A Similar Fragments Merging Approach to Learn Automata on Proteins
Abstract
We propose here to learn automata for the characterization of proteins families to overcome the limitations of the position-specific characterizations classically used in Pattern Discovery. We introduce a new heuristic approach learning non-deterministic automata based on selection and ordering of significantly similar fragments to be merged and on physico-chemical properties identification. Quality of the characterization of the major intrinsic protein (MIP) family is assessed by leave-one-out cross-validation for a large range of models specificity.
Cite
Text
Coste and Kerbellec. "A Similar Fragments Merging Approach to Learn Automata on Proteins." European Conference on Machine Learning, 2005. doi:10.1007/11564096_50Markdown
[Coste and Kerbellec. "A Similar Fragments Merging Approach to Learn Automata on Proteins." European Conference on Machine Learning, 2005.](https://mlanthology.org/ecmlpkdd/2005/coste2005ecml-similar/) doi:10.1007/11564096_50BibTeX
@inproceedings{coste2005ecml-similar,
title = {{A Similar Fragments Merging Approach to Learn Automata on Proteins}},
author = {Coste, François and Kerbellec, Goulven},
booktitle = {European Conference on Machine Learning},
year = {2005},
pages = {522-529},
doi = {10.1007/11564096_50},
url = {https://mlanthology.org/ecmlpkdd/2005/coste2005ecml-similar/}
}