Compositional data in neuroscience: If you’ve got it, log it!

Modeling the neurocognitive dynamics of language across the lifespan

Healthy aging is associated with a heterogeneous decline across cognitive functions, typically observed between language comprehension and language production (LP). Examining resting-state fMRI and neuropsychological data from 628 healthy adults (age 18-88) from the CamCAN cohort, we performed state-of-the-art graph theoretical analysis to uncover the neural mechanisms underlying this variability. At the cognitive level, our findings suggest that LP is not an isolated function but is modulated throughout the lifespan by the extent of inter-cognitive synergy between semantic and domain-general processes. At the cerebral level, we show that default mode network (DMN) suppression coupled with fronto-parietal network (FPN) integration is the way for the brain to compensate for the effects of dedifferentiation at a minimal cost, efficiently mitigating the age-related decline in LP. Relatedly, reduced DMN suppression in midlife could compromise the ability to manage the cost of FPN integration. This may prompt older adults to adopt a more cost-efficient compensatory strategy that maintains global homeostasis at the expense of LP performances. Taken together, we propose that midlife represents a critical neurocognitive juncture that signifies the onset of LP decline, as older adults gradually lose control over semantic representations. We summarize our findings in a novel synergistic, economical, nonlinear, emergent, cognitive aging model, integrating connectomic and cognitive dimensions within a complex system perspective.

Statistical power and false positive rates for interdependent outcomes are strongly influenced by test type: Implications for behavioral neuroscience

Statistical errors in preclinical science are a barrier to reproducibility and translation. For instance, linear models (e.g., ANOVA, linear regression) may be misapplied to data that violate assumptions. In behavioral neuroscience and psychopharmacology, linear models are frequently applied to interdependent or compositional data, which includes behavioral assessments where animals concurrently choose between chambers, objects, outcomes, or types of behavior (e.g., forced swim, novel object, place/social preference). The current study simulated behavioral data for a task with four interdependent choices (i.e., increased choice of a given outcome decreases others) using Monte Carlo methods. 16,000 datasets were simulated (1000 each of 4 effect sizes by 4 sample sizes) and statistical approaches evaluated for accuracy. Linear regression and linear mixed effects regression (LMER) with a single random intercept resulted in high false positives (>60%). Elevated false positives were attenuated in an LMER with random effects for all choice-levels and a binomial logistic mixed effects regression. However, these models were underpowered to reliably detect effects at common preclinical sample sizes. A Bayesian method using prior knowledge for control subjects increased power by up to 30%. These results were confirmed in a second simulation (8000 datasets). These data suggest that statistical analyses may often be misapplied in preclinical paradigms, with common linear methods increasing false positives, but potential alternatives lacking power. Ultimately, using informed priors may balance statistical requirements with ethical imperatives to minimize the number of animals used. These findings highlight the importance of considering statistical assumptions and limitations when designing research studies.

Clustering compositional data using Dirichlet mixture model

A model-based clustering method for compositional data is explored in this article. Most methods for compositional data analysis require some kind of transformation. The proposed method builds a mixture model using Dirichlet distribution which works with the unit sum constraint. The mixture model uses a hard EM algorithm with some modification to overcome the problem of fast convergence with empty clusters. This work includes a rigorous simulation study to evaluate the performance of the proposed method over varied dimensions, number of clusters, and overlap. The performance of the model is also compared with other popular clustering algorithms often used for compositional data analysis (e.g. KMeans, Gaussian mixture model (GMM) Gaussian Mixture Model with Hard EM (Hard GMM), partition around medoids (PAM), Clustering Large Applications based on Randomized Search (CLARANS), Density-Based Spatial Clustering of Applications with Noise (DBSCAN) etc.) for simulated data as well as two real data problems coming from the business and marketing domain and physical science domain, respectively. The study has shown promising results exploiting different distributional patterns of compositional data.

Composition of time in movement behaviors and weight change in Latinx, Black and white participants

Background

The relationship between time-use behaviors and prospective weight change is poorly understood.

Methods

A subset of Cancer Prevention Study-3 participants (n = 549, 58% women, 66% non-Latinx white) self-reported weight in 2015 and 2018 and completed an accelerometer protocol for seven days. Sedentary time, sleep, light, moderate, and vigorous intensity physical activity (PA) were treated as a compositional variable and multiple linear regression was used to examine associations between activity composition and weight change stratified by sex and race/ethnicity. Compositional isotemporal substitution analysis was used to quantify change in weight associated with reallocating 30 min./day.

Results

Activity composition was associated with weight change among women (p = 0.007), but not men (p = 0.356), and among Latinx (p = 0.032) and white participants (p = 0.001), but not Black participants (p = 0.903). Replacement of 30 min./day sedentary time with moderate-vigorous PA was associated with 3.49 lbs. loss (-6.76, -0.22) in Latinx participants and replacement with sleep was associated with 1.52 (0.25, 2.79) and 1.31 (0.40, 2.21) lbs. gain in white women and men.

