A neural network study on the dynamic identification of a fermentation system

The objective of this paper is to propose neural networks for the study of dynamic identification and prediction of a fermentation system which produces mainly 2,3-butanediol (2,3-BDL). The metabolic products of the fermentation, acetic acid, acetoin, ethanol, and 2,3-BDL were measured on-line via a mass spectrometer modified by the insertion of a dimethylvinylsilicone membrane probe. The measured data at different sampling times were included as the input and output nodes, at different learning batches, of the network. A fermentation system is usually nonlinear and dynamic in nature. Measured fermentation data obtained from the complex metabolic pathways are often difficult to be entirely included in a static process model, therefore, a dynamic model was suggested instead. In this work, neural networks were provided by a dynamic learning and prediction process that moved along the time sequence batchwise. In other words, a scheme of two-dimensional moving window (number of input nodes by the number of training data) was proposed for reading in new data while forgetting part of the old data. Proper size of the network including proper number of input/output nodes were determined by trained with the real-time fermentation data. Different number of hidden nodes under the consideration of both learning performance and computation efficiency were tested. The data size for each learning batch was determined. The performance of the learning factors such as the learning coefficient η and the momentum term coefficient α were also discussed. The effect of different dynamic learning intervals, with different starting points and the same ending point, both on the learning and prediction performance were studied. On the other hand, the effect of different dynamic learning intervals, with the same starting point and different ending points, was also investigated. The size of data sampling interval was also discussed. The performance from four different types of transfer functions, x/(1+|x|), sgn(x)·x ²/(1+x ²), 2/(1+e ^{? x} )?1, and 1/(1+e ^{? x} ) was compared. A scaling factor b was added to the transfer function and the effect of this factor on the learning was also evaluated. The prediction results from the time-delayed neural networks were also studied.     

Improvement of anther culture methods for doubled haploid production in barley breeding

There is potential to accelerate cultivar development with a doubled haploid system for breeding line production. Anther culture methodology was evaluated for U.S.A. spring barley (Hordeum vulgare L.) breeding applications. Gelrite was found to be an acceptable replacement for ficoll in the induction medium to reduce costs while maintaining embryoid and plant production levels. Beneficial effects of 28 d cold pretreatment of donor spikes for anther culture were confirmed with Pacific Northwest USA barley genotypes. A 3 d mannitol solution pretreatment of fresh anthers was shown to be less effective for green plant production compared to 28 d cold pretreatment of donor spikes. Extended donor spike cold pretreatment from 28 to 42 d did not reduce anther culture productivity. Based on this research, anther culture techniques show promise for economical and convenient application in spring barley breeding.Abbreviations DH doubled haploid - LS Linsmaier and Skoog basal medium - BAP benzylaminopurine - GLM Generalized Linear Model - SAS Statistical Analysis System     

Boundary conditions at the cartilage-synovial fluid interface for joint lubrication and theoretical verifications

The objective of this study is to establish and verify the set of boundary conditions at the interface between a biphasic mixture (articular cartilage) and a Newtonian or non-Newtonian fluid (synovial fluid) such that a set of well-posed mathematical problems may be formulated to investigate joint lubrication problems. A "pseudo-no-slip" kinematic boundary condition is proposed based upon the principle that the conditions at the interface between mixtures or mixtures and fluids must reduce to those boundary conditions in single phase continuum mechanics. From this proposed kinematic boundary condition, and balances of mass, momentum and energy, the boundary conditions at the interface between a biphasic mixture and a Newtonian or non-Newtonian fluid are mathematically derived. Based upon these general results, the appropriate boundary conditions needed in modeling the cartilage-synovial fluid-cartilage lubrication problem are deduced. For two simple cases where a Newtonian viscous fluid is forced to flow (with imposed Couette or Poiseuille flow conditions) over a porous-permeable biphasic material of relatively low permeability, the well known empirical Taylor slip condition may be derived using matched asymptotic analysis of the boundary layer at the interface.     

Modeling with in vitro kinetic parameters for the elaboration of transfer RNA identity in vivo

A tRNA with "double identity" was created, and this tRNA was demonstrated in vitro to aminoacylate quantitatively with either of two amino acids. In contrast, acceptance of only one of these amino acids was observed in vivo, and a simple manipulation determined which one was accepted. Kinetic parameters were obtained for aminoacylation with each amino acid of the tRNA with double identity and of related tRNAs. Modeling with these parameters largely explains which amino acid specificity is observed in vivo. The results delineate some of the kinetic boundaries for the design and accommodation of tRNA sequence variations in the elaboration of identity in vivo.     

