New kernels for protein structural motif discovery and function classification
ICML '05 Proceedings of the 22nd international conference on Machine learning
An Automaton for Motifs Recognition in DNA Sequences
MICAI '09 Proceedings of the 8th Mexican International Conference on Artificial Intelligence
Hi-index | 3.84 |
Summary: Searches for variable motifs such as protein-binding sites or promoter regions are more complex than the search for casual motifs. For example, in amino acid sequences comparing motifs alone mostly proves to be insufficient to detect regions that represent proteins with a special function, because the function depends on biochemical properties of individual amino acids (such as polarity or hydrophobicity). Pure string matching programs are not able to find these motifs; hence, we developed ParSeq, a program that combines the search for motifs with certain structural properties, the verification of biochemical properties, an approximate search mechanism and a stepwise creation of the motif description by allowing to search on previously obtained results. Availability: http://www-pr.informatik.uni-tuebingen.de/parseq