any development and migration that occurs. Five newly-born littermate piglets were given a virtually continuous label of [³H]TdR over 24 hr, sacrificed at 1, 3, 5, 7 and 10 days thereafter, and samples of the gastric mucosa taken. the percentage of labelled parietal cells as a function of position in the oxyntic gland was measured for each pig. A generalized log linear model was fitted to the data using the statistical package GLIM, confirming a significant trend for labelled cells to occupy higher sites in the oxyntic gland as the time since labelling of cells increased. Goodness of fit tests showed that the trend effect was highly unlikely to be due to the variability of cell distribution from animal to animal. the dynamics of the parietal cell population and the strengths of GLIM for analysing cell labelling data are discussed.