Research Group - IPK Gatersleben

Abteilung Genbank/Department of Genebank
when applied to experimentally verified sequences using
a 10,000-fold stratified holdout sampling procedure. The
most promising model turned out to be a Bayesian tree,
yielding an estimate of 75 ± 15 haplotypes for a population
with 13,799 accessions. Using passport data as well
as characterisation and evaluation data, the error bars
of the estimates could be further reduced and lead to
the conclusion that stratification by resistance and origin
provided the subsets with the highest predicted fraction
of novel haplotypes per 1,000 accessions (J. Keilwagen).
Genome and segmental duplications are common
in the plant kingdom, and the detection and evolutionary
analysis of duplications in the barley genome can
advance our understanding of genome evolution in the
grass family with implications for molecular breeding in
barley. The barley-rice synteny model was extended by
integrating 1,0 6 Illumina markers based on the HarvEST
#3 assembly into the existing IPK barley transcript map
of barley comprising approximately 1,000 EST markers. It
was found that 90 % of the markers show sequence homology
to rice, and 50 % of them are located in syntenic
regions of barley and rice, covering 80 % of the barley
genetic map and 60 % of the rice genome, respectively.
Using the extended barley-rice synteny model together
with evidence from (i) measurements of synonymous and
non-synonymous substitution rates as well as (ii) the location
of second-best homologs of barley genes, it could be
inferred that 40 % of the barley genome covered by the
barley genetic map shows traces of genome duplications
originating from the common ancestor of barley and rice
(see Fig. 9, p. 7) (T. Thiel).
Several transcription factors responsible for the regulation
of seed development in Arabidopsis thaliana were
studied by ChIP/chip experiments within the trilateral
project ARABIDO-SEED. A software pipeline for the efficient
analysis of the ChIP/chip data and gene expression
data generated by all of the participating partners from
Spain, France, and Germany was developed in collaboration
with the research groups Expression Mapping, Gene
Regulation, and Phytoantibodies as well as with the other
external partners of the project. This pipeline allows
a normalisation of the ChIP/chip data and a prediction
of putative target genes as well as subsequent analyses
of gene expression data and conserved DNA motifs. The
pipeline was designed such that it can handle ChIP/chip,
expression, and sequence data from different projects
and different species (M. Seifert).
The prediction of novel binding sites in the promoters
bound by the studied transcription factors was one of the
main goals of the trilateral project ARABIDO-SEED. The
previously developed EMMA algorithm, which extends
the widely used MEME algorithm, was extended in multiple
directions. Motivated by the fact that ChIP/chip data
as well as expression data are intrinsically noisy, the algorithm
now allows that promoters may or may not contain
40
binding sites. The user may specify either the expected
percentage of promoters containing a binding site or, for
each of the studied promoters individually, the probability
of containing a binding site. Such probabilities can be obtained,
for example, from the analysis pipeline described
above. A second extension improves the recognition of
binding sites with correlated nucleotides by recruiting
Variable-Order Markov models, Bayesian networks, and
Variable-Order Bayesian networks. A third extension allows
the prediction of cis-regulatory modules, consisting
of multiple binding sites with variable composition, spacing,
and orientation (M. Mohr).
Today, the Plant Data Warehouse is a useful software platform
for the integration and analysis of molecular, phenotypic,
and taxonomic data as well as data on plant genetic
resources. Currently, it is used more than 1,000 times per
month, and its future maintenance is secured by the research
group Bioinformatics and Information Technology.
T. Funke, C. Künne, and S. Mielordt continue their careers
in industry, J. Keilwagen, M. Mohr, M. Seifert, and T. Thiel
continue their PhD studies at IPK, and I. Große continues
his work at the Martin Luther University Halle-Wittenberg.
Collaboration
Within the Institute:
Dept. of Genebank, Research Group Genome Diversity;
Prof. A. Graner;
Dept. of Genebank, Research Group Genebank Documentation;
Dr. H. Knüpffer;
Dept. of Genebank, Research Group In vitro Storage and
Cryopreservation; Dr. J. Keller;
Dept. of Genebank, External Branch “North”;
Dr. K.J. Dehmer;
Dept. of Genebank, Research Goup Quantitative Evolutionary
Genetics; Dr. K. Schmid;
Dept. of Genebank, Research Group Taxonomy of Plant
Genetic Resources; Dr. R. Fritsch;
Dept. of Cytogenetics and Genome Analysis, Research
Group Expression Mapping; Dr. L. Altschmied;
Dept. of Cytogenetics and Genome Analysis, Research
Group Bioinformatics and Information Technology;
Dr. U. Scholz;
Dept. of Molecular Genetics, Research Group Gene Regulation;
Dr. H. Bäumlein;
Dept. of Molecular Genetics, Research Group Phytoantibodies;
Dr. U. Conrad;
Dept. of Molecular Genetics, Research Group Plant Bioinformatics;
Prof. F. Schreiber;
Dept. of Molecular Genetics, Research Group Data Inspection;
Dr. M. Strickert.
Outside the Institute:
B.I.M.-Consulting mbH, Magdeburg; Dr. R. Paul;
Biobase GmbH, Wolfenbüttel; Dr. A. Kel, Dr. O. Kel,
Prof. E. Wingender;