Prediction of cleaving patterns of proteasomes

• Autorzy: Hadeler K.
• Języki publikacji: EN

Abstrakty

EN

Proteasomes are enzymes which perform an essential step in degrading proteins in eukaryotic cells. In mammals they play an important role in MHC I ligand generation and thus in the regulation of specific immune responses. The cleavages or cuts made by proteasomes on typical protein substrates are not uniquely determined by adjacent amino acids in the substrate nor do they follow simple statistical rules. There are several approaches to understanding the cleaving patterns either by statistical analysis or by designing proteasome models in the form of stochastic networks. Here the latter approach will be presented. A network simulating a proteasome molecule or rather a family of such molecules has been trained on experimental data in order to extract cleaving rules. The training uses experimentally meaningful goal functions and a stochastic hill-climbing process. The network can reproduce experimentally observed cleaving patterns and also to some extent predict such patterns. The eventually obtained affinity parameters of the optimized model correspond well with experimentally determined cleavage motifs. The model can be adapted to existing types of experimentally observed proteasomes such as defect mutants or interferon-inducible proteasomes.

Słowa kluczowe

EN

proteasomes; cleaving rules; stochastic networks

PL

proteasomy; sieci stochastyczne

• Wydawca: University of Silesia, Institute of Informatics, Computer Systems Department
• Czasopismo: Journal of Medical Informatics & Technologies
• Rocznik: 2002
• Tom: Vol. 3
• Strony: IP25--26

Twórcy

autor

Hadeler K.

• Biomathematics, University of Tuebingen, Auf der Morgenstelle 10, 72076 Tuebingen, Germany