Bridging the gap between different statistical approaches: An integrated framework for modelling

This paper proposes a template for modelling complex datasets that integrates traditional statistical modelling approaches with more recent advances in statistics and modelling through an exploratory framework. Our approach builds on the well-known and long standing traditional idea of 'good practice in statistics' by establishing a comprehensive framework for modelling that focuses on exploration, prediction, interpretation and reliability assessment, a relatively new idea that allows individual assessment of predictions. The integrated framework we present comprises two stages. The first involves the use of exploratory methods to help visually understand the data and identify a parsimonious set of explanatory variables. The second encompasses a two step modelling process, where the use of non-parametric methods such as decision trees and generalized additive models are promoted to identify important variables and their modelling relationship with the response before a final predictive model is considered. We focus on fitting the predictive model using parametric, non-parametric and Bayesian approaches. This paper is motivated by a medical problem where interest focuses on developing a risk stratification system for morbidity of 1,710 cardiac patients given a suite of demographic, clinical and preoperative variables. Although the methods we use are applied specifically to this case study, these methods can be applied across any field, irrespective of the type of response.

A study of five cervicocephalic relocation tests in three different subject groups

Objective: To compare head relocation accuracy in traumatic ( whiplash), insidious onset neck pain patients and asymptomatic subjects when targeting a natural head posture (NHP) and complex predetermined positions. Design: A case-control study. Setting: University-based musculoskeletal research clinic. Participants: Sixty-three volunteers divided into three groups of similar gender and age: Group 1 (n=21) an asymptomatic group; group 2 (n=20) insidious onset neck pain; group 3 (n=22) a history of whiplash injury. Intervention: Five randomly ordered tests designed to detect relocation accuracy of the head. Outcome measures: A 3-Space Fastrak system measured the mean absolute relocation error of three trials of each relocation test. Results: A significant difference was found between groups in one of the tests targeting the NHP (p=0.001). Post-hoc pairwise comparisons revealed a significant difference (pless than or equal to0.05) between the asymptomatic group and each symptomatic group. The difference between the symptomatic groups just failed to reach significance (p=0.07). None of the other four tests revealed significant differences. Conclusion: The test of targeting the NHP indicates that relocation inaccuracy exists in patients with neck pain with a trend to suggest that the deficit may be greater in whiplash patients. Tests employing unfamiliar postures or more complex movement were not successful in differentiating subject groups.

Ras proteins: Different signals from different locations

Ras signalling has classically been thought to occur exclusively at the inner surface of a relatively uniform plasma membrane. Recent studies have shown that Ras proteins interact dynamically with specific microdomains of the plasma membrane as well as with other internal cell membranes. These different membrane microenvironments modulate Ras signal output and highlight the complex interplay between Ras location and function.

Mineral uptake in tobacco leaf discs during different developmental stages of shoot organogenesis

Relationships between mineral uptake and tobacco shoot organogenesis were investigated during three morphogenic phases: phase 1, days 0-10, pre-meristem formation; phase 2, days 10-20, meristem initiation and formation; and phase 3, days 20-35, growth and differentiation of induced meristems into leafy shoots. The mineral content of both shoot-forming (SF) and non-shoot-forming (NSF) media was examined over the 35-day culture period. Both SF and NSF explants rapidly consumed iron during phase 1. Nitrate uptake in SF explants was high and independent of explant growth during phases 1 and 2, but greatest and strongly correlated with growth during phase 3. Phosphorus uptake was highest in SF explants during phases 2 and 3, and correlated with explant growth. Uptake of potassium, calcium and sulphur was strongly associated with explant growth during phase 3 whereas magnesium uptake was only poorly correlated with growth. Results from this study indicate that particular minerals may have an important role in regulating development as well as generally supporting growth.

alpha-conotoxins PnIA and [A10L] PnIA stabilize different states of the alpha 7-L247T nicotinic acetylcholine receptor

The effects of the native alpha-conotoxin PnIA, its synthetic derivative [ A10L] PnIA and alanine scan derivatives of [ A10L] PnIA were investigated on chick wild type alpha7 and alpha7-L247T mutant nicotinic acetylcholine receptors (nAChRs) expressed in Xenopus oocytes. PnIA and [A10L] PnIA inhibited acetylcholine (ACh)-activated currents at wtalpha7 receptors with IC50 values of 349 and 168 nM, respectively. Rates of onset of inhibition were similar for PnIA and [ A10L] PnIA; however, the rate of recovery was slower for [ A10L] PnIA, indicating that the increased potency of [ A10L] PnIA at alpha7 receptors is conveyed by its slower rate of dissociation from the receptors. All the alanine mutants of [ A10L] PnIA inhibited ACh-activated currents at wtalpha7 receptors. Insertion of an alanine residue between position 5 and 13 and at position 15 significantly reduced the ability of [ A10L] PnIA to inhibit ACh-evoked currents. PnIA inhibited the non-desensitizing ACh-activated currents at alpha7-L247T receptors with an IC50 194 nM. In contrast, [ A10L] PnIA and the alanine mutants potentiated the ACh-activated current alpha7-L247T receptors and in addition [ A10L] PnIA acted as an agonist. PnIA stabilized the receptor in a state that is non-conducting in both the wild type and mutant receptors, whereas [ A10L] PnIA stabilized a state that is non-conducting in the wild type receptor and conducting in the alpha7-L247T mutant. These data indicate that the change of a single amino acid side-chain, at position 10, is sufficient to change the toxin specificity for receptor states in the alpha7-L247T mutant.

Resolution enhancement by subtraction of confocal signals taken at different pinhole sizes

Subtractive imaging in confocal fluorescence light microscopy is based on the subtraction of a suitably weighted widefield image from a confocal image. An approximation to a widefield image can be obtained by detection with an opened confocal pinhole. The subtraction of images enhances the resolution in-plane as well as along the optic axis. Due to the linearity of the approach, the effect of subtractive imaging in Fourier-space corresponds to a reduction of low spatial frequency contributions leading to a relative enhancement of the high frequencies. Along the direction of the optic axis this also results in an improved sectioning. Image processing can achieve a similar effect. However, a 3D volume dataset must be acquired and processed, yielding a result essentially identical to subtractive imaging but superior in signal-to-noise ratio. The latter can be increased further with the technique of weighted averaging in Fourier-space. A comparison of 2D and 3D experimental data analysed with subtractive imaging, the equivalent Fourier-space processing of the confocal data only, and Fourier-space weighted averaging is presented. (C) 2003 Elsevier Ltd. All rights reserved.