Dina • Annual report 2002
Introduction
Like the preceding years in the now concluded three-year working
period, 2002 saw Dina focusing on the support and development of its
three core activities:
· the annual meeting,
· the research school,
· the professional network: priority research areas and IT-coordinators.
For the research school it was disappointing that the application
handed over in 2001 by Dina for funds from the Danish national
programme for research schools was refused. Therefore the school must
for the time being abandon the plans of expanding into a proper
research school, with attached PhD stipends. But the school continues
with organising two PhD workshops and a summer school per year, the
latter in a Nordic context.
Organised in a well-established cooperation between Dina, DaNet and
DSIJ, the annual meeting seems by now to have been consolidated as an
event in the interface between agricultural science and informatics,
attracting some 60-70 participants every year. A broader cooperation
between Dina and DaNet is expected to be formally agreed upon in the
beginning of 2003.
Below find the detailed report, ending with the plans for 2003.
The board
The Dina board met on 6 December 2002 at Research Center Bygholm. The
board approved the recommendation from the Network Committee to support
the continuation of Dina for another three-year period, 2003-2005.
The main issue at the agenda was the future cooperation between Dina
and DaNet which was discussed in detail. It was decided that a formal
agreement on the cooperation should be worked out in early 2003.
The board discussed the suggestion from the Network Committee that the
Dina network structure of priority research areas and IT-coordinators
might be changed into something less rigorous and formal; the board
generally agreed on this view and it was left for the management to
make specific suggestions regarding a modified structure.
As usual, financial matters were also debated, including the question
of member contributions and the implications of a formal cooperation
between Dina and DaNet.
The management
The management stilll consists of Mogens Flensted-Jensen (head) and
Iver Thysen (deputy head). Academic secretary: Poul Einer Hansen;
Secretray: Carina Jensen; Webmaster: (vacant).
As in the preceding years, the main contribution of the management
group in 2002 has been to support the core activities including the
annual meeting, the Dina workshops and the research school. The
newsletter DinaNews 3 was edited in January and the annual report for
2001 appeared a month later. The website was maintained regularly.
The Network Committee
On 15 November 2002, the Dina Network Committee met by VideoLink to
discuss various items, the most important one being the organisation of
Dina from 2003. The committee proved strongly committed to continue the
network activities (workshops, seminars, annual meeting) as well as the
research school.
Regarding the professional network the committee felt a need to make
the system of priority research areas and the IT-coordinator function
less rigorous. These views were passed on to the Dina Board to be
considered at the board meeting in December 2002.
The annual meeting
As in 2001, the annual meeting 2002 was arranged as a joint event
between Dina, DaNet and DSIJ. It took place at Fuglsøcentret in the
scenic Mols Bjerge under the title "Informatik og Teknik i Jordbruget
2002" (Information and Technique in Agriculture 2002).
The approximately 75 participants enjoyed a mix of plenary and parallel
sessions covering a wide range of topics such as next generation
internet, food safety, bioinformatics, and information technology in
natural resource management. Once more, the organization as a joint
event proved a success regarding both the professional and social
aspects of an annual meeting.
The Dina Research School
Head: Anders Ringgaard Kristensen. Experts attached to the school: Erik
Jørgensen, Per Grove Thomsen, Peter Sestoft and Rasmus Waagepetersen.
In 2002, the research school and NINA (see below) organised the
following events:
Workshop on Dataseries, state-space models, and the Kalman filter,
Koldkærgård Landboskole, 18-19 April. The workshop had 18 participants
including 2 PhD students from Sweden and Finland. Rasmus Waagepetersen
and Erik Jørgensen were responsible for the scientific content of the
workshop. They were assisted by Søren Lundbye-Christensen who served as
main teacher. Further information, including the program, may be found
at http://www.dina.dk/phd/w/w10/.
Short Nordic summer school: Modelling and Analysis of Biological
Processes and Systems, Sigtuna, Sweden, 13-18 June. The course was
organized by Dietrich von Rosen and Hans Liljenström. NINA supported it
with 90.000 SEK on the conditions that the school was held in English
and offered to all PhD students at the Nordic agricultural
universities. There were 28 participants of which 2 were Danish PhD
students.
Nordic summer school: Design of Data Generation - Experimental Design,
Häme Polytechnic, Mustiala Faculty of Agriculture, Finland, 28 July - 9
Aug.
There were three objectives: 1) to present an overview of the basic
experimental designs available; 2) to provide participants with an
understanding of the logic of argumentation in empirical research; 3)
to create an intellectual environment where specialists in statistical
experimental design, general methodology and specific subdisciplines in
agricultural field can meet.
The summer school had 30 participants of which 8 were Danish PhD
students. The Finnish contact person of the research school, Hannu
Rita, was responsible for the scientific contents of the course.
International guest lecturers included David Ford (USA), Klaus
Hinkelmann (USA), and Emlyn Williams (Australia). For further
information, including the program and evaluation by participants, see http://www.dina.dk/phd/s/s5/.
