Innovation and market structure in the dynamics of the pharmaceutical industry and biotechnology: towards a history-friendly model

This paper is a first attempt at modelling the long-term dynamicsof market structure and innovation in the pharmaceutical industryin a history-friendly way. The model examines the relationshipsbetween the nature of the search space, demand, the patternsof competition, and industry evolution in the age of randomscreening and in the age of molecular biology, and shows thatconcentration in the pharmaceutical industry is shaped by lackof cumulativeness in innovative activities and market fragmentation.The model conforms to our appreciative understanding and respondsto changes in parameters concerning demand, costs, economiesof scale, opportunity conditions and the relative advantagesof new biotechnology firms (NBFs) vis-à-vis incumbents.With the exception of cost increases, the model is quite robustto these changes in its essential features: it is quite difficultto raise substantially concentration and to have NBFs displacingincumbents.     

Volatile components of artemisia vallesiaca all. (Mountain wormwood)

Summary The essential oil obtained by steam distillation of the leaves and flowers ofArtemisia vallesiaca All., a plant widely employed in the liqueur making industry and of almost completely unknown chemical composition, after separation on a silica gel column, was investigated by means of GLC, IR and mass spectrometry. Camphene, 1,8-cineole, camphor and borneol accounted for 67–68% of the oil. Over one hundred minor compounds were found by capillary gas chromatography and 52 were identified (Table 1).

---

Die flüchtigen Aromastoffe von Artemisia vallesiaca All. (Berg-Genippi)

Zusammenfassung Das ätherische Öl — erhalten durch Wasserdampfdistillation aus Blättern und Blüten derArtemisia vallesiaca All., einer Pflanze, die in der Wermut-und Likörproduktion viel verwendet wird, aber chemisch fast unbekannt ist — wurde nach Säulenchromatographie an Kieselgel analysiert mittels GLC, IR und GLC-MS. Camphen, 1,8-Cineol, Campher und Borneol bilden 67–68% des ätherischen Öl. Mehr als 100 weitere Bestandteile wurden durch GLC gefunden und 52 von diesen wurden identifiziert (Tabelle 1).

     

Influence of sucrose on cold gelation of heat‐denatured whey protein isolate

A solution of heat‐denatured whey proteins was prepared by heating 100 g kg⁻¹ whey protein isolate (WPI) at pH 7.0 to 75 °C for 15 min in the absence of salt. Heat treatment caused the globular protein molecules to unfold, but electrostatic repulsion opposed strong protein–protein aggregation and so prevented gel formation. When the heat‐denatured whey protein solution was cooled to room temperature and mixed with 15 mM CaCl₂, it formed a gel. We investigated the influence of the presence of sucrose in the protein solutions prior to CaCl₂ addition on the gelation rate. At relatively low concentrations (0–100 g kg⁻¹), sucrose decreased the gelation rate, presumably because sucrose increased the aqueous phase viscosity. At higher concentrations (100–300 g kg⁻¹), sucrose decreased the gelation rate, probably because sugar competes for the water of hydration and therefore increases the attraction between proteins. These data have important implications for the application of cold‐setting WPI ingredients in sweetened food products such as desserts. © 2000 Society of Chemical Industry     

Semi-supervised ensemble classification in subspaces

Graph-based semi-supervised classification depends on a well-structured graph. However, it is difficult to construct a graph that faithfully reflects the underlying structure of data distribution, especially for data with a high dimensional representation. In this paper, we focus on graph construction and propose a novel method called semi-supervised ensemble classification in subspaces, SSEC in short. Unlike traditional methods that execute graph-based semi-supervised classification in the original space, SSEC performs semi-supervised linear classification in subspaces. More specifically, SSEC first divides the original feature space into several disjoint feature subspaces. Then, it constructs a neighborhood graph in each subspace, and trains a semi-supervised linear classifier on this graph, which will serve as the base classifier in an ensemble. Finally, SSEC combines the obtained base classifiers into an ensemble classifier using the majority-voting rule. Experimental results on facial images classification show that SSEC not only has higher classification accuracy than the competitive methods, but also can be effective in a wide range of values of input parameters.