エピスタシス尺度に基づくリンケージ同定手法の提案  [in Japanese] Proposal of a Linkage Identification Method Based on Epistasis Measure  [in Japanese]

Access this Article

Search this Article

Author(s)

Abstract

遺伝的アルゴリズム(Genetic Algorithm ,GA )はビルディングブロックを交叉により組み合わせることで効果的な探索を実現しているが,そのためにどの遺伝子座がビルディングブロックを構成しうるのかを調べるリンケージ同定が重要となる.リンケージ同定に関してはこれまでにも確率モデルに基づく方法や非線形性もしくは非単調性を基に判断する手法が提案されている.本論文では,遺伝子座間に存在する非線形性を検出することでリンケージ同定を行うLINC (Linkage Identi cation by Nonlinearity Check )を発展させ,それぞれの遺伝子座のペアに対してエピスタシス(非線形性)尺度を定義し,それに基づいてリンケージの同定を実現する手法を提案する.Genetic Algorithms realize e ective search by exchanging building blocks through genetic recombinations.To realize e ective genetic search,linkage identi cation becomes important which detects a set of loci tightly linked to form a building block.Several methods are already proposed to identify linkage such as linkage learning algorithms based on probabilistic models and linkage identi cation procedures based on nonlinearity or non-monotonicity detection. In this paper,we extend the Linkage Identi cation by Nonlinearity Check (LINC)which identies linkage based on nonlinearity detection by de ning an epistasis measure for each pair of loci to realize linkage identi cation with the epistasis measure.

Genetic Algorithms realize effective search by exchanging building blocks through genetic recombinations. To realize effective genetic search, linkage identification becomes important which detects a set of loci tightly linked to form a building block. Several methods are already proposed to identify linkage such as linkage learning algorithms based on probabilistic models and linkage identification procedures based on nonlinearity or non-monotonicity detection. In this paper, we extend the Linkage Identification by Nonlinearity Check (LINC) which identifies linkage based on nonlinearity detection by defining an epistasis measure for each pair of loci to realize linkage identification with the epistasis measure.

Journal

  • 情報処理学会論文誌数理モデル化と応用(TOM)

    情報処理学会論文誌数理モデル化と応用(TOM) 43(SIG10(TOM7)), 6-13, 2002-11-15

    Information Processing Society of Japan (IPSJ)

References:  15

Cited by:  5

Codes

  • NII Article ID (NAID)
    110002711512
  • NII NACSIS-CAT ID (NCID)
    AA11464803
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    1882-7780
  • NDL Article ID
    6362044
  • NDL Call No.
    Z74-C192
  • Data Source
    CJP  CJPref  NDL  NII-ELS  IPSJ 
Page Top