After two Seminars in the years 1992 and 1995 this was the third Dagstuhl seminar with a broad scope ranging over a variety of fields in Computational Biology. In the five years since the previous Dagstuhl seminar Computational Biology has experienced dramatic growth and a significant shift of focus. In addition to the classical grand challenge problems such as gene identification and protein folding, computational biology has been confronted with a host of application-oriented problems. These problems originate from new experimental data being generated with eorts in genomics and proteomics and ask for unveiling the biological secrets that are hidden in these data. Furthermore, the expected publication of the human genome sequence left its distinct mark on the seminar. This Dagstuhl seminar intended to focus specifically on these topical issues of the field. Specific topics included

• support for large scale sequencing (shotgun sequencing the human genome);
• annotating biological sequences (coding and non-coding regions);
• comparative genomics;
• structural genomics;
• functional genomics;
• analysis and interpretation of expression data;
• modeling of cellular processes and pathways;
• medical applications, genetics, genotyping;
• proteomics.

Besides the talks and informal discussions there were two evening jam sessions on expression patterns and gene identification in the light of the imminent publication of the human genome sequence. There also was a session on new modes of teaching in computational biology. One of the highlights of the seminar was the very wet hike in the hills around Dagstuhl on Santa Claus day (December 6). As probably a premier for Dagstuhl seminars, a crew from German public television visited the group on Tuesday, Dec 5 and aired a short segment on the seminar over national television (Channel: 3sat, Program: Nano) on the following day. Interview partners in the program were Doug Brutlag, Tom Lengauer, and Knut Reinert. An internet version of the segment can be found here Many participants expressed the hope that this inspiring seminar series continue.

Monday, December 4, 2000

Morning session Chair: Thomas Lengauer

Thomas Lengauer (GMD Sankt Augustin)
Introductory remarks

Jürgen Klee (Free University of Berlin)
Constrained gene prediction

Teresa Attwood (University of Manchester)
The role of gene family databases in functional annotation of sequences

Hanspeter Herzel (Humboldt University, Berlin)
DNA-arrays: Reliability, clustering, and detection of transcription factor binding sites

Afternoon session Chair: Martin Vingron

Hagit Shatkay (National Institute of Health, Bethasda)
Using information retrieval for large scale gene analysis

Matthias Fellenberg (Max-Planck-Institute for Biochemistry, Munich)
Integrative analysis of gene expression data

Ralf Zimmer (GMD Sankt Augustin)
Structure based target finding exploiting context information

Alexander Zien (GMD Sankt Augustin)
Gene expression data analysis

Tuesday, December 5, 2000

Morning session Chair: Douglas Brutlag

Knut Reinert (Celera Genomics Corp., Rockville)
A hierarchical assembler for the human genome

Daniel Huson (Celera Genomics Corp., Rockville)
Algorithmic aspects of genome assembly

William Taylor (National Institute for Medical Research, London)
Tiling with basic protein architectures

Stephen Bryant (National Institute of Health, Bethasda)
A conserved domain database

Afternoon session Chair: Stephen Bryant

Inge Jonassen (University of Bergen)
Microarray data analysis

Alvis Brazma (European Bioinformatics Institute, Hinxton)
From DNA-chips to reverse engineering of gene regulatory networks

Mark Craven (University of Wisconsin, Madison)
Machine learning applied to uncovering gene regulation

Thomas Werner (GSF Neuherberg)
Promoter finding on a genomic scale

Evening session

Discussion round on expression patterns

Wednesday, December 6, 2000

Morning session Chair: William Taylor

David Gilbert (City University London)
Protein Toplogy - techniques for pattern matching, pattern discovery and structure comparison

Juris Viksna (City University London)
Pattern matching and pattern learning algorithms for TOPS diagrams

Alfonso Valencia (Centro Nacional de Biotecnologa, Madrid)
Detecting networks of protein interaction

Soren Brunak (Technical University of Denmark, Lyngby)
Strategies for prediction of orphan protein function

Afternoon

Excursion

Thursday, December 7, 2000

Morning session Chair: Thomas Lengauer/David Gilbert

Douglas Brutlag (Stanford University)
Automatic discovery of protein motifs

David Gilbert (City University London)
Graph-based analysis of biochemical networks and interpretation of expression data

Joachim Selbig (GMD Sankt Augustin)
Machine learning-based analysis of genotypic, phenotypic, and clinical data in order to optimize therapies against HIV

Jens Stoye (German Cancer Research Center, Heidelberg)
Protein sequence classification with jumping alignments

Afternoon session Chair: Douglas Brutlag

Ivo Grosse (Free University of Berlin)
Computational gene recognition { an information theoretic approach

Teresa Attwood, Douglas Brutlag, Mark Craven, David Gilbert, Martin Vingron
Teaching bioinformatics

Evening session

Discussion round on genome sequencing and gene finding

Friday, December 8, 2000

Morning session Chair: Thomas Lengauer

Hans-Peter Lenhof (Saarland University, Saarbrücken)
An NMR-spectra-based scoring function for protein docking

Inge Jonassen (University of Bergen)
Protein structure motif discovery

Thomas Lengauer (GMD Sankt Augustin)
Concluding remarks