Showing papers by "University of Southern Denmark published in 1992"

Filtration capacity of the ascidian Ciona intestinalis and its grazing impact in a shallow fjord

Abstract: Filtration capacity of the ascidian Ciona intestinalis (L.) was measured in the laboratory and determined from the exponential reduction in algal cell (Rhodomonas sp.) concentration as a function of time. Filtration rate (F, m1 min-l) as a function of dry weight (W, g) was found to be F= 118~:;$ and F = 199~:;6:,, for total dry weight and dry weight of organs, respectively. Ascidians starved at least 12 h had, after addition of algal cells, an initial lag-phase (1 to 2 h) with low filtration rates before constant high rates were attained. The duration of the initial lag-phase was longer at low algal cell concentrations. At algal cell concentrations above 10 to 15 X 103 cells m]-', the filtration rate declined to a lower level after some time, possibly correlated with the filling of the gut. In the temperature range 4 to 21 'C, maximum filtration rate (F,,,,,, m1 min-' ind:') increased linearly with increasing temperature (T 'C) according to Fm, = 1.4611.21. Above 21 \"C filtration rate declined rapidly with increasing temperature. The population density of C. intestinalis in the shallow cove Kertinge Nor, Fyn, Denmark, was estimated using analysis of stereophotographs. Population density varied greatly over the year with maximum densities in autumn. The filtration potent~al of the C. intestinalis population varled during the year between 0.1 and 1.0 times a volume equivalent to the total water volume of the cove per day. The results suggest that the population of C, intestinalis in Kertinge Nor may have an important grazing impact on the phytoplankton in late summer-early fall.

Cytogenetic analysis of 57 primary prostatic adenocarcinomas.

Abstract: Cytogenetic analysis after short-term culture in vitro of primary tumor samples was attempted in 82 patients with prostatic cancer. Tumor material was obtained by radical prostatectomy or transurethral resection. Successful cytogenetic studies were performed on 57 tumors of which five were well, 30 moderately, and 22 poorly differentiated adenocarcinomas. Only normal karyotypes were found in 24 tumors. Structural nonclonal aberrations were detected in 18 and clonal karyotypic abnormalities in 15 tumors. The most common clonal numerical aberration was loss of the Y chromosome; a missing Y was found in six tumors, in three of these as the sole anomaly. Clonal structural chromosomal rearrangements, usually accompanied by numerical changes, were detected in 12 tumors. The rearrangements involved 18 of the 22 autosomes and the X chromosome. Chromosomes 1, 7, and 10 were most frequently affected. Deletions, duplications, inversions, insertions, and balanced as well as unbalanced translocations were represented. The breakpoints in chromosome 1 were scattered along both the short and long arms with no obvious clustering, whereas those in chromosomes 7 and 10 were clustered at bands 7q22 (two deletions and two duplications in four different tumors) and 10q24 (two translocations, one deletion, and one inversion in four tumors). One additional tumor displayed a derivative chromosome 10 with a breakpoint in 10q23, and one had monosomy 10. Altogether, these abnormalities resulted in loss of 10q24----qter in five tumors. Monosomy 8 and rearrangements of the short arm of chromosome 8 leading to loss of 8p21----pter were seen in four tumors. Double minute chromosomes were found in two tumors.

Adjuvant Radiotherapy and Risk of Contralateral Breast Cancer

Abstract: BACKGROUND The risk of contralateral breast cancer is increased twofold to fivefold for breast cancer patients A registry-based cohort study in Denmark suggested that radiation treatment of the first breast cancer might increase the risk for contralateral breast cancer among 10-year survivors PURPOSE Our goal was to assess the role of radiation in the development of contralateral breast cancer METHODS A nested case-control study was conducted in a cohort of 56,540 women in Denmark diagnosed with invasive breast cancer from 1943 through 1978 Case patients were 529 women who developed contralateral breast cancer 8 or more years after first diagnosis Controls were women with breast cancer who did not develop contralateral breast cancer One control was matched to each case patient on the basis of age, calendar year of initial breast cancer diagnosis, and survival time Radiation dose to the contralateral breast was estimated for each patient on the basis of radiation measurements and abstracted treatment information The anatomical position of each breast cancer was also abstracted from medical records RESULTS Radiotherapy had been administered to 824% of case patients and controls, and the mean radiation dose to the contralateral breast was estimated to be 251 Gy Radiotherapy did not increase the overall risk of contralateral breast cancer (relative risk = 104; 95% confidence interval = 074-146), and there was no evidence that risk varied with radiation dose, time since exposure, or age at exposure The second tumors in case patients were evenly distributed in the medial, lateral, and central portions of the breast, a finding that argues against a causal role of radiotherapy in tumorigenesis CONCLUSIONS The majority of women in our series were perimenopausal or postmenopausal (53% total versus 38% premenopausal and 9% of unknown status) and received radiotherapy at an age when the breast tissue appears least susceptible to the carcinogenic effects of radiation Based on a dose of 251 Gy and estimates of radiation risk from other studies, a relative risk of only 118 would have been expected for a population of women exposed at an average age of 51 years Thus, our data provide additional evidence that there is little if any risk of radiation-induced breast cancer associated with exposure of breast tissue to low-dose radiation (eg, from mammographic x rays or adjuvant radiotherapy) in later life

