Variable selection in nonlinear function-on-scalar regression

We develop a new method for variable selection in a nonlinear additive function-on-scalar regression (FOSR) model. Existing methods for variable selection in FOSR have focused on the linear effects of scalar predictors, which can be a restrictive assumption in the presence of multiple continuously measured covariates. We propose a computationally efficient approach for variable selection in existing linear FOSR using functional principal component scores of the functional response and extend this framework to a nonlinear additive function-on-scalar model. The proposed method provides a unified and flexible framework for variable selection in FOSR, allowing nonlinear effects of the covariates. Numerical analysis using simulation study illustrates the advantages of the proposed method over existing variable selection methods in FOSR even when the underlying covariate effects are all linear. The proposed procedure is demonstrated on accelerometer data from the 2003–2004 cohorts of the National Health and Nutrition Examination Survey (NHANES) in understanding the association between diurnal patterns of physical activity and demographic, lifestyle, and health characteristics of the participants.     

豚鼠耳蜗电图慢成分的研究

选用0.8-150Hz的带通滤波,从豚鼠圆窗记录出一负一正相慢电位.实验结果表明.其对0.5kHz短音的反应阈为729dB nHL.较耳蜗电图快成分低38.27dB nHL.通过离断听觉传导径路不同水平对该电位的观察,表明它是毛细胞及听神经电反应的慢成分.而且可能受听觉传出神经的调控.     

5个番石榴品种果实食用品质和香气特征分析

为比较番石榴(Psidium guajava)不同品种果实的外观、营养价值和香气特征,采用国家标准方法,对5个番石榴品种(‘红宝石’、‘粉红蜜’、‘西瓜’、‘水蜜’、本地种)果实的外观和营养成分进行测定,采用顶空气质联用(HS-GC-MS)技术对5个品种果实的香气成分进行测定。结果表明,不同品种果实外观及营养成分差异明显。‘水蜜’的果形指数最低,果实扁圆形,种籽最少;大多数糖类物质(总糖、还原糖、蔗糖等)、糖酸比、总酚含量最高;果糖、VC和总黄酮含量位居第2;总酸、脂肪、粗纤维、灰分含量最低。主成分分析表明,‘水蜜’作为鲜食水果的食用品质最高。己醛和石竹烯是红肉型品种的特征风味物质,3-羟基-2-丁酮是白肉型品种的特征香气成分。     

两阶抽样中比率型估计量的平均精度

本文给出了两阶抽样中总体均值的比率型估计量的平均精度,它当样本容量充分大时主项不劣于无偏估计量的平均精度．     

Measures of nematode community structure and sources of variability among and within agricultural fields

Whole nematode communities, extracted from soil samples taken from agricultural fields, were enumerated by taxonomic family and trophic group (i.e., bacterivores, fungivores, omnivores, plant-parasites, and predators) to evaluate nematode community structure as an indicator for monitoring ecological condition of soil. No differences were found in mixing treatments or methods of packing or shipping samples. However, extraction using Cobb's sifting and gravity method, followed by sucrose centrifugation, gave greater recovery of free-living nematodes than elutriation followed by sucrose centrifugation. Population means and variance of the sampled area were similar when sampled using different strategies for collecting soil samples within fieds, including several patterns, directions and repetitions of transects. Components of variation associated with ratios among the five trophic groups of nematodes and selected indices of community structure were quantified as variation among regions, among counties, among agricultural fields (2-ha area), among transects within agricultural fields, and within composite soil samples. The variance component for'within composite soil samples' was relatively large compared to the other components of variance. Variation within composite soil samples was less for maturity indices (based on life-history strategy characteristics), ratio of bacterivores to plant-parasites, sum of bacterivores and fungivores, populations of plant-parasites, and populations of bacterivores than for trophic diversity indices, populations of fungivores, populations of omnivores, populations of predators, or the ratio of fungivores to bacterivores. With a single composite sample per field, the ability to differentiate ecological condition of soils among fields within a region improved if the variance among and within fields exceeded the variance within composite samples. Given the variance components, power curves indicated that detection of a 10% change (with 0.8 power) in the ecological condition of soils within a region between two time periods would require sampling a minimum of 25 and 50 fields with one composite soil sample analyzed per field for the maturity and trophic diversity index, respectively. More than 100 fieldsper region would be required to detect temporal change in populations of individual trophic groups. Biplots of maturity indices, but not of trophic diversity or populations of individual trophic groups, identified clear differences among fields. Thus, maturity indices, which differentiated among sampling sites better and more efficiently than trophic diversity indices or measures based on populations of individual trophic groups, may be appropriate for use in a regional and/or national monitoring program.     