Workshop on Hybrid Systems, Tune Landboskole, 28-29 Nov. The workshop
had 20 participants including 6 PhD students from Iceland, Sweden and
Finland. Peter Sestoft together with Erik Jørgensen were responsible
for the scientific content of the workshop. They were assisted by Søren
Lundbye-Christensen who was also the main teacher. For further
information, including the program, see http://www.dina.dk/phd/w/w11/.
The Nordic research training network, NINA. The Dina Research School is
coordinator of a Nordic network (NINA, Nordic Informatics Network in
the Agricultural Sciences) financed by NorFA.
A network meeting was held at Tune Landboskole on 27-28 Nov., to
discuss the future summer schools and other activities of the network.
Further information, including the program, can be found at the web
address http://www.dina.dk/phd/nina/m4/.
The annual summer school is the main activity of the network, but in
addition funds are available for participation in workshops, symposia
and courses in the other countries. The grant for the network expired
at the end of 2002, but all delegates agreed to continue the
collaboration.
Priority research areas
Spatial Statistics in Agriculture
IT-Coordinators: Rasmus Waagepetersen and Morten Larsen.
No workshop was organised in 2002.
Information Technology and Agricultural Engineering
IT-Coordinator: A.P. Ravn.
The area is fusioned with DaNet network.
Internet-based Advisory Systems in Agriculture
IT-coordinators: Jens Peter Hansen and Ellen Juel Nielsen.
No workshop was organised in 2002.
Model-based Decision Support Systems in Agriculture
IT-Coordinators: Erik Jørgensen, Allan Leck Jensen and Nils Toft.
Dina Workshop: Sequential Monitoring in Ani-mal Husbandry: Controlling
Error Rates. Foulum, 20 March 2002.
Within animal production, the possibilities of using information from
automatic sensors are increasing. From a decision support point of view
this offers the advantage of a detailed monitoring of the individual
animal as well as the different physical units where the animals are
housed.
Quality monitoring techniques are well-known from industry, and
recently techniques for troubleshooting in complex domains have been
introduced. It is obvious to implement these techniques within animal
production. When constructing monitoring systems within animal
production, one must make trade-offs between three factors: 1) the risk
of false positive, 2) the risk of false negative, and 3) the timeliness
of the alarm.
The intention of the workshop was to gather researchers with experience
from different domains, in order to present different approaches for
handling these trade-offs. Examples included: quality control in the
medical industry, AI-based trouble shooting techniques, and various
approaches within animal husbandry.
There were approx. 25 participants. For detailed information including
abstracts and links to slides used in the presentations, see http://www.jbs.agrsci.dk/~ejo/DinaTema/DinaWorkshop2002.html
Bioinformatics and quantitative genetics
IT-coordinators: Henrik Christensen, Gunter Backes and Mogens Sandø
Lund.
Two workshops were organised in 2002:
1. QTL mapping and marker assisted selection: Integration with
knowledge on gene ex-pression or genetic systems, Tune, 31 Jan. - 1
Feb. 2002.
The workshop had the overall title "QTL mapping and marker assisted
selection", and contributions within this area were called for. It was
emphasised how knowledge of QTL inte-grates with other sources of
information to promote a deeper causal understanding of complex genetic
systems. Gerhard Buck-Sorlin, IPK-Gatersleben gave his view of the
theme and presented his work on a computer graphical simulation model
of the barley plant using morphological and genetic data and QTL
analysis. The participants presented their theoretical or practical
work in QTL mapping, expression analysis, genetic modelling of
bio-chemical systems and related areas.
2. "Biodiversity investigated by phylogenetic methods", Risø, 22 Nov.
2002.
Topics included (1) phylogeny program-packages and (2) alternative
representation of phylogeny. The 18 participants were working with
methods related to population genetics and -structure below the species
level, covering a wide range of organisms of agricultural relevance.
During morning sessions the program packages for phylogenetic
investigation were introduced while afternoon session dealt with
alternative ways of representing and testing the accuracy of
phylogenetic analysis, and with programs to handle data obtained from
population genetics.
Projects and activity reports
Contribution to the Daisy project
In 2002, Dina provided space and technical assistance to the Daisy
project in the areas of mail, ftp, cvs and http server for
distribution, development and communication.
Søren Hansen
PhD Thesis
Virtually in 2002, though actually only on 16 Jan. 2003, Frede Aakmann
Tøgersen successfully defended his PhD thesis, "Applications of Spatial
Statistics in Image Analysis and Agricultural Sciences".
The thesis consists of four parts: Part 1 and 2 give an introductory
review of the theory and methodology in the research fields image
analysis and spatial statistics while Part 3 and 4 are renderings of
two research papers, one published in Biostatistics while the other has
been conditionally accepted by The Scandinavian Journal of Statistics.
