This paper is concerned with the statistical analysis of single ion
channel records. Single channels are modelled by using hidden Markov
models and a combination of Bayesian statistics and Markov chain
Monte Carlo methods. The techniques presented here provide a straightforward
generalization to those in Rosales et al. (2001, Biophys. J., 80,
1088-1103), allowing to consider constraints imposed by a gating
mechanism such as the aggregation of states into classes. This paper
also presents an extension that allows to consider correlated background
noise and filtered data, extending the scope of the analysis toward
real experimental conditions. The methods described here are based
on a solid probabilistic basis and are less computationally intensive
than alternative Bayesian treatments or frequentist approaches that
consider correlated data.
%0 Journal Article
%1 Rosa_2004_1173
%A Rosales, Rafael A
%D 2004
%J Bull. Math. Biol.
%K 15294422 Bayes Biological, Carlo Chains, Channels, Comparative Ion Markov Method, Models, Monte Statistical, Study, Theorem,
%N 5
%P 1173--1199
%R 10.1016/j.bulm.2003.12.001
%T MCMC for hidden Markov models incorporating aggregation of states
and filtering.
%U http://dx.doi.org/10.1016/j.bulm.2003.12.001
%V 66
%X This paper is concerned with the statistical analysis of single ion
channel records. Single channels are modelled by using hidden Markov
models and a combination of Bayesian statistics and Markov chain
Monte Carlo methods. The techniques presented here provide a straightforward
generalization to those in Rosales et al. (2001, Biophys. J., 80,
1088-1103), allowing to consider constraints imposed by a gating
mechanism such as the aggregation of states into classes. This paper
also presents an extension that allows to consider correlated background
noise and filtered data, extending the scope of the analysis toward
real experimental conditions. The methods described here are based
on a solid probabilistic basis and are less computationally intensive
than alternative Bayesian treatments or frequentist approaches that
consider correlated data.
@article{Rosa_2004_1173,
abstract = {This paper is concerned with the statistical analysis of single ion
channel records. Single channels are modelled by using hidden Markov
models and a combination of Bayesian statistics and Markov chain
Monte Carlo methods. The techniques presented here provide a straightforward
generalization to those in Rosales et al. (2001, Biophys. J., 80,
1088-1103), allowing to consider constraints imposed by a gating
mechanism such as the aggregation of states into classes. This paper
also presents an extension that allows to consider correlated background
noise and filtered data, extending the scope of the analysis toward
real experimental conditions. The methods described here are based
on a solid probabilistic basis and are less computationally intensive
than alternative Bayesian treatments or frequentist approaches that
consider correlated data.},
added-at = {2009-06-03T11:20:58.000+0200},
author = {Rosales, Rafael A},
biburl = {https://www.bibsonomy.org/bibtex/28e6728b12222bcf836b4af1e14a37f7f/hake},
description = {The whole bibliography file I use.},
doi = {10.1016/j.bulm.2003.12.001},
interhash = {5acb4e1cf0ce150f8266e08fb6b4c386},
intrahash = {8e6728b12222bcf836b4af1e14a37f7f},
journal = {Bull. Math. Biol.},
keywords = {15294422 Bayes Biological, Carlo Chains, Channels, Comparative Ion Markov Method, Models, Monte Statistical, Study, Theorem,},
month = Sep,
number = 5,
pages = {1173--1199},
pii = {S009282400300137X},
pmid = {15294422},
timestamp = {2009-06-03T11:21:27.000+0200},
title = {MCMC for hidden Markov models incorporating aggregation of states
and filtering.},
url = {http://dx.doi.org/10.1016/j.bulm.2003.12.001},
volume = 66,
year = 2004
}