A classification-based generative approach to selective targeting of global slow oscillations during sleep

Background

Given sleep's crucial role in health and cognition, numerous sleep-based brain interventions are being developed, aiming to enhance cognitive function, particularly memory consolidation, by improving sleep. Research has shown that Transcranial Alternating Current Stimulation (tACS) during sleep can enhance memory performance, especially when used in a closed-loop (cl-tACS) mode that coordinates with sleep slow oscillations (SOs, 0.5-1.5Hz). However, sleep tACS research is characterized by mixed results across individuals, which are often attributed to individual variability.

Objective/Hypothesis

This study targets a specific type of SOs, widespread on the electrode manifold in a short delay ("global SOs"), due to their close relationship with long-term memory consolidation. We propose a model-based approach to optimize cl-tACS paradigms, targeting global SOs not only by considering their temporal properties but also their spatial profile.

Methods

We introduce selective targeting of global SOs using a classification-based approach. We first estimate the current elicited by various stimulation paradigms, and optimize parameters to match currents found in natural sleep during a global SO. Then, we employ an ensemble classifier trained on sleep data to identify effective paradigms. Finally, the best stimulation protocol is determined based on classification performance.

Results

Our study introduces a model-driven cl-tACS approach that specifically targets global SOs, with the potential to extend to other brain dynamics. This method establishes a connection between brain dynamics and stimulation optimization.

Conclusion

Our research presents a novel approach to optimize cl-tACS during sleep, with a focus on targeting global SOs. This approach holds promise for improving cl-tACS not only for global SOs but also for other physiological events, benefiting both research and clinical applications in sleep and cognition.

The effect of ascertainment on penetrance estimates for rare variants: Implications for establishing pathogenicity and for genetic counselling

Next-generation sequencing has led to an explosion of genetic findings for many rare diseases. However, most of the variants identified are very rare and were also identified in small pedigrees, which creates challenges in terms of penetrance estimation and translation into genetic counselling in the setting of cascade testing. We use simulations to show that for a rare (dominant) disorder where a variant is identified in a small number of small pedigrees, the penetrance estimate can both have large uncertainty and be drastically inflated, due to underlying ascertainment bias. We have developed PenEst, an app that allows users to investigate the phenomenon across ranges of parameter settings. We also illustrate robust ascertainment corrections via the LOD (logarithm of the odds) score, and recommend a LOD-based approach to assessing pathogenicity of rare variants in the presence of reduced penetrance.

A genome-wide association analysis of loss of ambulation in dystrophinopathy patients suggests multiple candidate modifiers of disease severity

The major determinant of disease severity in Duchenne muscular dystrophy (DMD) or milder Becker muscular dystrophy (BMD) is whether the dystrophin gene (DMD) mutation truncates the mRNA reading frame or allows expression of a partially functional protein. However, even in the complete absence of dystrophin, variability in disease severity is observed, and candidate gene studies have implicated several genes as modifiers. Here we present the largest genome-wide search to date for loci influencing severity in N = 419 DMD patients. Availability of subjects for such studies is quite limited, leading to modest sample sizes, which present a challenge for GWAS design. We have therefore taken special steps to minimize heterogeneity within our dataset at the DMD locus itself, taking a novel approach to mutation classification to effectively exclude the possibility of residual dystrophin expression, and utilized statistical methods that are well adapted to smaller sample sizes, including the use of a novel linear regression-like residual for time to ambulatory loss and the application of evidential statistics for the GWAS approach. Finally, we applied an unbiased in silico pipeline, utilizing functional genomic datasets to explore the potential impact of the best supported SNPs. In all, we obtained eight SNPs (out of 1,385,356 total) with posterior probability of trait-marker association (PPLD) ≥ 0.4, representing six distinct loci. Our analysis prioritized likely non-coding SNP regulatory effects on six genes (ETAA1, PARD6G, GALNTL6, MAN1A1, ADAMTS19, and NCALD), each with plausibility as a DMD modifier. These results support both recurrent and potentially new pathways for intervention in the dystrophinopathies.

0112 Classification of reconstructed depth profiles shows Global and non-Global slow oscillations differentiate in the hippocampus and thalamus

Introduction

Sleep slow oscillations (SOs, 0.5-1.5 Hz) can be classified on the scalp as Global, Local or Frontal, where Global SOs are found in most electrodes within a short time delay and gate long-range information flow during NREM sleep. In this study, we estimate the current density within the brain that generates a Global SO, to evaluate which sub-cortical structures are involved in Global SO dynamics. We then train multiple machine learning algorithms to distinguish between Global SOs and other SO types, and probe variance of Global/non-Global SO profiles within and across subjects.Sleep slow oscillations (SOs, 0.5-1.5 Hz) can be classified on the scalp as Global, Local or Frontal, where Global SOs are found in most electrodes within a short time delay and gate long-range information flow during NREM sleep. In this study, we estimate the current density within the brain that generates a Global SO, to evaluate which sub-cortical structures are involved in Global SO dynamics. We then train multiple machine learning algorithms to distinguish between Global SOs and other SO types, and probe variance of Global/non-Global SO profiles within and across subjects.