Chromosome aberrations in 35 primary ovarian carcinomas

Abstract: Cytogenetic analysis was performed on short-term cultures of primary ovarian carcinomas from 62 patients. Cytogenetic analysis was successful in 59 cases. Clonal chromosome aberrations were detected in 35 tumors. Only numerical changes or a single structural change were found in five carcinomas: trisomy 12 was the sole anomaly in two tumors, one tumor had the karyotype 50,XX, + 5, + 7, + 12, + 14, a fourth tumor had a balanced t(1;5), and the fifth tumor had an unbalanced t(8;15). The fact that four of these five carcinomas were well differentiated suggests that simple karyotypic changes are generally characteristic of these less aggressive ovarian tumors. The majority of the cytogenetically abnormal tumors (n = 30) had complex karyotypes, with both numerical and structural aberrations and often hypodiploid or near-triploid stemlines. The numerical imbalances (comparison with the nearest euploid number) were mostly losses, in order of decreasing frequency -17, -22, -13, -8, -X, and -14. The structural aberrations were mostly deletions and unbalanced translocations. Recurrent loss of genetic material affected chromosome arms 1p, 3p, 6q, and 11p. The breakpoints of the clonal structural abnormalities clustered to several chromosome bands and segments: 19p13, 11p13-15, 1q21-23, 1p36, 19q13, 3p12-13, and 6q21-23. The most consistent change (16 tumors) was a 19p + marker, and in 12 of the tumors the 19p + markers looked alike.

Time-Dependent Association Measures for Bivariate Survival Distributions

Abstract: We propose time-dependent association measures for application to bivariate survival analysis. Such association measures provide informative summaries for data on twins, ophthalmic and auditory studies, and for other matched-pair designs. We develop several desirable properties of time-dependent association measures and study three measures motivated by these properties. We examine the measures from a general bivariate survival perspective and for the proportional hazards frailty model. We use monozygotic (MZ) and dizygotic (DZ) twin data from the Danish Twin Registry to illustrate how these measures depend on the specification of the proportional hazards frailty model. This model consists of two components: a baseline hazard function and a frailty distribution. We produce gamma and nonparametric maximum likelihood estimates of the frailty distribution and estimate a Gompertz baseline hazard function. For two of the measures, a nonparametric estimate provides a comparison to the model-based estim...

Statistical theory of charged-particle stopping and straggling in the presence of charge exchange

Abstract: General expressions have been derived for the mean energy loss and straggling of a heavy charged particle penetrating through a thin slice of matter. The main input parameter is a differential energy loss rate that allows for transitions within an arbitrary number of projectile states. Both spontaneous and collision-induced events are potentially induced, but different events are assumed statistically independent. Asymptotically, the mean energy loss is shown to approach a linear dependence on time or path length. The slope is given by the equilibrium energy loss rate, and the intercept at zero path length depends on the incident projectile state. Energy-loss straggling can be split into a collisional part and a contribution from variations in projectile state. The latter turns out to be related to the intercept in the mean energy loss for the pertinent projectile states. Previous results applying to a two-state system have been generalized to an arbitrary number of states. A full evaluation has been presented for a three-state system. A procedure is outlined that allows to empirically determine the magnitude of charge-exchange straggling.

