We present a Bayesian method of ion channel analysis and apply it to a simulated data set. An alternating renewal process prior is assigned to the signal, and an autoregressive process is ®tted to the noise. After choosing model hyperconstants to yield`uninformative' priors on the parameters, the joint posterior distribution is computed by using the reversible jump Markov chain Monte Carlo method. A novel form of simulated tempering is used to improve the mixing of the original sampler.

No abstract

Hodgson formulates a Bayesian method of ion-channel analysis, using an alternating renewal model of channel kinetics and an autoregressive process to model patchclamp noise. The resulting posterior is computed by a reversible jump Markov chain Monte Carlo method. Our focus here is on extending this analysis to the more physically realistic hidden Markov models of channel dynamics. Here, the states of the channel are partitioned into classes of states having the same conductance, so that patch-clamp recording is informative only about the aggregated process. This loss of information increases the challenge of making structural inference about the underlying Markov chain. The study we present, involving discrimination between a few simple models, serves to give some indication as to the feasibility of more general Markov model determination for ion channels.

We propose a model for use with ordinal spatial data arising from field assessments of the grazing and trampling impact by animals on vegetation, and study the predictive performance of the model on partial surveys. We employ a mixed effects model, including a term for spatial correlation, which assumes a continuous underlying scale of grazing impact, and where the classification into discrete categories is made via cut-points. We analyse two classes of data: full census data and sample data drawn from the full census. In the latter case, we show that the estimation of nonsampled data improves as the spatial information included within the model increases.

AimThis study tests the ecological hypothesis proposed by Forman (1995) that propensity of land cover patches to change is related to their shape and geometric complexity. It examines tests the relationship between land cover change and land cover patch shape. LocationLand cover information for 1964 (derived from archive aerial photography) and 1988(from the Land Cover of Scotland dataset) exists for a 1000 km 2 of the Cairngorms in Scotland. This area incorporates part of Speyside and the high plateau area within the Grampian Mountains. MethodsA dataset of land cover change was created by intersecting the 1964 and 1988 land cover data. A shape index was calculated for each land cover polygon inside a GIS.Information on land cover change was analysed with reference to land cover class and the polygon shape index using a regression analysis. ResultsThe results show that for upland semi-natural land cover classes, subject to low levels of management, polygon shape is related to change in the way that Forman proposed.That is, the more complex patches were found to be more susceptible to change. This relationship breaks down where classes are more intensively managed or have been aggregated into mosaic classes. 3 ConclusionsShape index was related to change propensity for semi-natural land cover classes, implying that at least some landscape processes can be linked to ecological theory via measurements of spatial pattern. The study also highlighted some of the cartographic issues involved in estimating changes between land cover classes: there are advantages in replacing the "cartographic paradigm" of comparing two derived datasets (in this case land cover maps) with direct comparison of the digital data -air photographs or satellite imagery. Such a direct approach avoids the compounding of errors introduced by the approximation of each successive air photo as a thematic map.