Methods

32 volunteers (18 females) slept in the lab with polysomnography including 24 head EEG channels; their sleep was scored according to AASM criteria. SOs were algorithmically detected at each channel and classified as Global or non-Global using our method (Malerba et al., 2019). The depth profile of each SO was reconstructed with current source estimation (in Brainstorm followed by sLORETA), with a standardized head model including 17 regions. Each depth profile was embedded in a matrix averaging current by region and in three 200ms-long time bins: before, during and after the SO trough. Thirty classifiers were applied to this dataset, leveraging Matlab’s supervised learning application. We compared accuracy within and across subjects and identified best-performing algorithms across dataset size. We then used univariate feature selection to quantify the relevance of each region-time pair to successful classification.

Results

Global/non-Global SOs current depth profiles have higher variance across subjects, and accuracy improves when data is sampled between rather than within individuals. Ensemble subspace methods reached highest accuracy (98.5%). Feature selectivity identified cortical, hippocampal, and thalamic currents at the trough of the SO as the most relevant for Global/non-Global SO classifications.

Conclusion

We introduce an analytical framework enabling the study of SO depth profiles, including their time evolution, as matrices. The predominant differentiation of Global/non-Global SOs in cortical, hippocampal, and thalamic currents supports the potential functional difference of these SO types.

Support (If Any)

NIH grant (R01 AG046646) to S.C.M.

Global and non-Global slow oscillations differentiate in their depth profiles

Sleep slow oscillations (SOs, 0.5-1.5 Hz) are thought to organize activity across cortical and subcortical structures, leading to selective synaptic changes that mediate consolidation of recent memories. Currently, the specific mechanism that allows for this selectively coherent activation across brain regions is not understood. Our previous research has shown that SOs can be classified on the scalp as Global, Local or Frontal, where Global SOs are found in most electrodes within a short time delay and gate long-range information flow during NREM sleep. The functional significance of space-time profiles of SOs hinges on testing if these differential SOs scalp profiles are mirrored by differential depth structure of SOs in the brain. In this study, we built an analytical framework to allow for the characterization of SO depth profiles in space-time across cortical and sub-cortical regions. To test if the two SO types could be differentiated in their cortical-subcortical activity, we trained 30 machine learning classification algorithms to distinguish Global and non-Global SOs within each individual, and repeated this analysis for light (Stage 2, S2) and deep (slow wave sleep, SWS) NREM stages separately. Multiple algorithms reached high performance across all participants, in particular algorithms based on k-nearest neighbors classification principles. Univariate feature ranking and selection showed that the most differentiating features for Global vs. non-Global SOs appeared around the trough of the SO, and in regions including cortex, thalamus, caudate nucleus, and brainstem. Results also indicated that differentiation during S2 required an extended network of current from cortical-subcortical regions, including all regions found in SWS and other basal ganglia regions, and amygdala and hippocampus, suggesting a potential functional differentiation in the role of Global SOs in S2 vs. SWS. We interpret our results as supporting the potential functional difference of Global and non-Global SOs in sleep dynamics.

The PPLD has advantages over conventional regression methods in application to moderately sized genome-wide association studies

In earlier work, we have developed and evaluated an alternative approach to the analysis of GWAS data, based on a statistic called the PPLD. More recently, motivated by a GWAS for genetic modifiers of the X-linked Mendelian disorder Duchenne Muscular Dystrophy (DMD), we adapted the PPLD for application to time-to-event (TE) phenotypes. Because DMD itself is relatively rare, this is a setting in which the very large sample sizes generally assembled for GWAS are simply not attainable. For this reason, statistical methods specially adapted for use in small data sets are required. Here we explore the behavior of the TE-PPLD via simulations, comparing the TE-PPLD with Cox Proportional Hazards analysis in the context of small to moderate sample sizes. Our results will help to inform our approach to the DMD study going forward, and they illustrate several respects in which the TE-PPLD, and by extension the original PPLD, offer advantages over regression-based approaches to GWAS in this context.

