Nonnegative decomposition of functional count data

We present a novel decomposition of nonnegative functional count data that draws on concepts from nonnegative matrix factorization. Our decomposition, which we refer to as NARFD (nonnegative and regularized function decomposition), enables the study of patterns in variation across subjects in a highly interpretable manner. Prototypic modes of variation are estimated directly on the observed scale of the data, are local, and are transparently added together to reconstruct observed functions. This contrasts with generalized functional principal component analysis, an alternative approach that estimates functional principal components on a transformed scale, produces components that typically vary across the entire functional domain, and reconstructs observations using complex patterns of cancellation and multiplication of functional principal components. NARFD is implemented using an alternating minimization algorithm, and we evaluate our approach in simulations. We apply NARFD to an accelerometer dataset comprising observations of physical activity for healthy older Americans.     

A unique tubulin antiserum attenuates the rate of poleward chromosome movement in anaphase.

An antiserum against tubulin, NS20, was previously shown to specifically attenuate both fast axonal transport in vivo (Johnston, K. M. et al., Brain Res. 385, 38-45 (1986)) and in vitro (Johnston, K. M. et al., Cell Motil. Cytoskel. 7, 110-115 (1987)) and flagellar motility (Goldsmith, M. et al., Cell Motil. Cytoskel. 20, 249-262 (1991)). We hypothesized that NS20 blocked motility by binding to a multifunctional motor binding domain on the microtubules (MTs), or axonemes. Here we have examined the effect of microinjecting NS20, at metaphase, into dividing PtK2 cells. Plotting chromosome separation (CS) as a function of time, we report here that CS rates for anaphase A (chromosome-to-pole movement) were reduced by approximately 50% relative to uninjected controls. CS rates for anaphase B (spindle pole elongation) were unaffected by the NS20 antiserum. The inhibition of CS rate during anaphase A by NS20 was significantly greater than the inhibition caused by a control antitubulin serum (PC5). Two possible mechanisms underlying NS20's inhibition of CS during anaphase A were considered. NS20 could block the binding of a kinetochore-associated motor to kinetochore MTs (kMTs) or, alternatively, NS20 could stabilize kMTs against depolymerization. Our results favor the first alternative. In a cold-induced depolymerization assay, NS20 had no selective stabilizing effect on MTs. Moreover, we show that NS20 can selectively block the binding of a well characterized MT-associated motor (kinesin) to MTs, in vitro. These results suggest that NS20 may be defining a unique tubulin binding domain common to the motors underlying vesicle transport, flagellar motility, and chromosome movements during anaphase A.     

A randomized clinical trial of remission induction,consolidation, and chemo-immunotherapy maintenance in adult acute myeloblastic leukemia

Summary The Southeastern Cancer Study Group conducted a randomized clinical trial in acute myeloblastic leukemia and the blastic phase of chronic myelocytic leukemia to compare: Two induction programs (Schedule A) cytosine arabinoside and 6-thioguanine or (Schedule B) cytosine arabinoside, 6-thioguanine and daunorubicin; two consolidation programs (Schedule C) continuation of induction programs at a reduced dose or (Schedule D) a combination of cyclophosphamide, methotrexate and vincristine; and two maintenance programs — (Schedule E) 1 month of BCG, followed by methotrexate or (Schedule F) methotrexate. Over a 3 year period 372 patients were entered and 295 were judged evaluable. None of 11 patients with blastic phase of chronic myelocytic leukemia responded. There were no significant differences between the schedules in the number of patients with acute myeloblastic leukemia achieving complete remissions (37%, Schedule A vs. 41% Schedule B). The relapse rates on consolidation were similar (43%, Schedule C and 39%, Schedule D). BCG significantly prolonged the duration of first remission following consolidation (P<0.05) from 13.0–23.9 weeks. Survival was not significantly prolonged (92.7 weeks vs. 71.7 weeks). There were no serious complications from BCG therapy. Contributors. The following members of the Southeastern Cancer Study Group participated in this study: John T. Carpenter, John R. Durant, Richard Gams, William J. Hammack, George A. Omura, Gayle Roberts, University of Alabama School of Medicine, Birmingham, Alabama; Harold Silberman, Donald S. Miller, Duke University School of Medicine, Durham, North Carolina; William B. Kremer, Durham Veterans Administration Hospital, Durham, North Carolina; Evert A. Bruckner, Lawrence E. Cooper, Charles C. Corley, Joseph E. Hardison, Charles M. Huguley, Jr., James Keller, Mason G. Robertson, John D. Schmale, Charles Vogel, W. R. Vogler, William H. Whaley, E. F. Winton, Emory University School of Medicine, Atlanta, Georgia; Chan Kon Chin, Guy Faguet, Claude-Starr Wright, Medical College of Georgia, Augusta, Georgia; Y. S. Ahn, Howard E. Lessner, University of Miami School of Medicine, Miami, Florida; Dov Gorshein, Scott Murphy, Presbyterian University of Pennsylvania Medical Center, Philadelphia, Pennsylvania; William E. Barry, Sharon P. Fischer, Rosaline R. Joseph, Richard V. Smalley, Temple University School of Medicine, Philadelphia, Pennsylvania; Virgil Loeb, Jr., Cary Presant, Edward Reinhard, Shabbir H. Safdar, Washington University School of Medicine, St. Louis, Missouri; Norman Maldonado, Enrique Velez-Garcia, University of Puerto Rico School of Medicine, San Juan, Puerto Rico; S. A. Gregory, William H. Knospe, Rush-Presbyterian-St. Luke's Medical Center, Chicago, Illinois; Stephen Krauss, University of Tennessee Memorial Research Center, Knoxville, Tennessee; Karl Tornyos, New Orleans Veterans Hospital, New Orleans, Louisiana; W. B. Forman, R. W. Kellermeyer, A. Rassiga, Case Western Reserve University School of Medicine, Cleveland, Ohio; William R. Arrowsmith, George Porter, Donald M. Samples, Ochsner Clinic, New Orleans, Louisiana; Lois W. Dow, Charles L. Neely, University of Tennessee School of Medicine, Memphis, Tennessee; G. O. Broun, Jr., St. Louis University School of Medicine, St. Louis, Missouri     

The ecological requirements of Hierochloë odorata in Nova Scotia

Hierochloë odorata (L.) Beauv. is restricted to the upper zone of salt marshes and rarely becomes dominant.
In the field, the species was found not to have strict N, P, K requirements. It grew in a range of pH values from 4.3 to 7.9, tolerated salinities up to 500 mhos and favoured soils with a moisture content from 25 to 30%. It grew in situations with a mean water table between 14 and 28 cm below the surface. Natural shading in the field was found to increase the heights of plants by about 30%. The application of fertilizer in the field increased the height of sweet grass but also stimulated the growth of associated species. In the greenhouse, the effects of different soils and fertilizers on plant growth were assessed. Inorganic fertilizer (12:18:12) application produced more leaves and tillers than organic fertilizer (6:2:0).
Sweet grass grows in a zone of reduced competition between the dune species and the salt-marsh species. Near the salt-marsh, the species may be limited by high salt concentration. The low levels of competition offered by cultivation and the responsiveness of the species to fertilizer suggests that cultivation of Hierochloë odorata may be successful.