Geometrical Formation of Compound Starch Grains in Rice Implements Voronoi Diagram

Ryo Matsushima, Masahiko Maekawa, Wataru Sakamoto

Research output: Contribution to journalArticlepeer-review

13 Citations (Scopus)


Starch forms transparent grains, called starch grains (SGs), in amyloplasts. One of the major morphological SG forms in Poaceae, called a compound SG, is formed by assemblies of small starch granules in an amyloplast. Starch granules assemble as a well-ordered structure; however, the mechanism that regulates this organization has not been identified. In this study, we examined how starch granules grow and converge into the final SG morphology. First, we found that the number of starch granules in an amyloplast is almost constant from the early developmental stage until endosperm maturity. Next, we quantitatively evaluated the geometrical similarities between starch granules and a Voronoi diagram, which is a mathematical tessellation of space based on the distance to a specific set of points in the space. The in silico growth simulation showed that the geometrical patterns of compound SGs resembling a Voronoi diagram is determined by physical interactions among the free-growing starch granules and the amyloplast envelope membrane. The geometrical similarity between compound SGs and a Voronoi diagram is likely a result of maximum loading and storage of starch in the amyloplast. The simulation described in this study provides a greater understanding of how compound SGs are formed and also has the potential to explain morphological variations of SGs.

Original languageEnglish
Pages (from-to)2150-2157
Number of pages8
JournalPlant and Cell Physiology
Issue number11
Publication statusPublished - Nov 1 2015


  • Amyloplast
  • Endosperm
  • Rice
  • Simulation
  • Starch granule
  • Voronoi diagram

ASJC Scopus subject areas

  • Physiology
  • Plant Science
  • Cell Biology


Dive into the research topics of 'Geometrical Formation of Compound Starch Grains in Rice Implements Voronoi Diagram'. Together they form a unique fingerprint.

Cite this