On the limitations of comparing mean square forecast errors

Linear models are invariant under non-singular, scale-preserving linear transformations, whereas mean square forecast errors (MSFEs) are not. Different rankings may result across models or methods from choosing alternative yet isomorphic representations of a process. One approach can dominate others for comparisons in levels, yet lose to another for differences, to a second for cointegrating vectors and to a third for combinations of variables. The potential for switches in ranking is related to criticisms of the inadequacy of MSFE against encompassing criteria, which are invariant under linear transforms and entail MSFE dominance. An invariant evaluation criterion which avoids misleading outcomes is examined in a Monte Carlo study of forecasting methods.     

Structure of astacin and implications for activation of astacins and zinc-ligation of collagenases.

Astacin, a digestive zinc-endopeptidase from the crayfish Astacus astacus L., is the prototype for the 'astacin family', which includes mammalian metallo-endopeptidases and developmentally regulated proteins of man, fruitfly, frog and sea urchin. Here we report the X-ray crystal structure of astacin, which reveals a deep active-site cleft, with the zinc at its bottom ligated by three histidines, a water molecule and a more remote tyrosine. The third histidine (His 102) forms part of a consensus sequence, shared not only by the members of the astacin family, but also by otherwise sequentially unrelated proteinases, such as vertebrate collagenases. It may therefore represent the elusive 'third' zinc ligand in these enzymes. The amino terminus of astacin is buried forming an internal salt-bridge with Glu 103, adjacent to His 102. Astacin pro-forms extended at the N terminus, as observed for some 'latent' mammalian astacin homologues, did not exhibit this 'active' conformation, indicating an activation mechanism reminiscent of trypsin-like serine proteinases.     

Development and function of T cells in mice rendered interleukin-2 deficient by gene targeting.

Interleukin-2 (IL-2) is a lymphocytotropic hormone which is thought to have a key role in the immune response of mammalian cells. It is produced by a subpopulation of activated T-lymphocytes and acts in vitro as the principal auto- and paracrine T-cell growth factor (for reviews see refs 1-3). IL-2 is, however, not the sole T-cell growth factor, nor does it act exclusively on T cells, also promoting growth of NK cells and differentiation of B cells. A role for IL-2 in T-cell development has been postulated but remains controversial. Here we test the requirement for IL-2 in vivo using IL-2-deficient mice generated by targeted recombination. We find that mice homozygous for the IL-2 gene mutation are normal with regard to thymocyte and peripheral T-cell subset composition, but that a dysregulation of the immune system is manifested by reduced polyclonal in vitro T-cell responses and by dramatic changes in the isotype levels of serum immunoglobulins.     

A small GTP-binding protein dissociates from synaptic vesicles during exocytosis.

Low-molecular-weight GTP-binding proteins are strong candidates for regulators of membrane traffic. In yeast, mutations in the sec4 or ypt1 genes encoding small GTP-binding proteins inhibit constitutive membrane flow at the plasma membrane or Golgi complex, respectively. It has been suggested that membrane fusion-fission events are regulated by cycling of small GTP-binding proteins between a membrane-bound and free state, but although most of these small proteins are found in both soluble and tightly membrane-bound forms, there is no direct evidence to support such cycling. In rat brain a small GTP-binding protein, rab3A, is exclusively associated with synaptic vesicles, the secretory organelles of nerve terminals. Here we use isolated nerve terminals to study the fate of rab3A during synaptic vesicle exocytosis. We find that rab3A dissociates quantitatively from the vesicle membrane after Ca2(+)-dependent exocytosis and that this dissociation is partially reversible during recovery after stimulation. These results are direct evidence for an association-dissociation cycle of a small GTP-binding protein during traffic of its host membrane.     

The European Science Foundation: Excerpts from the annual report for 1987

Science Policy News

The European Science Foundation: Excerpts from the annual report for 1987     

Clustering the rows and columns of a contingency table

A number of ways of investigating heterogeneity in a two-way contingency table are reviewed. In particular, we consider chi-square decompositions of the Pearson chi-square statistic with respect to the nodes of a hierarchical clustering of the rows and/or the columns of the table. A cut-off point which indicates significant clustering may be defined on the binary trees associated with the respective row and column cluster analyses. This approach provides a simple graphical procedure which is useful in interpreting a significant chi-square statistic of a contingency table.The author gratefully acknowledges the constructive comments of the referees and the editor.     

Pharmacodynamic profile of CQP 201-403, a novel 8α-amino-ergoline

Summary The profile of action in animals of CQP 201-403, a novel 8-amino-ergoline, is in most aspects that of a very potent dopaminomimetic, both as a prolactin secretion inhibitor, and at the levels of the CNS and the cardiovascular system. Qualitatively CQP 201-403 differs slightly from bromocriptine and apomorphine in its effects on the CNS (no influence on serotonin metabolism in the rat cortex; induction of masculine mounting behavior in rats) and the cardiovascular system of the dog (reflex tachycardia in response to a blood-pressure fall). In man the new compound proved to be highly active in lowering prolactin serum levels and to be more potent than bromocriptine (Parlodel^®).In memory of Dr Annemarie Closse, who died 14 June 1987.