A new linear regression-like residual for survival analysis, with application to genome wide association studies of time-to-event data

In linear regression, a residual measures how far a subject's observation is from expectation; in survival analysis, a subject's Martingale or deviance residual is sometimes interpreted similarly. Here we consider ways in which a linear regression-like interpretation is not appropriate for Martingale and deviance residuals, and we develop a novel time-to-event residual which does have a linear regression-like interpretation. We illustrate the utility of this new residual via simulation of a time-to-event genome-wide association study, motivated by a real study seeking genetic modifiers of Duchenne Muscular Dystrophy. By virtue of its linear regression-like characteristics, our new residual may prove useful in other contexts as well.

Host-to-host variation of ecological interactions in polymicrobial infections

Host-to-host variability with respect to interactions between microorganisms and multicellular hosts are commonly observed in infection and in homeostasis. However, the majority of mechanistic models used to analyze host-microorganism relationships, as well as most of the ecological theories proposed to explain coevolution of hosts and microbes, are based on averages across a host population. By assuming that observed variations are random and independent, these models overlook the role of differences between hosts. Here, we analyze mechanisms underlying host-to-host variations of bacterial infection kinetics, using the well characterized experimental infection model of polymicrobial otitis media (OM) in chinchillas, in combination with population dynamic models and a maximum entropy (MaxEnt) based inference scheme. We find that the nature of the interactions between bacterial species critically regulates host-to-host variations in these interactions. Surprisingly, seemingly unrelated phenomena, such as the efficiency of individual bacterial species in utilizing nutrients for growth, and the microbe-specific host immune response, can become interdependent in a host population. The latter finding suggests a potential mechanism that could lead to selection of specific strains of bacterial species during the coevolution of the host immune response and the bacterial species.

Cell responses only partially shape cell-to-cell variations in protein abundances in Escherichia coli chemotaxis

Cell-to-cell variations in protein abundance in clonal cell populations are ubiquitous in living systems. Because protein composition determines responses in individual cells, it stands to reason that the variations themselves are subject to selective pressures. However, the functional role of these cell-to-cell differences is not well understood. One way to tackle questions regarding relationships between form and function is to perturb the form (e.g., change the protein abundances) and observe the resulting changes in some function. Here, we take on the form-function relationship from the inverse perspective, asking instead what specific constraints on cell-to-cell variations in protein abundance are imposed by a given functional phenotype. We develop a maximum entropy-based approach to posing questions of this type and illustrate the method by application to the well-characterized chemotactic response in Escherichia coli. We find that full determination of observed cell-to-cell variations in protein abundances is not inherent in chemotaxis itself but, in fact, appears to be jointly imposed by the chemotaxis program in conjunction with other factors (e.g., the protein synthesis machinery and/or additional nonchemotactic cell functions, such as cell metabolism). These results illustrate the power of maximum entropy as a tool for the investigation of relationships between biological form and function.

Data-driven quantification of the robustness and sensitivity of cell signaling networks

Robustness and sensitivity of responses generated by cell signaling networks has been associated with survival and evolvability of organisms. However, existing methods analyzing robustness and sensitivity of signaling networks ignore the experimentally observed cell-to-cell variations of protein abundances and cell functions or contain ad hoc assumptions. We propose and apply a data-driven maximum entropy based method to quantify robustness and sensitivity of Escherichia coli (E. coli) chemotaxis signaling network. Our analysis correctly rank orders different models of E. coli chemotaxis based on their robustness and suggests that parameters regulating cell signaling are evolutionary selected to vary in individual cells according to their abilities to perturb cell functions. Furthermore, predictions from our approach regarding distribution of protein abundances and properties of chemotactic responses in individual cells based on cell population averaged data are in excellent agreement with their experimental counterparts. Our approach is general and can be used to evaluate robustness as well as generate predictions of single cell properties based on population averaged experimental data in a wide range of cell signaling systems.

In silico modeling of Itk activation kinetics in thymocytes suggests competing positive and negative IP4 mediated feedbacks increase robustness

The inositol-phosphate messenger inositol(1,3,4,5)tetrakisphosphate (IP₄) is essential for thymocyte positive selection by regulating plasma-membrane association of the protein tyrosine kinase Itk downstream of the T cell receptor (TCR). IP₄ can act as a soluble analog of the phosphoinositide 3-kinase (PI3K) membrane lipid product phosphatidylinositol(3,4,5)trisphosphate (PIP₃). PIP₃ recruits signaling proteins such as Itk to cellular membranes by binding to PH and other domains. In thymocytes, low-dose IP₄ binding to the Itk PH domain surprisingly promoted and high-dose IP₄ inhibited PIP₃ binding of Itk PH domains. However, the mechanisms that underlie the regulation of membrane recruitment of Itk by IP₄ and PIP₃ remain unclear. The distinct Itk PH domain ability to oligomerize is consistent with a cooperative-allosteric mode of IP₄ action. However, other possibilities cannot be ruled out due to difficulties in quantitatively measuring the interactions between Itk, IP₄ and PIP₃, and in generating non-oligomerizing Itk PH domain mutants. This has hindered a full mechanistic understanding of how IP₄ controls Itk function. By combining experimentally measured kinetics of PLCγ1 phosphorylation by Itk with in silico modeling of multiple Itk signaling circuits and a maximum entropy (MaxEnt) based computational approach, we show that those in silico models which are most robust against variations of protein and lipid expression levels and kinetic rates at the single cell level share a cooperative-allosteric mode of Itk regulation by IP₄ involving oligomeric Itk PH domains at the plasma membrane. This identifies MaxEnt as an excellent tool for quantifying robustness for complex TCR signaling circuits and provides testable predictions to further elucidate a controversial mechanism of PIP₃ signaling.

