Minimal Spanning Tree Analysis of Fungal Spore Spatial Patterns

• L. Jones Cameron

  Centre for Applied Colloid and BioColloid Science, School of Chemical Sciences, Swinburne University of Technology

• T. Lonergan Greg

  Centre for Applied Colloid and BioColloid Science, School of Chemical Sciences, Swinburne University of Technology

• E. Mainwaring David

  Department of Applied Chemistry Royal Melbourne Institute of Technology

抄録

Fungal colonies reproduce via asexual spores which differentiate on selected hyphae. At low magnification, the spores appear as a point set distribution. A practical method to empirically evaluate these spatial point sets is developed which is premised on finding the Minimal Spanning Tree (MST). This is a graph theoretic approach to solving the generalized ‘Travelling Salesman Problem’ - that is, how to connect a set of isolated points in the most efficient way. This paper applies a computerised method using the S-Plus object oriented programming language for cluster analysis of these spatial patterns. The MST returns a unique branching, continuously connected pattern which summarises the shortest distance path which connects all the spores. We can hypothesise that this pathway is one geometric representation of the minimum physiological connectedness needed for the coordinated structural development of the asexual reproduction mechanism in fungi. Sporulation is generally considered to be an adaptive response which allows epigenic control of growth in hostile conditions. The MST therefore provides empirical measurement of the spatial cluster-correlation of the pattern.

収録刊行物

• bioimages

  bioimages 4 (3), 91-98, 1996

  日本バイオイメージング学会

キーワード

• minimal spanning tree
• fungal spores
• spatial patterns
• graph theory
• S-Plus
• nearest-neighbour distributions

詳細情報

• CRID: 1390570796245706368
• NII論文ID: 10002037779
• NII書誌ID: AA11084187
• DOI: 10.11169/bioimages.4.91
• ISSN: 09192719
• 本文言語コード: en
• データソース種別:
  • JaLC
  • CiNii Articles
• 抄録ライセンスフラグ: 使用不可