Genetic control of PI and GC variants in the American Mink

Genetic polymorphism of the serum α-protease inhibitor (PI) and group-specific component (GC) in minks was revealed using one-dimensional polyacrylamide gel electrophoresis and immunoblotting. Two codominant alleles were identified at each of the two loci. The data ruled out the possibility of any linkage between the PI, GC and the coat colour gene Crystal ( Cr ).     

Studies on well-coupled Photosystem-I-enriched subchloroplast vesicles. Electron transfer by b- and c-type cytochromes in relation to the origin of the ‘slow’ electric potential component

Cytochrome redox changes and electric potential generation are kinetically compared during cyclic electron transfer in Photosystem-I-enriched and Photosystem-II-depleted subchloroplast vesicles (i.e., stroma lamellae membrane vesicles) supplemented with ferredoxin using a suitable electron donating system. In response to a single-turnover flash, the sequence of events is: (1) fast reduction of cytochrome b-563 (t_0.5 ≈ 0.5 ms) (2) oxidation of cytochrome c-554 (t_0.5 ≈ 2 ms), (3) slower reduction of cytochrome b-563 (t_0.5 ≈ 4 ms), (4) generation of the ‘slow’ electric potential component (t_0.5 ≈ 15–20 ms), (5) re-reduction of cytochrome c-554 (t_0.5 ≈ 30 ms) and (6) reoxidation of cytochrome b-563t_0.5 ≈ 90 ms). Per flash two cytochrome b-563 species turn over for one cytochrome c-554. These b-563 cytochromes are reduced with different kinetics via different pathways. The fast reductive pathway proceeds probably via ferredoxin, is insensitive to DNP-INT, DBMIB and HQNO and is independent on the dark redox state of the electron transfer chain. In contrast, the slow reductive pathway is sensitive to DNP-INT and DBMIB, is strongly delayed at suboptimal redox poising (i.e., low $\text{NADPH}\text{NADP}{}^{\mathrm{+}}$ ratio) and is possibly coupled to the reduction of cytochrome c-554. Each reductive pathway seems obligatory for the generation of about 50% of the slow electric potential component. Also cytochrome c-559_LP (LP, low potential) is involved in Photosystem-I-associated cyclic electron flow, but its flash-induced turnover is only observed at low preestablished electron pressure on the electron-transfer chain. Data suggest that cyclic electron flow around Photosystem I only proceeds if cytochrome b-559_LP is in the reduced state before the flash, and a tentative model is presented for electron transfer through the cyclic system.     

The binding of human milk lactoferrin to immunoglobulin A

To define the step at which translational initiation factor IF1 exercises its stimulation, initial rate kinetic analyses of 30 S initiation complex formation were carried out in the presence and absence of this factor. It was shown that, without affecting the affinity of the ribosomes either for the initiator tRNA or for the poly(AUG) used as template, IF1 increases approximately 2.5-fold the limiting Vmax of the 'pre-ternary complex'----ternary complex transition which represents the rate-limiting step in 30 S initiation complex formation. This kinetic effect titrates with the 30 S ribosomal subunit which must therefore represent the target of IF1 action.     

Rat liver membrane secretory component is larger than free secretory component in bile: evidence for proteolytic conversion of membrane form to free form

Secretory component is a receptor for polymeric immunoglobulins on epithelial cells and hepatocytes that facilitates transport of polymeric immunoglobulins into external secretions. Little is known about the transcellular migration of secretory component-polymeric IgA complexes or the membrane forms of secretory component. We therefore examined rat bile and liver membranes to identify and compare the various molecular species of secretory component. Bile or liver membrane proteins were electrophoresed in sodium dodecyl sulfate-polyacrylamide gels and electrophoretically transferred to nitrocellulose membranes. Protein profiles on blots were probed with antisecretory component antiserum, and the immunoreactive bands were visualized by indirect immunoperoxidase staining. Bile collected in the presence of proteolytic inhibitors showed an immunoreactive doublet band (Mr = 82,000 and 78,000) in the molecular weight range of free secretory component. By contrast, free secretory component in bile collected in the absence of proteolytic inhibitors and purified by affinity chromatography migrated as a single protein with an Mr = 70,000. Both components of the free secretory component doublet bound dimeric IgA when blots were probed with human dimeric IgA. Crude liver membranes prepared in the presence of proteolytic inhibitors showed two immunoreactive secretory component-containing bands, Mr = 107,000 and 99,000, whereas membranes prepared without proteolytic inhibitors showed two smaller immunoreactive bands; one of these proteolytically severed proteins comigrated with the 82,000-dalton free secretory component in bile. These results indicate that membrane forms of secretory component are present in rat liver. The observations that the membrane secretory component is larger than biliary free secretory component and yields biliary SC-like forms of secretory component upon proteolysis support the hypothesis that free secretory component in bile is a proteolytic product of larger liver membrane-associated secretory component.