Competing positive and negative feedbacks mediated by PH domain ligand interactions regulate Itk activation kinetics in T Cells (178.3)

Inositol phosphate second messengers often regulate crosstalk between receptor signaling and lipid metabolic networks critically affecting cell decision processes, however, molecular mechanisms underlying such cross-regulation are poorly understood. Pairing mathematical modeling and experiments we elucidate these mechanisms in thymocyte activation initiated by T cell receptor (TCR) and antigen interactions. Thymocyte activation is carefully controlled by transient activation kinetics of the Tec-family protein tyrosine kinase Itk generated by TCR signaling, and, production of the membrane lipid phosphatidylinositol(3,4,5)trisphosphate (PIP3) and soluble inositol(1,3,4,5) tetrakisphosphate (IP4). By combining PLC-g (substrate of active Itk) activation kinetics in experiments with maximum entropy based computational approaches we show multiple possible in-silico models describing different modes of molecular interactions between Itk, PIP3, and IP4, can be distinguished. We show that models displaying maximum robustness share a cooperative-allosteric mode of Itk regulation by IP4 involving oligomeric Itk PH domains which induces dueling positive and negative feedbacks in Itk activation. Models lacking the feedbacks or containing monomeric Itk are significantly less robust. We also elucidate key mechanisms regulating the "shape" of the transient Itk kinetics that can be manipulated in experiments for developing therapeutic strategies targeting TCR signaling.

KELVIN: a software package for rigorous measurement of statistical evidence in human genetics

This paper describes the software package KELVIN, which supports the PPL (posterior probability of linkage) framework for the measurement of statistical evidence in human (or more generally, diploid) genetic studies. In terms of scope, KELVIN supports two-point (trait-marker or marker-marker) and multipoint linkage analysis, based on either sex-averaged or sex-specific genetic maps, with an option to allow for imprinting; trait-marker linkage disequilibrium (LD), or association analysis, in case-control data, trio data, and/or multiplex family data, with options for joint linkage and trait-marker LD or conditional LD given linkage; dichotomous trait, quantitative trait and quantitative trait threshold models; and certain types of gene-gene interactions and covariate effects. Features and data (pedigree) structures can be freely mixed and matched within analyses. The statistical framework is specifically tailored to accumulate evidence in a mathematically rigorous way across multiple data sets or data subsets while allowing for multiple sources of heterogeneity, and KELVIN itself utilizes sophisticated software engineering to provide a powerful and robust platform for studying the genetics of complex disorders.

Fast and accurate calculation of a computationally intensive statistic for mapping disease genes

Many statistical methods in biology utilize numerical integration in order to deal with moderately high-dimensional parameter spaces without closed form integrals. One such method is the PPL, a class of models for mapping and modeling genes for complex human disorders. While the most common approach to numerical integration in statistics is MCMC, this is not a good option for the PPL for a variety of reasons, leading us to develop an alternative integration method for this application. We utilize an established sub-region adaptive integration method, but adapt it to specific features of our application. These include division of the multi-dimensional integrals into three separate layers, implementing internal constraints on the parameter space, and calibrating the approximation to ensure adequate precision of results for our application. The proposed approach is compared with an empirically driven fixed grid scheme as well as other numerical integration methods. The new method is shown to require far fewer function evaluations compared to the alternatives while matching or exceeding the best of them in terms of accuracy. The savings in evaluations is sufficiently large that previously intractable problems are now feasible in real time.

A matrix-based multilevel approach to identify functional protein modules

Identifying functional modules is believed to reveal most cellular processes. There have been many computational approaches to investigate the underlying biological structures (Bader and Hogue, 2003; Dhillon et al., 2005; Krogan et al., 2006; Ramadan et al., 2005; Xiong et al., 2005; Zhang et al., 2004). A spectral clustering method plays a critical role identifying functional modules in a yeast protein-protein network (Ramadan et al., 2005). We present an unweighted-graph version of a multilevel spectral algorithm which more accurately identifies protein complexes with less computational time.