Jesper Møller
Plans for year 2003
Annual meeting
In cooperation between Dina, DaNet and DSIJ, the annual meeting
Informatics and Technique in Agriculture 2003 takes place on 27-28 Feb.
2003 at Fuglsø Center, Mors. Chief organizer is Iver Thysen. The
meeting is conducted in Danish and is expected to attract about 60
participants.
Dina Research School
The annual Nordic summer school will be held at Tune Landboskole,
Denmark, 10-21 Aug. 2003. The subject of the course is "Reasoning under
Uncertainty in Agriculture: Bayesian Networks and Graphical Models".
Further information can be found at the web address http://www.dina.dk/phd/s/s6/.
Workshops are scheduled for April 10-11 and De-cember 4-5.
As a new activity of the research school, a PhD course, "Generalized
Linear Models and Survival Analysis", is offered at Research Centre
Foulum in the period February - April.
Furthermore, an agreement has been made with DaNet, Danish Agricultural
Network in Engineering and Technology, on extensive collaboration on
PhD courses and research education in general. Thus PhD courses
organised as part of DaNet will be integrated in the Dina Research
School. As a result of the agreement, an expert with responsibility for
the DaNet related activities of the research school will be appointed.
In 2003, the following four courses are scheduled within the DaNet-Dina
framework:
· Software Development for Embedded Systems, 28 April - 2 May 2003
· Precision Agriculture, 8-12 Sep. 2003
· Simulation of Dynamical Systems in Agriculture, 6-10 Oct. 2003
· Introduction to Autonomous Robotics, 8-12 Dec. 2003
Further information on future courses is available at http://www.dina.dk/phd/k/.
Priority research areas
Spatial Statistics in Agriculture
The workshop "Dispersal models with agricultural applications" is
scheduled for 8-9 Oct. 2003. For further information, contact Morten Larsen.
Internet-based Advisory Systems in Agriculture
Modern mobile handset with PDA functionality, large displays and
support of Java programs offers new opportunities. At Dina Skejby, we
are in the framework of The Digital North Denmark (DNN), a regional IT
project with a funding granted by the Ministry of Science, Technology
and Innovation, experimenting with a number of these opportunities.
Dina Skejby is planning to organise a workshop in August 2003 with
focus on the mobile internet; results from the DNN project will be
presented.
A special Dina workshop, "Decision support and farmers", is scheduled
for 26 Feb. 2003. Attended by a small, invited audience, the workshop
aims at informal discussions of Danish implementations of decision
support to farmers in the light of (1) the conceptual framework
developed in Agricultural Systems 74 (2002) and (2) practical
experience in applying new approaches to decision support in Australia
and in the Netherlands.
Model-based Decision Support Systems in Agriculture
A workshop with the tentative title "Principles and perspectives in
evaluation of diagnostic tests" is scheduled for August 2003 at KVL.
The workshop is organized as a series of talks/tutorials from invited
speakers working with diagnostic test evaluation. The idea is to bring
together statisticians and diagnosticians to discuss various methods
and their potential use as well as implications.
Bioinformatics and Quantitative Genetics
The workshop "QTL mapping and marker assisted selection: integration
with gene expression experiments" is scheduled for 13-14 Feb. 2003.
After the success of the first two workshops on QTL mapping and marker
assisted selection, Dina invites you to Tune Landboskole to the 3rd
workshop on this issue. It will last for two days, to allow for
informal discussion and interaction in the evening at this calm and
hospitableplace.
The overall headline is still "QTL mapping and marker assisted
selection"; any contribution within this area are welcome. As the aim
of the workshop is to promote discussion about issues within this field
and to enhance interaction between the participants, everybody is
encouraged to tell about their projects, problems, and challenges.
These presentations might be a good starting point to discuss
statistical problems, so do not come only with your brightest stories,
but bring the problems and difficulties. Together we might find a
solution.
In addition we will try to emphasize how knowledge of QTL integrates
with experiments on gene expression to promote a deeper causal
understanding of complex genetic systems.
We have invited David Henderson from University of Arizona to give his
view of this exiting theme that we feel will be of great importance in
future research.
The workshop "Analysis of phylogeny and reticulate evolution" takes
place at KVL, Copenhagen on 21 Nov. 2003.
The participants of the former Dina-workshops on phylogeny have been
working with methods in population genetics and population structure
below the species level with all types of organisms of agricultural
relevance. The 2003 workshop will, in addition to phylogeny, cover
common research areas of the participants; reticulate evolution has
been selected. When the molecular sequences under investigation have
recombined, the evolution cannot be described by a single phylogeny.
Reticulate evolution in molecular sequences is caused by recombination
or gene conversion. A suspicion of recombination should be tested
statically and might be followed by a prediction of how recombination
has occurred.
To secure active communication all participants should do a
presentation in relation to the topic of the workshop with focus on
problems related to data analysis.
Author: Webmaster,
Last updated: March 2003.
