INTRAHOST EVOLUTION OF THE HIV-2 CAPSID CORRELATES WITH PROGRESSION TO AIDS

Background: HIV-2 infection will progress to AIDS in most patients without treatment, albeit at approximately half the rate of HIV-1 infection. HIV-2 p26 amino acid variations are associated with lower viral loads and enhanced processing of T cell epitopes, which may lead to protective Gag-specific CTL responses common in slower disease progressors. Lower virus evolutionary rates, and positive selection on conserved residues in HIV-2 env have been associated with slower progression to AIDS. We therefore aimed to determine if intrahost evolution of HIV-2 p26 is associated with disease progression. Methods: Twelve treatment-naive, HIV-2 mono-infected participants from the Guinea-Bissau Police cohort with longitudinal CD4+ T cell data and clinical follow-up were included in the analysis. CD4% change over time was analysed via linear regression models to stratify participants into relative faster and slower disease progressor groups. Gag amplicons of 735 nucleotides which spanned the p26 region were amplified by PCR and sequenced. We analysed p26 sequence diversity evolution, measured site-specific selection pressures and evolutionary rates, and determined if these evolutionary parameters were associated with progression status. Amino acid polymorphisms were mapped to existing p26 protein structures. Results: In total, 369 heterochronous HIV-2 p26 sequences from 12 male patients with a median age of 30 (IQR: 28-37) years at enrolment were analysed. Faster progressors had lower CD4% and faster CD4% decline rates. Median pairwise sequence diversity was higher in faster progressors (5.7x10-3 versus 1.4x10-3 base substitutions per site, P<0.001). p26 evolved under negative selection in both groups (dN/dS=0.12). Virus evolutionary rates were higher in faster than slower progressors - synonymous rates: 4.6x10-3 vs. 2.3x10-3; and nonsynonymous rates: 6.9x10-4 vs. 2.7x10-4 substitutions/site/year, respectively. Virus evolutionary rates correlated negatively with CD4% change rates (rho = -0.8, P=0.02), but not CD4% level. However, Bayes factor (BF) testing indicated that the association between evolutionary rates and CD4% kinetics was supported by weak evidence (BF=0.5). The signature amino acid at p26 positions 6, 12 and 119 differed between faster (6A, 12I, 119A) and slower (6G, 12V, 119P) progressors. These amino acid positions clustered near to the TRIM5 alpha/p26 hexamer interface surface. Conclusions: Faster p26 evolutionary rates were associated with faster progression to AIDS and were mostly driven by synonymous substitutions. Nonsynonymous evolutionary rates were an order of magnitude lower than synonymous rates, with limited amino acid sequence evolution over time within hosts. These results indicate the HIV-2 p26 may be an attractive vaccine or therapeutic target.

A New Bayesian Two-Sample t Test and Solution to the Behrens–Fisher Problem Based on Gaussian Mixture Modelling with Known Allocations

Testing differences between a treatment and control group is common practice in biomedical research like randomized controlled trials (RCT). The standard two-sample t test relies on null hypothesis significance testing (NHST) via p values, which has several drawbacks. Bayesian alternatives were recently introduced using the Bayes factor, which has its own limitations. This paper introduces an alternative to current Bayesian two-sample t tests by interpreting the underlying model as a two-component Gaussian mixture in which the effect size is the quantity of interest, which is most relevant in clinical research. Unlike p values or the Bayes factor, the proposed method focusses on estimation under uncertainty instead of explicit hypothesis testing. Therefore, via a Gibbs sampler, the posterior of the effect size is produced, which is used subsequently for either estimation under uncertainty or explicit hypothesis testing based on the region of practical equivalence (ROPE). An illustrative example, theoretical results and a simulation study show the usefulness of the proposed method, and the test is made available in the R package . In sum, the new Bayesian two-sample t test provides a solution to the Behrens–Fisher problem based on Gaussian mixture modelling.

Vaccine hesitancy for COVID-19 explored in a phenomic study of 259 socio-cognitive-behavioural measures in the UK-REACH study of 12,431 UK healthcare workers

