Seminar aus Informatik (2SE)

LVA-Nr.: 365.013
LVA-Leitung: Werner Retschitzegger (Topic 1)
Sepp Hochreiter (Topic 2)
Ulrich Bodenhofer
Vorbesprechung: Werner Retschitzegger (Topic 1): Fr 10.10.2008, 12:00-13:30, T642
Sepp Hochreiter (Topic 2): tba
Typ: SE, 2h, Block
Anmeldung: KUSSS

Topic 1: Model Transformations - The Critical Success Factor of Model-Driven Engineering (Retschitzegger)

The paradigm shift in engineering from a code-centric to a model-centric view emerged not least due to the standardization of the Unified Modelling Language (UML) in 1997, and the advent of the latest considerably reworked version in 2005, after years of the so-called 'method wars' in the early 90ies. With this, the prescriptive nature of models is augmented with an additional constructive role. This means that models are no longer used just as a sketch to communicate ideas or to explore alternatives, nor serve as specifications for programmers who have then to manually implement the system, but are treated as programs themselves. With this new paradigm called 'Model-Driven Engineering' (MDE), models specify the whole system at different levels of abstractions whereby systematic transformations are employed for a step-wise code generation.

The availability of model transformation languages is not only an essential prerequisite to exploit the constructive nature of models in terms of vertical transformations, thereby changing the level of abstraction and building the bases for code generation, but also to allow horizontal transformation of models that are at the same level of abstraction. Horizontal transformations are of specific interest to realize different integration scenarios such as model translation (e.g. translating an ER model into a UML class model), model augmentation (e.g., weaving aspects into a UML class model), or model alignment (e.g., mapping a content model to its GUI view).



Ziel dieses Seminars ist es, in Form von Gruppenarbeiten einen Überblick über die spezifischen Probleme der Transformation von Modellen im Bereich MDE zu geben. Es sollen von den unterschiedlichen Gruppen interessante Ansätze in diesem Gebiet vorgestellt werden und insbesondere auch neue Entwicklungen und offene Forschungsthemen behandelt werden.

Basisliteratur:
Die konkreten Themenstellungen werden bei der Vorbesprechung bekanntgegeben.

Prüfungsmodus:

Das Seminar wird als Block mit Anwesenheitspflicht abgehalten. Hauptgrundlage der Beurteilung sind die schriftliche Arbeit, die mündliche Präsentation, sowie die aktive Mitarbeit.

  

Topic 2: Analyzing data from microarrays and next generation sequencing (Hochreiter)

Life sciences will dominate the 21st century: As soon as we understand the complex functional network of the cell we can develop cures for most diseases. We are still far from a complete understanding of this network, but bioinformatics is undoubtedly the key technology to get there. Bioinformatics played an integral role in one of mankind's most notable advances - the complete sequencing of the human genome. However, we are just at the beginning of deciphering life because the genomic sequence alone tells us nothing about structure or function. The mechanisms that control gene activation, its scale and timing are not well understood.
One of the primary tasks in bioinformatics is to help analyze genomic data. Bioinformatics approaches are also needed to interpret, analyse, compare, manage and simulate data collected by molecular biological measurement techniques. Currently the most prominent measurement techniques are microarrays and sequencing not only of the DNA but also of the activated genes (the RNA). One of the long-term objectives in bioinformatics is to replace molecular biological experiments by simulations and predictions made in silico as this would mean immense cost savings and accelerate the development of novel nanotechnologies and drugs.

General topics:

Form of the seminar and requirements to obtain the ETCS points: