Sally Galbraith

A Study of Clustered Data and Approaches to Its Analysis

Statistical analysis is critical in the interpretation of experimental data across the life sciences, including neuroscience. The nature of the data collected has a critical role in determining the best statistical approach to take. One particularly prevalent type of data is referred to as “clustered data.” Clustered data are characterized as data that can be classified into a number of distinct groups or “clusters” within a particular study. Clustered data arise most commonly in neuroscience when data are compiled across multiple experiments, for example in electrophysiological or optical recordings taken from synaptic terminals, with each experiment providing a distinct cluster of data. However, there are many other types of experimental design that can yield clustered data. Here, we provide a statistical model for intracluster correlation and systematically investigate a range of methods for analyzing clustered data. Our analysis reveals that it is critical to take data clustering into account and suggests appropriate statistical approaches that can be used to account for data clustering.

Activin A is essential for neurogenesis following neurodegeneration.

It has long been proposed that excitotoxicity contributes to nerve cell death in neurodegenerative diseases. Activin A, a member of the transforming growth factor‐β superfamily, is expressed by neurons following excitotoxicity. We show for the first time that this activin A expression is essential for neurogenesis to proceed following neurodegeneration. We found that intraventricular infusion of activin A increased the number of newborn neurons in the dentate gyrus, CA3, and CA1 layers of the normal adult hippocampus and also, following lipopolysaccharide administration, had a potent inhibitory effect on gliosis in vivo and on microglial proliferation in vivo and in vitro. Consistent with the role of activin A in regulating central nervous system inflammation and neurogenesis, intraventricular infusion of follistatin, an activin A antagonist, profoundly impaired neurogenesis and increased the number of microglia and reactive astrocytes following onset of kainic acid‐induced neurodegeneration. These results show that inhibiting endogenous activin A is permissive for a potent underlying inflammatory response to neurodegeneration. We demonstrate that the anti‐inflammatory actions of activin A account for its neurogenic effects following neurodegeneration because co‐administration of nonsteroidal anti‐inflammatory drugs reversed follistatins inhibitory effects on neurogenesis in vivo. Our work indicates that activin A, perhaps working in conjunction with other transforming growth factor‐β superfamily molecules, is essential for neurogenesis in the adult central nervous system following excitotoxic neurodegeneration and suggests that neurons can regulate regeneration by suppressing the inflammatory response, a finding with implications for understanding and treating acute and chronic neurodegenerative diseases. STEM CELLS 2009;27:1330–1346

Accelerated longitudinal designs: An overview of modelling, power, costs and handling missing data

Longitudinal studies are often used to investigate age-related developmental change. Whereas a single cohort design takes a group of individuals at the same initial age and follows them over time, an accelerated longitudinal design takes multiple single cohorts, each one starting at a different age. The main advantage of an accelerated longitudinal design is its ability to span the age range of interest in a shorter period of time than would be possible with a single cohort longitudinal design. This paper considers design issues for accelerated longitudinal studies. A linear mixed effect model is considered to describe the responses over age with random effects for intercept and slope parameters. Random and fixed cohort effects are used to cope with the potential bias accelerated longitudinal designs have due to multiple cohorts. The impact of other factors such as costs and the impact of dropouts on the power of testing or the precision of estimating parameters are examined. As duration-related costs increase relative to recruitment costs the best designs shift towards shorter duration and eventually cross-sectional design being best. For designs with the same duration but differing interval between measurements, we found there was a cutoff point for measurement costs relative to recruitment costs relating to frequency of measurements. Under our model of 30% dropout there was a maximum power loss of 7%.

Peripheral Blood Gene Expression in Postinfective Fatigue Syndrome Following From Three Different Triggering Infections

BACKGROUND Several infections trigger postinfective fatigue syndromes, which share key illness characteristics with each other and with chronic fatigue syndrome (CFS). Previous cross-sectional case-control studies of CFS have suggested that unique gene expression signatures are evident in peripheral blood samples. METHODS Peripheral blood transcriptomes in samples collected longitudinally, in 18 subjects with a fatigue syndrome lasting ≥ 6 months after acute infection due to Epstein-Barr virus, Ross River virus, or Coxiella burnetii (Q fever), and 18 matched control subjects who had recovered promptly, were studied by microarray (n = 127) and confirmatory quantitative polymerase chain reaction (PCR). Gene expression patterns associated with CFS were sought by univariate statistics and regression modeling. RESULTS There were 23 genes with modest differential expression (0.6-2.3-fold change) in within-subject comparisons of early, symptomatic time points with late, recovered time points. There were modest differences found in 63 genes, either in cross-sectional comparison of cases and controls at 6 months after infection onset or in the regression model. There were 223 genes significantly correlated with individual symptom domains. Quantitative PCR confirmed 33 (73%) of 45 genes-none were consistent across cohorts. CONCLUSIONS Although the illness characteristics of patients with postinfective fatigue syndromes have more similarities than differences, no reliable peripheral blood gene expression correlate is evident.

