|
BTU Cottbus Institut für Informatik, Informations- und Medientechnik, Chair for Graphics Systems |
together with
|
of biological and chemical systems"
(combined with the first
GroIMP user and developer meeting)
Location: BTU Cottbus, Computer Science
Department,
Cottbus (Germany), Ewald-Haase-Str. 12/13,
room 001
Ole
Kniemeyer (BTU Cottbus): Relational Growth Grammars and the language XL
Michael
Henke (University of Göttingen): The 3D construction set for plant modelling
(+ print
version)
Reinhard
Hemmerling (BTU Cottbus): Simulation of chemical reactions with GroIMP and XL
Birka
Fonkeng (BTU Cottbus): Layouts and filters for the visualisation of graphs in
GroIMP (Diploma thesis presentation, ppt version, in German)
Jan
Dérer (BTU Cottbus): Descriptive tree data format and analysis functions in
GroIMP
(+ ppt
version)
Abstracts of the
contributions
The formalism of relational growth grammars as a combination of graph grammars and L-systems is presented in theory and practice. The programming language XL is used for the practical part, it is an extension of Java and provides an interface based on operator overloading for the implementation of rule-based formalisms like relational growth grammars. Applications in biology and architecture demonstrate the capabilities of the approach.
Michael Henke: The 3D
construction set for plant modelling
The above-ground structure of plants and
trees can be subdivided into the trunk, several levels of branches and
ultimately the leaves. Every plant organ produces its direct successors,
determines their number and their characteristic. In the same intuitive way it
is now possible with the 3D construction set integrated into XL to design
models. The "blocks" provided in this construction set allow the
structural modelling of a large bandwidth of not only herbal plants. Through
simple combinations of blocks, the basic structure of the object to be
modelled is defined. The characteristics of the individual components can be
controlled by a great number of parameters.
Tully Yates, A. Phillips, C. Rawlings, C. Hodgman, J. Köhler: Disecting the feedback control of gibberellin biosynthesis using a combined wetlab - in silico approach
The plant growth hormone gibberellin (GA) is a key regulator of growth and development in plants. The pathway is well characterized and is the main research topic of the hormone signalling group at Rothamsted Research. However, the coordinate genetic regulation by transcription factors, transport and feedback is unclear. The aim of the project is to use a combined experimental and computational approach to investigate the gene regulator mechanisms of GA biosynthesis in Arabidopsis. Functional Structural Plant Modelling (FSPM) will further be used to understand key issues of transport and tissue type specificity. GA biosynthesis can be divided into three parts based on the subcellular localization. Moreover the expression of many of the genes involved is tightly controlled by developmental stage and tissue type. FSPM allows us to tie together these qualitative factors to build a quantitative model of GA biosynthesis.
Reinhard Hemmerling:
Simulation of chemical reactions with GroIMP and XL
When modelling plant growth, one needs to incorporate structural
constraints, lighting conditions, distribution of nutrients and maybe more into
the simulation. The distribution of nutrients in the plant depends on
transportation and chemical kinetics.
In this presentation an introduction to chemical kinetics and its
numerical simulation in a computer will be shown. Then simple examples of how
to specify chemical networks in GroIMP/XL will be demonstrated and what their
simulation results look like. In the end, some problems that have been
discovered during research will be depicted and put up for discussion.
Christian Klukas, Falk
Schreiber: Methods for the dynamic exploration and editing of KEGG pathway
diagrams
The KEGG Pathway database is a very valuable information resource for researchers in the fields of life sciences. It contains metabolic and regulatory processes in the form of wiring diagrams, which can be used for browsing and information retrieval as well as a base for modeling and simulation. Thus it helps in understanding biological processes and higher-order functions of biological systems. Currently the KEGG website uses semi-static visualizations for the presentation and navigation of its pathway information. While this visualization style offers a good pathway presentation and navigation, it does not provide some of the possibilities related to dynamic visualizations, most importantly, the creation and visualization of userspecific pathways. This talk presents methods for the dynamic visualization, interactive navigation and editing of KEGG pathway diagrams. These diagrams, given as KEGG Markup Language (KGML) files, can be visually explored using novel approaches combining semi-static and dynamic visualization, but also edited or even newly created and then exported into KGML files.