Endotoxin removal by charge-modified filters

The effects of positively charged nylon and depth (cellulose-diatomaceous earth) filters on endotoxin removal from various solutions were evaluated. The charged filter media removed significant amounts of Escherichia coli and natural endotoxin from tap water, distilled water, sugars, and NaCl solutions; no significant removal of endotoxin was observed with negatively charged filter media. The extent of removal was influenced by pH, the presence of salts, and organic matter. Such media may be useful for the control of endotoxins in raw-product water or solutions used to prepare parenteral drug products or in other fluids where endotoxin control is desired.     

银杏科研的主要成就和展望

前言银杏又名白果,是现存种子植物中最古老的孑遗植物,也是源于我国的特有珍贵树种。学者们一向把它视为植物界的奇珍,在植物分类学、树木学、果树栽培学和古植物学上占有特殊的位置,至今国内外植物学家、地质学家,以及果树栽培学家们,仍对它进行着深入细致的研究。银杏的科学奥秘将逐步被揭开。     

Production of Methyl Ketones from Secondary Alcohols by Cell Suspensions of C2 to C4n-Alkane-Grown Bacteria

Nineteen new C₂ to C₄n-alkane-grown cultures were isolated from lake water from Warinanco Park, Linden, N.J., and from lake and soil samples from Bayway Refinery, Linden, N.J. Fifteen known liquid alkane-utilizing cultures were also found to be able to grow on C₂ to C₄n-alkanes. Cell suspensions of these C₂ to C₄n-alkane-grown bacteria oxidized 2-alcohols (2-propanol, 2-butanol, 2-pentanol, and 2-hexanol) to their corresponding methyl ketones. The product methyl ketones accumulated extracellularly. Cells grown on 1-propanol or 2-propanol oxidized both primary and secondary alcohols. In addition, the activity for production of methyl ketones from secondary alcohols was found in cells grown on either alkanes, alcohols, or alkylamines, indicating that the enzyme(s) responsible for this reaction is constitutive. The optimum conditions for in vivo methyl ketone formation from secondary alcohols were compared among selected strains: Brevibacterium sp. strain CRL56, Nocardia paraffinica ATCC 21198, and Pseudomonas fluorescens NRRL B-1244. The rates for the oxidation of secondary alcohols were linear for the first 3 h of incubation. Among secondary alcohols, 2-propanol and 2-butanol were oxidized at the highest rate. A pH around 8.0 to 9.0 was found to be the optimum for acetone or 2-butanone formation from 2-alcohols. The temperature optimum for the production of acetone or 2-butanone from 2-propanol or 2-butanol was rather high at 60°C, indicating that the enzyme involved in the reaction is relatively thermally stable. Metal-chelating agents inhibit the production of methyl ketones, suggesting the involvement of a metal(s) in the oxidation of secondary alcohols. Secondary alcohol dehydrogenase activity was found in the cell-free soluble fraction; this activity requires a cofactor, specifically NAD. Propane monooxygenase activity was also found in the cell-free soluble fraction. It is a nonspecific enzyme catalyzing both terminal and subterminal oxidation of n-alkanes.     

Preferential degradation of the terminal carbohydrate moiety of membrane glycoproteins in rat hepatoma cells and after transfer to the membranes of mouse fibroblasts

Glycoproteins in the plasma membrane of rat hepatoma cells were labeled at their externally exposed tyrosine residues with 131I and at their galactose and sialic acid residues with 3H. The degradation of both isotopes in the total cell protein fraction, in glycoproteins purified by concanavalin A, and in glycoproteins separated on two-dimensional gels was determined. Similarly, the total cellular membrane glycoproteins were metabolically labeled with [35S]methionine and [3H]fucose. The fate of both incorporated labels was followed by lectin chromatography or by precipitation of the proteins with specific antibodies followed by electrophoretic gel separation. In both labeling experiments, the carbohydrate markers were lost from the ligand- recognized fraction with similar kinetics as from the total cell protein fraction. In some glycoprotein species which were separated by two-dimensional gel electrophoresis, the polypeptide portion exhibited up to a twofold slower rate of degradation relative to that of the carbohydrate moiety. This difference is most pronounced in carbohydrate- rich glycoproteins. To corroborate this finding, double-labeled membrane glycoproteins were incorporated into reconstituted phospholipid vesicles which were then transferred via fusion into the plasma membrane of mouse fibroblasts. Both the polypeptide and carbohydrate moieties of the transferred membrane glycoproteins were degraded with the same relative kinetics as in the original hepatoma cells. The rate of degradation is mostly a function of the structural properties of the membrane components as shown by the preservation of metabolically stable fucogangliosides of Reuber H-35 hepatoma cells transferred onto the fibroblasts. The technique of insertion of membrane components into the plasma membrane of another cell should assist in the elucidation of the exact route and mechanism of membrane protein destruction.