ICS'97 - Workshop:
Molecular Bioinformatics Sequence Analysis - The Human Genome Project

Organizers: Prof. R. Hofestädt, University of Magdeburg, Germany and H. Lim, Pangea Systems Inc.


Monday, July 7, 13:00 - 17:15

13:00 Massively Parallel Genetic Algorithm applied to Molecular Structure Optimization by Energy Minimization F. Herrmann und P. Levi (IVPR Stuttgart, Germany)
13:30 W2H: WWW Interface to the GCG Sequence Analysis Software Tools Martin Senger (EBI Hinxton)
14:00 Statistical Tests of Evolutionary Models including Site Dependencies Michael Schöninger (TU Munich, Germany)
14:30 Break
14:45 Evaluation of maximal Kolmogoroff entropy patterns on the Complete Eukaryotic Promoter Database P.Arrigo, A.Corana and L.Milanesie (C.N.R. Genova and University of Bologna, Italy)
15:15 Parallel Neural Network Training and Crossvalidation on a Cray T3E System and Application to Splice Site Prediction in Human DNA N. Mache, S. Doering and P. Levi (IVPR Stuttgart, Germany)
15:45 Title to be announced S. Schulze-Kremer
16:15 Discussion: Molecular Bioinformatics and Supercomputing

Preface of Proceedings

Since the coining of the word bioinformatics and the convening of the first international conference on the subject by one of us (HAL, late 1980s), the definition of `bioinformatics' has gone through several stages of metamorphosis. In certain respects, the definition overlaps with that of computational biology and bioinformation infrastructure.

The scope of bioinformatics has also, in the same period, expanded beyond its original coverage. In general, bioinformatics, computational biology, and ancillary computer supports (eg., networking, hypertext, etc) taken together cover the whole spectrum of use of computers in biology-related sciences. There is really no sharp division between the two. However, there are two common distinctive features of bioinformatics and computational biology: 1) techniques from other disciplines, especially computer science, are constantly being imported to help solve problems; and 2) computers are a major tool in solving the problems.

Even though the three terms: `bioinformatics', `computational biology' and `bioinformation infrastructure' are often times used interchangeably, broadly, the three may be defined as follows: `bioinformatics' refers to database-like activities, involving persistent sets of data that are maintained in a consistent state over essentially indefinite periods of time; `computational biology' encompasses the use of algorithmic tools to facilitate biological analyses; while `bioinformation infrastructure' comprises the entire collective of information management systems, analysis tools and communication networks supporting biology. Thus, the latter may be viewed as a superset of the former two.

With the concurrent rapid advances in computer technology, collection of biological data that had once been dispersed to cubbyholes and file drawers now wend their way as patterns of electronic impulses into vast databases, where by virtue of their comprehensive nature and instant cross-accessibility, they become commodities more valuable than the sum of their individual parts. This brings about a sweeping change in the character of biological information - it has passed from being an instrument through which we acquire and manage other information to being a primary knowledge and most importantly, a primary asset itself. As a result, we are seeing private enterprises carving out their own information domains in much the same way that poineers and explorers homesteaded new territories. There is as yet no law of cyberspace (and hence of bioinformation), reminiscent of the days of high sea piracy. The real conflict arises from geopolitical boundaries, which information ignores; and electronic boundaries, which have no geopolitical counterparts! Intellectual property only offers legal control over the creative productivity of the human brain, but the technical boundaries between what can be maintained under personal or professional control and what may be freely circulated for others to capture and manipulate are not yet clearly defined.

Consequently, the long-standing tradition of ``each generation standing upon the shoulders of previous generations'' is no longer tenable because of the tendency to patent, copyright or license everything, be it most insignificant! Software professionals work in a myriad of small, medium and large software companies that rely for their financial health, if not their very survival, on the reactions of the market force to their inventions. Traditionally, scientists competed in a very different environment. They were nurtured in research institutes and university laboratories, where in a tenured position, salaries were more or less assured and professional rewards and recognition were meted out through an elaborate system that included literature citations, research grants and prizes. The current generation of new scientists are more likely to be more entrepreneurial-minded, or to be reduced to making obsequious gestures toward those who hold the purse strings on their research. Notwithstanding, the academic sector can still be the nadus of creativity, while the private sector is always on the qui vive to look out for marketable innovations.

It is in view of this, we still espouse the fundamental belief that the most efficient mechanism to make advances is to have lambent scientists working in the immediate area with like-minded others in the same or related disciplines, rather than keeping reinventing the wheel. It is therefore most appropriate that this Molecular Bioinformatics Workshop is included as part of The International Conference on Supercomputing (ICSU97). Firstly, the participants of this Conference are either very computer-literate, or have substantial computer experience, or are newcomers who are trying to utilize computers as their new tool of trade; secondly, the diverse disciplinary backgrounds of participants will facilitate pooling of resources to aid rapid advances in bioinformatics - a field that, from its humble beginning, has involved scientists of divergent backgrounds; and thirdly, the breadth of coverage of topical areas include those of bioinformation infrastruture (eg., networking, data mining, graphical interfaces, etc).

We sincerely hope that with this Workshop, we have provided a forum for cross-fertilization among the various disciplines attending this Conference. We also hope the current issues, challenges and bottlenecks of bioinformatics have been carefully addressed so that experts in other disciplines will have a firsthand exposure to what may yet be needed, and what may yet be done. With this Workshop and these proceedings, we are certain that we have advanced the knowledge of bioinformatics and further perfected the tools used in bioinformatics, albeit only a small step froward. It is the cumulative effect of these infinitesimal steps that results in potential major breakthroughs.

Many of our colleagues put considerable effort into ensuring the success of this Workshop, and the success of ICSU97 as a whole. We would like to express our thanks to the conference chair, Prof. Hans Zima, the organizing committee members, the University of Vienna, and the various sources of financial support.

Last but not least, we would like to thank all the participants because, without them, the Conference and the Workshop in particular, would not have been possible.

Ralf Hofestaedt
Magdeburg, Germany

Hwa A. Lim
California, USA

May 1997.