Birka Fonkeng: Layouts and filters for the visualisation of graphs in
GroIMP
(Diploma thesis presentation, in German)
In this diploma thesis, various 2-dimensional layout algorithms for graphs have been investigated and implemented in the GroIMP software in order to visualise the graphs which are transformed by Relational Growth Grammars. These layouts can be subdivided in grid-based (including circular), hierarchical, force- (or energy-) based and edge-based layouts. The edge-based layout algorithm is a specific development for RGGs where the different edge types (corresponding to relations) get assigned type-specific directions. Furthermore, some filter operations for graphs (for emphasizing or hiding parts of a graph) were implemented.
Gerhard Buck-Sorlin: Virtual
rose: simulating rose architecture with GroIMP
to optimize flower production
in glasshouse production systems
Cut roses (Rosa hybrida) represent a high input and
high-value ornamental glasshouse crop. Many factors affect number and quality
of flowers, among them most prominently the manipulation strategy exercised by
the grower. The latter consists in a constant interaction with the growing crop
which is “fuzzy” and difficult to predict both in its temporal and spatial
pattern. In this presentation, a preliminary version of a cut rose model
written in XL within GroIMP will be shown. It will then be discussed how such
an essentially morphological model could become a true functional-structural
plant model by linking it with various submodels, which simulate external
environmental (light, temperature) and internal processes (transport of
nutrients, signal transduction). Finally, GroIMP’s potential for user
interaction and global context sensitivity will be explored in the context of
cut rose management practices and their effects on crop physiology at different
scales: bending of shoots to increase source strength, pruning of undesired
shoots, and rose cutting for harvest.
Dirk Lanwert: Ecophysiological forest stand modelling with XL/Java and
GroIMP via internet
A simple ecophysiological
single tree model of conifer stands considering spatial needle biomass
distribution and resulting light conditions calculates the net photosynthesis
production over annual periods. Stand structure description and process
calculation are realised using XL/Java and the modelling plattform GroIMP. The
integration of the GroIMP http interface within the model shows an examplary
method to provide model calculation facilities as a webservice.
Jan Dérer: Descriptive tree
data format and analysis functions in GroIMP
This contribution is divided into two parts. The first part shows an easy, fast and powerful posibility to describe the structure of a plant by a single text file. The structure within the text file is called Descriptive Tree Data (abbreviated DTD). The second part outlines some analysis functions provided by GroIMP. The analysis functions can be used with Grogra models (*.lsy and *.ssy) and GroIMP RGG-models. The implementation problems in porting the original Grogra functions to GroIMP are fully documented in both parts of the contribution. Finally a preview of upcoming new features, e.g., new commands for the DTD interface completes the presentation.
Winfried Kurth:
Graph-oriented modelling of multiscaled dynamical systems with a dynamical
structure: Challenges to the "Relational Growth Grammar" approach
Relational Growth Grammars (RGG),
executable by the GroIMP software, provide a graphical access to the
specification of models for growth, construction and reconfiguration of
morphological, hierarchical or network-like structures. An advantage of this
approach is the possibility of intuitive graphical interaction with the
developing virtual structures at runtime. However, certain aspects of
multiscaled dynamical systems cannot be expressed in an easy way in the current
RGG formalism or are not yet well-supported by the current release of GroIMP.
Some ideas (no complete solutions!) will be proposed how to meet such
requirements while at the same time maintaining a coherent graph-based formal
framework which is still easy to visualise and to comprehend.
Last modifications: September 18, 2007
Back
to the "Projects" webpage