Predicting population response to pollutants: the significance of sex

Abstract: Two of the primary goals of ecotoxicology are to assess past and present impacts of pollutant contamination on ecosystems and to predict diverse population and community responses to future pollutant exposure. In an attempt to meet these goals, great efforts are currently being devoted to devising measures that are ecologically relevant and sensitive, in addition to being practical and cost effective. Approaches using biochemical or physiological effect indicators are largely empirical in that they require some prior knowledge of the system or pollutant before predictions of population response can be made. Information gained in such studies can greatly facilitate our understanding of the mechanistic basis of pollutant toxicity and may indeed prove useful in predicting future pollutant impacts. Theoretical predictions, based on, for example, the concepts of community diversity and stability (Fisher et al. 1973; Fisher 1977; reviewed in Green 1979; Gray et al. 1988) or life-history theory (e.g. Calow 1983,1989; Caswell 1985; Sibly & Calow 1985, 1989; Maltby et al. 1987), have also proved their worth in providing insight into the factors governing population and community response to pollutant stress. Several studies have demonstrated the importance of reproductive mode in determining the responses of individuals and populations to environmental perturbations from pollution (Grassle & Grassle 1974; Gray 1979). Indeed, Gray (1979) suggested that the presence of a species in a polluted area may be more a question of life-history strategy than the degree of tolerance that individuals exhibit to adverse conditions. Gray was referring specifically to those aspects of life-history related to larval development mode (i.e. benthic vs planktonic), and reproductive rate. However, there are other aspects of life history that can have more obvious effects on ecological and evolutionary processes in populations and which have received even less recognition in ecotoxicological studies. The genetic consequences associated with

Prognostic impact of chromosome aberrations in ovarian cancer.

Abstract: Clinico-cytogenetic correlations were assessed in 88 patients with malignant ovarian tumours. Cytogenetic analysis of the primary tumours yielded normal karyotype (N) in 33 patients and abnormal karyotypes (A) in 55 patients. Within the A group, seven tumours had simple abnormalities (AS), i.e., numerical changes only or a single structural aberration, and 48 had karyotypes with complex aberrations (AC). A correlation analysis between groups N and A revealed that cytogenetic abnormalities were more often found among seropapillary tumours, and that cases with abnormal karyotypes on average were of higher stage and more often had residual tumour mass after initial surgery (P less than 0.05 for all variables). When the three groups N, AS, and AC were compared, they were found to be significantly different with regard not only to the three parameters mentioned above, but now tumour grade also appeared to correlate with karyotypic pattern (P = 0.001), with poorly differentiated tumours having the most complex karyotypes. In a correlation analysis between karyotypic pattern and survival, group A patients had shorter survival than group N (P = 0.049). In the corresponding analysis between groups N, AS, and AC, the differences were also significant (P = 0.039), with shorter survival in group AC than in groups N and AS. Stage, grade, residual tumour after primary surgery, and performance status also correlated with survival time. A multivariate analysis identified abnormal karyotype as being independently associated with short survival in advanced clinical stages (P = 0.030) of ovarian carcinoma. We conclude that cytogenetic analysis of tumour cells may be of clinical value in the assessment of prognosis in patients with malignant ovarian tumours.

The role of occupational exposure and immunodeficiency in B-cell malignancies

Abstract: Excesses of non-Hodgkin's lymphoma have been observed among farmers exposed to phenoxyacetic acid herbicides and, less persuasively, among workers exposed to insecticides. Exposure to organic solvents (particularly chlorinated hydrocarbons) has also been associated with an increased risk of NHL. TCDD (which is a contaminant of phenoxy herbicides), DDT, and chlorinated solvents have all been reported to induce impairment or suppression of cell-mediated immunity. We hypothesize that NHL is caused by common viruses, such as the Epstein-Barr virus, that induce proliferation and immortalization of B-cells, followed by T-cell impairment entailing cell-mediated immunodeficiency. The increased risk of NHL with HIV infection and heart or kidney transplantation, in which immunodeficiency also occurs, is consistent with this hypothesis.

First-order nonadiabatic coupling matrix elements from multiconfigurational self-consistent-field response theory

Abstract: A new scheme for obtaining first‐order nonadiabatic coupling matrix elements (FO‐NACME) for multiconfigurational self‐consistent‐field (MCSCF) wave functions is presented. The FO‐NACME are evaluated from residues of linear response functions. The residues involve the geometrical response of a reference MCSCF wave function and the excitation vectors of response theory. Advantages of the method are that the reference state is fully optimized and that the excited states, represented by the excitation vectors, are strictly orthogonal to each other and to the reference state. In a single calculation the FO‐NACME between the reference state and several excited states may be obtained simultaneously. The method is most well suited to describe situations where the dominant configurations for the two states differ mainly by a single electron replacement. When the dominant configurations differ by two electrons many correlating orbitals are required in the MCSCF reference state calculation to accurately describe the FO‐NACME. FO‐NACME between various states of H2, MgH2, and BH are presented. These calculations show that the method is capable of giving quantitatively correct results that converge to the full configuration interaction limit. Comparisons are made with state‐averaged MCSCF results for MgH2 and finite‐difference configuration interaction by perturbation with multi‐configurational zeroth‐order wave function reflected by interactive process (CIPSI) results for BH.

Re-emergence in remission of primary clone in acute myelogenous leukaemias with multiple chromosomal aberrations at diagnosis

Abstract: We describe the clinical, haematological and cytogenetic features of three patients who had acute myelogenous leukaemia (AML) with complex bone marrow karyotypes when first cytogenetically examined. Induction chemotherapy led to remission from the acute leukaemia. However, neither clinically nor morphologically did this remission mean a return to normal haematopoiesis. The two patients who displayed myelodysplastic features before and when AML was diagnosed, again developed myelodysplasia, and the third patient, who had a long history of polycythaemia vera, returned to this myeloproliferative condition. Nor was cytogenetic normalization achieved; instead, abnormal cell clones were found in which all but one of the karyotypic aberrations present at acute leukaemia diagnosis had disappeared. The solitary anomalies that were detected in these reemerging clones must correspond to the primary cytogenetic aberrations of the patients pre-leukaemic diseases. They were del(5) (q11q33) and del(17) (p11) in the two myelodysplastic cases, and der(18)t(9;18) (p11;p11) in the patient with long-standing polycythaemia vera. The other, secondary, aberrations were probably the leukaemogenic changes, and with the eradication or reduction of the subclones containing them, the leukaemic phenotype disappeared. The three cases add cytogenetic evidence to the growing understanding that the remission obtained in some AMLs is actually a return to a preleukaemic, myeloproliferative or myelodysplastic, syndrome.

Identification of a whole-arm translocation by in situ hybridization with directly fluorochrome-labeled probes in a myelodysplastic syndrome.

Abstract: A case of myelodysplasia was found to have a complex bone marrow karyotype, involving an apparent whole-arm translocation between 17q and 18q. The application of a simplified fluorescence in situ hybridization technique, using directly fluorochrome-labeled centromere-specific alpha-satellite DNA probes, demonstrated the presence of sequences from both chromosomes 17 and 18 in the centromere of the derivative chromosome. This proves that a true whole-arm translocation had occurred. The case exemplifies how in situ hybridization analysis can be used to resolve interpretation problems in cancer cytogenetics.

AVL Trees With Relaxed Balance

Abstract: AVL trees with relaxed balance were introduced with the aim of improving runtime per formance by allowing a greater degree of concurrency. This is obtained by uncoupling updating from rebalancing. An additional benefit is that rebalancing can be controlled separately. In particular, it can be postponed completely or partially until after peak working hours. We define a new collection of rebalancing operations which allows for a significantly greater degree of concurrency than the original proposal. Additionally, in contrast to the original proposal, we prove the complexity of our algorithm. If N is the maximum size of the tree, we prove that each insertion gives rise to at most I_ log_Phi(N + 3/2) + log_Phi(squareroot{5}) - 3 _I rebalancing operations and that each deletion gives rise to at most I_ log_Phi(N + 3/2) + log_Phi(squareroot{5}) - 4 _I rebalancing operations, where Phi is the golden ratio.

Improving the Interpretation of Cluster Analysis: Developing a new Rescaling Algorithm