Conclusion

The distribution of time spent in daily behaviors was associated with three-year weight change in women, Latinx, and white participants. This was the first longitudinal compositional study of weight change; thus, more studies are needed.

The µ-opioid system in midline thalamic nuclei modulates defence strategies towards a conditioned fear stimulus in male mice

BACKGROUND

Nuclei located in the dorsal midline thalamus, such as the paraventricular nucleus of the thalamus (PVT), are crucial to modulate fear and aversive behaviour. In addition, the PVT shows a dense expression of µ-opioid receptors (MORs) and could mediate the anxiolytic effects of opioids.

METHODS

We analysed the contribution of MORs in the dorsal midline thalamus (i.e. the PVT) to the performance of mice in a classical fear conditioning paradigm. We locally injected a specific agonist (DAMGO), an antagonist (CTAP) of MOR or saline as a control into the dorsal midline thalamus of male mice, prior to fear extinction training. We assessed freezing as a typical measure of fear and extended our analysis by evaluation of aversive, non-aversive and neutral behavioural features using compositional data analysis.

RESULTS

Pharmacological blockade of MORs through CTAP in the dorsal midline thalamus induced a fear memory extinction deficit, as evidenced by maintained freezing during extinction sessions. Stimulation of MORs by DAMGO resulted in an overall increase in locomotor activity, associated with decreased freezing during recall of extinction. Compositional data analysis confirmed the freezing-related pharmacological effects and revealed specific differences in basic behavioural states. CTAP-treated mice remained in an aversive state, whereas DAMGO-treated mice displayed predominantly neutral behaviour.

CONCLUSIONS

Fear extinction requires the integrity of the µ-opioid system in the dorsal midline thalamus. Pharmacological stimulation of MOR and associated facilitation of fear extinction recall suggest a potential therapeutic avenue for stress-related or anxiety disorders.

Host sex and transplanted human induced pluripotent stem cell phenotype interact to influence sensorimotor recovery in a mouse model of cortical contusion injury

Traumatic brain injury (TBI) is a substantial cause of disability and death worldwide. Primary head trauma triggers chronic secondary injury mechanisms in the brain that are a focus of therapeutic efforts to treat TBI. Currently, there is no successful clinical strategy to repair brain injury. Cell transplantation therapies have demonstrated promise in attenuating secondary injury mechanisms of neuronal death and dysfunction in animal models of brain injury. In this study, we used a unilateral cortical contusion injury (CCI) model of sensorimotor brain injury to examine the effects of human induced pluripotent stem cell (hiPSC) transplantation on pathology in male and female adult mice. We determined transplanted hiPSC-derived neural stem cells (NSCs) and neuroblasts but not astrocytes best tolerate the injured host environment. Surviving NSC and neuroblast cells were clustered at the site of injection within the deep layers of the cortex and underlying corpus callosum. Cell grafts extended neuritic processes that crossed the midline into the contralateral corpus callosum or continued laterally within the external capsule to enter the ipsilateral entorhinal cortex. To determine the effect of transplantation on neuropathology, we performed sensorimotor behavior testing and stereological estimation of host neurons, astrocytes, and microglia within the contused cortex. These measures did not reveal a consistent effect of transplantation on recovery post-injury. Rather the positive and negative effects of cell transplantation were dependent on the host sex, highlighting the importance of developing patient-specific approaches to treat TBI. Our study underscores the complex interactions of sex, neuroimmune responses and cell therapy in a common experimental model of TBI.

A Lachnospiraceae-dominated bacterial signature in the fecal microbiota of HIV-infected individuals from Colombia, South America

HIV infection has a tremendous impact on the immune system's proper functioning. The mucosa-associated lymphoid tissue (MALT) is significantly disarrayed during HIV infection. Compositional changes in the gut microbiota might contribute to the mucosal barrier disruption, and consequently to microbial translocation. We performed an observational, cross-sectional study aimed at evaluating changes in the fecal microbiota of HIV-infected individuals from Colombia. We analyzed the fecal microbiota of 37 individuals via 16S rRNA gene sequencing; 25 HIV-infected patients and 12 control (non-infected) individuals, which were similar in body mass index, age, gender balance and socioeconomic status. To the best of our knowledge, no such studies have been conducted in Latin American countries. Given its compositional nature, microbiota data were normalized and transformed using Aitchison's Centered Log-Ratio. Overall, a change in the network structure in HIV-infected patients was revealed by using the SPIEC-EASI MB tool. Genera such as Blautia, Dorea, Yersinia, Escherichia-Shigella complex, Staphylococcus, and Bacteroides were highly relevant in HIV-infected individuals. Differential abundance analysis by both sparse Partial Least Square-Discriminant Analysis and Random Forest identified a greater abundance of Lachnospiraceae-OTU69, Blautia, Dorea, Roseburia, and Erysipelotrichaceae in HIV-infected individuals. We show here, for the first time, a predominantly Lachnospiraceae-based signature in HIV-infected individuals.