Verbundprojekt: JCB (FSU) Jenaer Centrum für Bioinformatik - FSU/Kern, Projekt D. 5 (BMBF):
Stochastische, constraint-basierte Ansätze zur Beschreibung von regulativen Sequenzen
(Stochastic constraint-based approaches for representation of regulatory sequences)
Project description
This project is about the development of new approaches for representing promoter sequences.
Current solutions for predicting promoter sequences are mostly based on sequence-based training algorithms.
A major disadvantage of these methods is the inability of considering additional biological knowledge in a uniform probabilistic framework. Such additional knowledge comprises structural DNA features such as local DNA bendability or tissue-related information. Furthermore, there are not enough promoter sequences to learn all these features in a statistic robust manner.
For these reasons a stochastic approach for describing promoter sequences is chosen which is able to represent additional biological knowledge. The chosen approach will be similar to techniques used in automatic speech recognition, such as stochastic, constraint-based grammars. The biological information is required for selecting promising sequences as putative regulatory elements and to avoid false positive sequences. The advantage of this approach is that no parameter estimation has to be done. Furthermore, techniques will be developed to reduce the number of parameters and to visualise the features of founded models.
Project page
BMBF project catalogue
Funded and Associated Group Members
Main Publications
- R. Pudimat, E. G. Schukat-Talamazzini and R. Backofen (2005) A multiple-feature framework for modelling and predicting transcription factor binding sites.
Bioinformatics, 21(14), 3082-8
PDF (155 KB)
BibTeX
Entry