Background Vaccination is key to successful prevention of COVID-19 particularly nosocomial acquired infection in health care workers (HCWs). 'Vaccine hesitancy' is common in the population and in HCWs, and like COVID-19 itself, hesitancy is more frequent in ethnic minority groups. UK-REACH (United Kingdom Research study into Ethnicity and COVID-19 outcomes) is a large-scale study of COVID-19 in UK HCWs from diverse ethnic backgrounds, which includes measures of vaccine hesitancy. The present study explores predictors of vaccine hesitancy using a 'phenomic approach', considering several hundred questionnaire-based measures. Methods UK-REACH includes a questionnaire study encompassing 12,431 HCWs who were recruited from December 2020 to March 2021 and completed a lengthy online questionnaire (785 raw items; 392 derived measures; 260 final measures). Ethnicity was classified using the Office for National Statistics' five (ONS5) and eighteen (ONS18) categories. Missing data were handled by multiple imputation. Variable selection used the islasso package in R, which provides standard errors so that results from imputations could be combined using Rubin's rules. The data were modelled using path analysis, so that predictors, and predictors of predictors could be assessed. Significance testing used the Bayesian approach of Kass and Raftery, a 'very strong' Bayes Factor of 150, N=12,431, and a Bonferroni correction giving a criterion of p<4.02 x 10-8 for the main regression, and p<3.11 x 10-10 for variables in the path analysis. Results At the first step of the phenomic analysis, six variables were direct predictors of greater vaccine hesitancy: Lower pro-vaccination attitudes; no flu vaccination in 2019-20; pregnancy; higher COVID-19 conspiracy beliefs; younger age; and lower optimism the roll-out of population vaccination. Overall 44 lower variables in total were direct or indirect predictors of hesitancy, with the remaining 215 variables in the phenomic analysis not independently predicting vaccine hesitancy. Key variables for predicting hesitancy were belief in conspiracy theories of COVID-19 infection, and a low belief in vaccines in general. Conspiracy beliefs had two main sets of influences: i) Higher Fatalism, which was influenced a) by high external and chance locus of control and higher need for closure, which in turn were associated with neuroticism, conscientiousness, extraversion and agreeableness; and b) by religion being important in everyday life, and being Muslim. ii) receiving information via social media, not having higher education, and perceiving greater risks to self, the latter being influenced by higher concerns about spreading COVID, greater exposure to COVID-19, and financial concerns. There were indirect effects of ethnicity, mediated by religion. Religion was more important for Pakistani and African HCWs, and less important for White and Chinese groups. Lower age had a direct effect on hesitancy, and age and female sex also had several indirect effects on hesitancy. Conclusions The phenomic approach, coupled with a path analysis revealed a complex network of social, cognitive, and behavioural influences on SARS-Cov-2 vaccine hesitancy from 44 measures, 6 direct and 38 indirect, with the remaining 215 measures not having direct or indirect effects on hesitancy. It is likely that issues of trust underpin many associations with hesitancy. Understanding such a network of influences may help in tailoring interventions to address vaccine concerns and facilitate uptake in more hesistant groups.

Computing Bayes factors from data with missing values

The Bayes factor is increasingly used for the evaluation of hypotheses. These may betraditional hypotheses specified using equality constraints among the parameters of thestatistical model of interest or informative hypotheses specified using equality andinequality constraints. So far no attention has been given to the computation of Bayesfactors from data with missing values. A key property of such a Bayes factor should bethat it is only based on the information in the observed values. This paper will show thatsuch a Bayes factor can be obtained using multiple imputations of the missing values.

Putting the individual into reliability: Bayesian testing of homogeneous within-person variance in hierarchical models

Measurement reliability is a fundamental concept in psychology. It is traditionally considered a stable property of a questionnaire, measurement device, or experimental task. Although intraclass correlation coefficients (ICC) are often used to assess reliability in repeated measure designs, their descriptive nature depends upon the assumption of a common within-person variance. This work focuses on the presumption that each individual is adequately described by the average within-person variance in hierarchical models. And thus whether reliability generalizes to the individual level, which leads directly into the notion of individually varying ICCs. In particular, we introduce a novel approach, using the Bayes factor, wherein a researcher can directly test for homogeneous within-person variance in hierarchical models. Additionally, we introduce a membership model that allows for classifying which (and how many) individuals belong to the common variance model. The utility of our methodology is demonstrated on cognitive inhibition tasks. We find that heterogeneous within-person variance is a defining feature of these tasks, and in one case, the ratio between the largest to smallest within-person variance exceeded 20. This translates into a tenfold difference in person-specific reliability! We also find that few individuals belong to the common variance model, and thus traditional reliability indices are potentially masking important individual variation. We discuss the implications of our findings and possible future directions. The methods are implemented in the R package vICC

Get to grips with motivation: slipping and gripping movements are biased by approach-avoidance context

People are better at approaching appetitive cues signalling reward and avoiding aversive cues signalling punishment than vice versa. This action bias has previously been shown in approach-avoidance tasks involving arm movements in response to appetitive or aversive cues. It is not known whether appetitive or aversive stimuli also bias more distal dexterous actions, such as gripping and slipping, in a similar manner. To test this hypothesis, we designed a novel task involving grip force control (gripping and slipping) to probe gripping-related approach and avoidance behaviour. 32 male volunteers, aged 18-40 years, were instructed to either grip (“approach”) or slip (”avoid”) a grip-force device with their right thumb and index finger at the sight of positive or negative images. In one version of this pincer grip task, participants were responding to graspable objects and in another version of the task they were responding to happy or angry faces. Bayesian repeated measures Analysis of variance revealed extreme evidence for an interaction between response type and cue valence (Bayes factor = 296). Participants were faster to respond in affect-congruent conditions (“approach appetitive”, “avoid aversive”) than in affect-incongruent conditions (“approach aversive”, “avoid appetitive”). This bias towards faster response times for affect-congruent conditions was present regardless of whether it was a graspable object or a face signalling valence. Since our results mirror the approach and avoidance effects previously observed for arm movements, we conclude that a tendency favouring affectively congruent cue-response mappings is an inherent feature of motor control and thus also includes precision grip.

Principles of Model Specification in ANOVA Designs

ANOVA---the workhorse of experimental psychology--seems well understood in that behavioral sciences have agreed-upon contrasts and reporting conventions. Yet, we argue this consensus hides considerable flaws in common ANOVA procedures, and these flaws become especially salient in the within-subject and mixed-model cases. The main thesis is that these flaws are in model specification. The specifications underlying common use are deficient from a substantive perspective, that is, they do not match reality in behavioral experiments. The problem, in particular, is that specifications rely on coincidental rather than robust statements about reality. We provide specifications that avoid making arguments based on coincidences, and note these Bayes factor model comparisons among these specifications are already convenient in the BayesFactor package. Finally, we argue that model specification necessarily and critically reflects substantive concerns, and, consequently, is ultimately the responsibility of substantive researchers. Source code for this project is at github/PerceptionAndCognitionLab/stat_aov2