Abstract: During the last two decades segmentation of cuslDmers has received increasing attention amongst the market research community. Usually, in such research some kind of clustering algorithm is used as an analytical tool for identifying life style segments. SAS Institute lnc.i& provides a number of clustering procedures based on different algorithms (i.e. linkage methods single, average, complete, (two-stage)density centroid, maximumlikelihood, Lance-Williams flexible-beta., Gower's median, McQuitty's similarity analysis, Ward's minimum-variance, and Anderberg's centroid sorting). While some procedures like PRCe FASTCLUS provide statistical clues regarding the 'number-of-clusters' problem. no help is given with interpreting the operational significance of the identified cluster structure. Interpreting clusters can be a cumbersome if not impossible task which represents a serious obstacle to the successful practical implementation of the analysis. This paper demonstrates how this pitfall can be avoi~ by combining several SA~ procedures with a rescaling technique proposed by the authors. The rescaling, called Weighted Factor Average (W'F'A) and a related distance-measure, d, is used in combination with SASISTA 1"* PAOC PRINCOMP, PROC FACTOR, PROC FASTCLUS, PAOC COAAESP, and SASiGRAP~ PROC GPLOT & PAOC G3D. The technique is evaluated using a 1000 respondents x 57 variables datamatrix originating from a commercial study concerning convenience goods. INTRODUCTION When segmenting a heterogeneous market of consumers, cluster analysis is an indispensable analytical tool. Frequently, the input data set used for the clustering phase is a matrix of target-market respondents' answers to a set of life-style oriented AIO-Ouestions. Every researcher dealing wiltl quantitative segmentation analysis faces several crucial questions: 1, What questions to ask (which variables to indude) 2. What scaling to apply 3. What algorithm to use for clustering 4. How to handle outlier and missing data 5. How to determine the right number 01 dusters 6. How to get interpretational meaningful cluster profiles While much literature has been published regarding 1 through 5 I see Sudman & Bradburn (82), Brockhoff (85), Everitt (79,81), Lorr (83), Romesburg (84), and Spath (8S) for detailed discussions on those topics 1 -very few researchers have dealt with the last of the above problems perhaps the most important question of all. One might argue that one and two, which belong to the input phase, can be handled in an appropriate way if treated with care. Regarding 3 it is possible to produce a mathematical solution in an euclidian hyperspace. However, results might differ remarkably, depending on the algorithm used, i.e. single linkage will favour one 'snake like' super .. duster and several micro-clusters while other algorithms tend to favour clusters of roughly equal size. One might also develop guidelines for performing an outlier check (4) .. See i .. e .. SAS STAT Version 6, Vol I P 842-49 Ukewise statistical cues have been developed which give recommendations regarding (5) .. The Cubic Clustering Criterion (CCe) of PROC FASTCLUS is such an example. eec measures the deviation of the identified clusters from the distJ:ibution of poinll:! expected if the points where actually generated from a uniform distribution (However the CCC-measure is valid only in a hyperspace consisting of uflCO(fealated variables). Using ecc sometimes causes decisional difficulties since several cluster solutions might seem to be of equal quality. That is, it may be impossible to make an incontrovertible decision as to which of the n solutions to choose for subsequent analysis, Perhaps the CCC is at a peak when using 6 clusters, but only slightly beyond the peak when using 4. So, is it justifiable to use Occhams razor by trading some statistical quality for interpretational simplicity. However, the opposite situation might occur. Although statistics may favour 4 clusters, using 6 might prove superiOf from an interpretational point of view .. 1318 Regarding (6), very little has been published. This is due to the fact that it is nearly impossible to develop lonna! procedures which possess universal validity, that is those which are not problem specific. This methodological difficulty is even more 01 an issue when comparing across scales and research disciplines. Is it at all possible to use the same type of interpretational tools (e.g. rules 01 thumb) regarding market research, astrophysics, and medicine? In the former case one usually dusters data which are scaled according to a semantic differential ('soft' data), whiie in the latter case the data are ratio scaled i.e. distances between galactical structures or differences in wavelength and luminosity of stars ('hard' data). ALGORITHMIC CONSIDERATIONS With the conventions, that we start out with and I) 0 observations, which we end up classifying into m clusters 2) N variables from which we extract n factors we denote by X. the original OxN data matrix (standardized) and by C the Oxm "clustering matrix" defined by _11 if obs i bBlongs to cluster j ct,.-\O otherwise Furthermore, let S be the mxm diagonal matrix having as its i'the diagonal element su.=size(clust:ec (il) Now it is possible to compute a new matrix A: In terms of the above defined matrices, A can be written as the matrixproduct A = x.-CS·' Assume a factor analysis has been performed (i.e. prindpal components extraction of factors using the matrix of correlation with subsequent VARIMAX-rotation). Thus finding F' (the transposed variables x factorloadings matrix) yields: { LOAD" LOAD" 1" ~1 LQA.D;u .. .. LOADlIl LOADZl2 Finally, a new diagonal matrix is needed: