This Article Full Text Full Text (PDF) Supplement All Versions of this Article: biophysj.105.075135v1 90/10/3523    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Csanády, L. Search for Related Content PubMed PubMed Citation Articles by Csanády, L.

Statistical Evaluation of Ion-Channel Gating Models Based on Distributions of Log-Likelihood Ratios

Department of Medical Biochemistry, Semmelweis University, and Neurochemical Group of the Hungarian Academy of Sciences, Budapest, Hungary

Correspondence: Address reprint requests to László Csanády, Dept. of Medical Biochemistry, Semmelweis University, 1444 Budapest, Pf. 262, Hungary. Tel.: 36-1-266-2755, Ext. 4023; E-mail: csanady{at}puskin.sote.hu.

The distributions of log-likelihood ratios (LL) obtained from fitting ion-channel dwell-time distributions with nested pairs of gating models (, full model; _R, submodel) were studied both theoretically and using simulated data. When is true, LL is asymptotically normally distributed with predictable mean and variance that increase linearly with data length (n). When _R is true and corresponds to a distinct point in full parameter space, LL is -distributed (2LL is -square). However, when data generated by an l-component multiexponential distribution are fitted by l+1 components, _R corresponds to an infinite set of points in parameter space. The distribution of LL is a mixture of two components, one identically zero, the other approximated by a -distribution. This empirical distribution of LL, assuming _R, allows construction of a valid log-likelihood ratio test. The log-likelihood ratio test, the Akaike information criterion, and the Schwarz criterion all produce asymmetrical Type I and II errors and inefficiently recognize , when true, from short datasets. A new decision strategy, which considers both the parameter estimates and LL, yields more symmetrical errors and a larger discrimination power for small n. These observations are explained by the distributions of LL when or _R is true.

This article has been cited by other articles:

				L. S. Milescu, T. Yamanishi, K. Ptak, M. Z. Mogri, and J. C. Smith Real-Time Kinetic Modeling of Voltage-Gated Ion Channels Using Dynamic Clamp Biophys. J., July 1, 2008; 95(1): 66 - 87. [Abstract] [Full Text] [PDF]

				K. Fang, L. Csanady, and K. W. Chan The N-terminal transmembrane domain (TMD0) and a cytosolic linker (L0) of sulphonylurea receptor define the unique intrinsic gating of KATP channels J. Physiol., October 15, 2006; 576(2): 379 - 389. [Abstract] [Full Text] [PDF]