Functional heterogeneity at dopamine release sites

Although drugs used to treat several neurological diseases are presumed to target synapses that secrete dopamine (DA), relatively little is known about synaptic vesicle (SV) release mechanisms at single DA synapses. We found that the relative probability of release (Pr) varied between individual DA synapses. Furthermore, DA terminals generally exhibited lower Pr than glutamatergic hippocampal (Hpc) terminals, suggesting that DA release is less reliable than the release of glutamate. Our mathematical model of fluorescence loss shows that Pr is regulated by two independent and heterogeneous elements. First, the size of the recycling SV pool regulates Pr. Second, Pr is also independently regulated by additional factors, which are reflected in the time constant of FM 1-43 destaining, τ. We found that the observed difference in Pr between Hpc and DA neurons results because the recycling SV pool is smaller in DA neurons than in Hpc neurons. However, τ does not vary between these two neuron populations. We also identified a population of functional nonsynaptic boutons in DA axons, which are not associated with a postsynaptic element and which are not functionally different from boutons that formed conventional synapses. Our work provides a new approach to the study of SV exocytosis in DA neurons and shows that synaptic terminals of DA neurons are functionally heterogeneous and differ from excitatory terminals in terms of Pr.

Correlates and characteristics of hepatitis C virus-specific T-cell immunity in exposed uninfected high-risk prison inmates.

Some hepatitis C (HCV)‐uninfected, high‐risk individuals have HCV‐specific cellular immunity without viraemia or seroconversion. The characteristics of these responses and the risk behavioural associations were studied in 94 subjects in a prospective cohort of recently seronegative prisoners reporting injecting drug use (IDU). Detailed behavioural data were collected. HCV antibody and PCR testing were performed. ELISpot assays for HCV‐induced interferon (IFN)‐γ and interleukin (IL)‐2 production by T lymphocytes, as well as multiplex in vitro cytokine production assays, were performed. Seventy‐eight subjects remained antibody and PCR negative and 16 seroconverted. Of the seronegative group, 22 (28%) had IFN‐γ ELISpot responses in comparison with 13 of the 16 seroconverters (82%). This seronegative immune status was associated positively with injecting anabolic steroids and negatively with a recent break from IDU. The IFN‐γ ELISpot responses involved both CD4 and CD8 T lymphocytes and were comparable in magnitude, but narrower in specificity, in uninfected subjects than in seroconverters. A subset of seronegative subjects had HCV‐induced cytokine production patterns comparable with the seroconverters with increased production of IFN‐γ, IL‐2 and tumour necrosis factor (TNF)‐α and reduced IL‐10 in response to nonstructural peptides. In conclusion, comparable patterns of HCV‐specific cellular immunity are found in recently infected subjects and in a minority of high‐risk, uninfected subjects. Further characterization of these responses and their protective efficacy will inform HCV vaccine development.

Gene Expression Correlates of Postinfective Fatigue Syndrome after Infectious Mononucleosis

BACKGROUND Infectious mononucleosis (IM) commonly triggers a protracted postinfective fatigue syndrome (PIFS) of unknown pathogenesis. METHODS Seven subjects with PIFS with 6 or more months of disabling symptoms and 8 matched control subjects who had recovered promptly from documented IM were studied. The expression of 30,000 genes was examined in the peripheral blood by microarray analysis in 65 longitudinally collected samples. Gene expression patterns associated with PIFS were sought by correlation with symptom factor scores. RESULTS Differential expression of 733 genes was identified when samples collected early during the illness and at the late (recovered) time point were compared. Of these genes, 234 were found to be significantly correlated with the reported severity of the fatigue symptom factor, and 180 were found to be correlated with the musculoskeletal pain symptom factor. Validation by analysis of the longitudinal expression pattern revealed 35 genes for which changes in expression were consistent with the illness course. These genes included several that are involved in signal transduction pathways, metal ion binding, and ion channel activity. CONCLUSIONS Gene expression correlates of the cardinal symptoms of PIFS after IM have been identified. Further studies of these gene products may help to elucidate the pathogenesis of PIFS.

LASSO model selection with post-processing for a genome-wide association study data set

Model selection procedures for simultaneous analysis of all single-nucleotide polymorphisms in genome-wide association studies are most suitable for making full use of the data for a complex disease study. In this paper we consider a penalized regression using the LASSO procedure and show that post-processing of the penalized-regression results with subsequent stepwise selection may lead to improved identification of causal single-nucleotide polymorphisms.

Dietary changes in migrant adolescents with increasing length of stay in Australia and associated risk of wheeze – a retrospective, cross sectional study

BackgroundRecent studies have reported that asthma prevalence increases on migration to Australia. We hypothesised that changes in dietary intake contribute to this phenomenon. The aim of this study was to assess dietary intake in relation to migration status, length of stay in Australia and the association with self-reported wheeze.MethodsStudents (n = 144) in a multicultural high school in Western Sydney completed the asthma symptoms ISAAC video questionnaire (AVQ3.0), spirometry and allergy skin prick tests. A dietitian administered a’Food Frequency’ and ‘Food Habits’ questionnaire and a dietary history interview.ResultsStudents who spoke a language other than English, consumed a traditional or mixed dietary pattern, with lower consumption of saturated fat, compared to students who spoke English only. Saturated fat intake increased and fibre intake decreased with length of time in Australia. Intake of foods high in saturated or trans fatty acids were positively associated with length of stay in Australia. No associations between nutrient intake or whole food intake and self-reported wheeze were observed.ConclusionAs time progressed, dietary intake of immigrant children changed. While this was not associated with the development of wheeze in the students in this cohort, these changes are likely to have negative health consequences.