計量生物学
Online ISSN : 2185-6494
Print ISSN : 0918-4430
ISSN-L : 0918-4430
20 巻 , 2 号
選択された号の論文の8件中1~8を表示しています
総説
  • 林 武司
    2000 年 20 巻 2 号 p. 109-129
    発行日: 2000/03/30
    公開日: 2012/02/27
    ジャーナル フリー
    Genetic maps with many markers are now available for agronomic plants and animals. Statistical methods for detecting QTL using genetic maps were mainly developed for plants, because it is easier to generate inbred lines in plants. Segregating populations derived from crossing of two inbred lines, such as BC1 and F2, are the most adequate materials for QTL mapping. However, methods devised for plant data obtained from the line crossing cannot be applied to the analysis of data of animals, since they cannot fully take account of the more complex date structures of outcrossing animal populations, i. e. data on several families with relationships across families and unknown linkage phases in parents.
    We report several methods for interval mapping of QTLs generally used in plants and animals, respectively. The following methods are described. (1) Maximum likelihood and least squares methods based on a simple regression model for plants and animals such as pigs, in which large full-sib families are available. (2) A method based on animal models in best linear unbiased prediction (BLUP) of breeding values for animals. (3) Sib-pair linkage tests, which was firstly introduced for human QTL mapping and is applicable to animals.
    We elucidate how the difference in statistical models for QTL mapping between plants and animals corresponds to difference in data structures.
原著
  • 了 偉, 根角 博久, 山田 昌彦, 高野 泰, 鵜飼 保雄
    2000 年 20 巻 2 号 p. 131-141
    発行日: 2000/03/30
    公開日: 2012/02/27
    ジャーナル フリー
    A method for quantitative evaluation of 3-dimensional (3D) fruit morphology was proposed, including measurement, quantitative description, normalization and statistical analysis of the 3D fruit shape. A measurement system consisting of a CCD laser displacement sensor and two pulse stages was constructed to obtain 3D surface data of fruit. By expanding the shape data measured in a series of spherical harmonic functions, a collection of coefficients to characterize the object surface were obtained. The surface data and coefficients expanded were normalized to provide viewpoint independent of object coordinates for comparison and analysis of different fruits. Based on the principal component analysis about the normalized spherical harmonic coefficients, the scores of some first principal components with high contributions were chosen to summarize the information about 3D shape. We applied the method to evaluate the genetic behavior for citrus and persimmon varieties.
  • Hiroshi Hakoyama, Yoh Iwasa
    2000 年 20 巻 2 号 p. 143-154
    発行日: 2000/03/30
    公開日: 2012/02/27
    ジャーナル フリー
    To estimate parameters without bias and to determine confidence intervals of the parameter in general cases are difficult. Even the maximum likelihood (ML) estimators may have significant bias if the sample-size is small. In this paper, we develop a simple method of removing large bias in parameter estimation and calculating confidence intervals based on the Monte Carlo sampling. The method is applicable to any parameteric models for which (1) the computer simulation can be performed and (2) a biased (but correctable) estimator *bias can be constructed. Let E[*|θ] be the expected value of the estimator *(s*) calculated for data s* that is generated by the model with parameter θ. For a given estimator with a large bias, * bias’ we can calculate bias-corrected estimator, *bc’ which satisfies the following relationship; E[*bias(s*)|*bc (s)] = *bias(s), where s is the observed data. We can find the bias corrected estimate *bc(s) and its confidence intervals by trial and error, using the Monte Carlo sampling repeatedly. We can prove that *bc is the unbiased estimator if θ and *bias are linearly related. To illustrate the use of this method, we apply it to a stochastic differential equation model for a logistically growing population with environmental and demographic stochasticities. An approximate maximum likelihood (AML) estimate of three parameters (intrinsic growth rate r, carrying capacity K, and environmental stochasticity **) has a significant bias, especially if the time series data of population size is short. However we can remove the bias very effectively by the Monte Carlo sampling.
  • Teruyuki Nakayama, Kenichi Satoh, Megu Ohtaki
    2000 年 20 巻 2 号 p. 155-166
    発行日: 2000/03/30
    公開日: 2012/02/27
    ジャーナル フリー
    We deal with the problem of whether or not there is an increase in female proportion among 0-4 year olds in the Japanese population during the last 21 years. The validity of the use of a binomial model for analyzing the municipality-specific data of female proportion is examined using a beta-binomial model. It is suggested from a logistic regression analysis that there exists a difference of the female proportion of infant population among population-size-specific groups of municipalities. The highest proportion was observed for the rural group, and the lowest one was observed for the city group. A slight but statistically signifant increasing time-trend in female proportion among the city group is detected via a test with the null hypothesis that the proportions of females are all equal against an ordered alternative.
  • 談 小健, 高野 泰, 岸野 洋久
    2000 年 20 巻 2 号 p. 167-179
    発行日: 2000/03/30
    公開日: 2012/02/27
    ジャーナル フリー
    Paired comparison is popular in preference trials. The estimated Scores have high accuracy, even if the data include measurement errors. However, incomplete comparison becomes inevitable, especially when the number of objects to be scored is large. In such cases, high precision by paired comparison is not guaranteed. In this paper, we propose a maximum likelihood estimation of scores, based on four-fold choice data. An empirical study of flower preference showed that the method of four-fold choice takes about the same time as a paired comparison, and the estimated scores were almost the same for the two procedures. It was clearly shown, by numerical simulation, that the precision of the estimated score of the most preferred object does not decrease with an increased number of objects to be compared in the method of four-fold choice. This is in great contrast to the paired comparison method. The estimated scores of less preferred objects have larger variances in the method of four-fold choice, whereas those from paired comparisons have similar variances for all objects. Thus, the four-fold choice method is effective when the scores of Most preferred objects are matters of concern.
  • Sadao Tomizawa, Nobuko Miyamoto, Hideki Horiguchi
    2000 年 20 巻 2 号 p. 181-191
    発行日: 2000/03/30
    公開日: 2012/02/27
    ジャーナル フリー
    For the analysis of square contingency tables with the same ordered row and column classifications Goodman (1985) considered the diamond (DD) model. This paper proposes a measure to represent the degree of departure from the DD model. The measure is somewhat similar to Goodman and Kruscal’s(1954) gamma measure. It is applied to two kinds of unaided distance vision data. For these data, the degrees and the patterns of departure from the DD model are compared.
研究速報
フォーラム
  • 上坂 浩之, 森川 敏彦, 魚井 徹
    2000 年 20 巻 2 号 p. 201-215
    発行日: 2000/03/30
    公開日: 2012/02/27
    ジャーナル フリー
    Showing the non-inferiority of a new drug to an existing treatment is one of the important ways of demonstrating efficacy of a new drug. However, it has been recognized that statistically rejecting a null hypothesis of inferiority does not directly imply true non-inferiority. There are several caveats in concluding non-inferiority based on the test of the inferiority hypothesis: the type I error rate may be infiated by biases which arise from inappropriate design and/or conduct of a non-inferiority trial, the active control may not be as effective as expected, or the non-inferiority margin may be too large to exclude ineffective treatment. We discuss 1) source of biases and means of eliminating them, 2) conditions of the non-inferiority margin, and 3) requirements for the design of a non-inferiority trial. We argue that the credibility of the conclusion from the trial results depends on not only the trial design but also on how the trial is conducted, protocol violations handled and results analyzed. We also present some means to ensure credibility. Finally, check lists for evaluating quality and credibility of the non-inferiority trial results are proposed.
feedback
Top