BN-BacArena: Bayesian network extension of BacArena for the dynamic simulation of microbial communities

MOTIVATION

Simulating gut microbial dynamics is extremely challenging. Several computational tools, notably the widely used BacArena, enable modeling of dynamic changes in the microbial environment. These methods, however, do not comprehensively account for microbe-microbe stimulant or inhibitory effects or for nutrient-microbe inhibitory effects, typically observed in different compounds present in the daily diet.

RESULTS

Here, we present BN-BacArena, an extension of BacArena consisting on the incorporation within the native computational framework of a Bayesian network model that accounts for microbe-microbe and nutrient-microbe interactions. Using in vitro experiments, 16S rRNA gene sequencing data and nutritional composition of 55 foods, the output Bayesian network showed 23 significant nutrient-bacteria interactions, suggesting the importance of compounds such as polyols, ascorbic acid, polyphenols and other phytochemicals, and 40 bacteria-bacteria significant relationships. With test data, BN-BacArena demonstrates a statistically significant improvement over BacArena to predict the time-dependent relative abundance of bacterial species involved in the gut microbiota upon different nutritional interventions. As a result, BN-BacArena opens new avenues for the dynamic modeling and simulation of the human gut microbiota metabolism.

AVAILABILITY AND IMPLEMENTATION

MATLAB and R code are available in https://github.com/PlanesLab/BN-BacArena.

Feature Saliencies in Asymmetric Hidden Markov Models

Many real-life problems are stated as nonlabeled high-dimensional data. Current strategies to select features are mainly focused on labeled data, which reduces the options to select relevant features for unsupervised problems, such as clustering. Recently, feature saliency models have been introduced and developed as clustering models to select and detect relevant variables/features as the model is learned. Usually, these models assume that all variables are independent, which narrows their applicability. This article introduces asymmetric hidden Markov models with feature saliencies, i.e., models capable of simultaneously determining during their learning phase relevant variables/features and probabilistic relationships between variables. The proposed models are compared with other state-of-the-art approaches using synthetic data and real data related to grammatical face videos and wear in ball bearings. We show that the proposed models have better or equal fitness than other state-of-the-art models and provide further data insights.

Learning massive interpretable gene regulatory networks of the human brain by merging Bayesian networks

We present the Fast Greedy Equivalence Search (FGES)-Merge, a new method for learning the structure of gene regulatory networks via merging locally learned Bayesian networks, based on the fast greedy equivalent search algorithm. The method is competitive with the state of the art in terms of the Matthews correlation coefficient, which takes into account both precision and recall, while also improving upon it in terms of speed, scaling up to tens of thousands of variables and being able to use empirical knowledge about the topological structure of gene regulatory networks. To showcase the ability of our method to scale to massive networks, we apply it to learning the gene regulatory network for the full human genome using data from samples of different brain structures (from the Allen Human Brain Atlas). Furthermore, this Bayesian network model should predict interactions between genes in a way that is clear to experts, following the current trends in explainable artificial intelligence. To achieve this, we also present a new open-access visualization tool that facilitates the exploration of massive networks and can aid in finding nodes of interest for experimental tests.

NeuroSuites: An online platform for running neuroscience, statistical, and machine learning tools

Nowadays, an enormous amount of high dimensional data is available in the field of neuroscience. Handling these data is complex and requires the use of efficient tools to transform them into useful knowledge. In this work we present NeuroSuites, an easy-access web platform with its own architecture. We compare our platform with other software currently available, highlighting its main strengths. Thanks to its defined architecture, it is able to handle large-scale problems common in some neuroscience fields. NeuroSuites has different neuroscience-oriented applications and tools to integrate statistical data analysis and machine learning algorithms commonly used in this field. As future work, we want to further expand the list of available software tools as well as improve the platform interface according to user demands.

Autoregressive Asymmetric Linear Gaussian Hidden Markov Models

In a real life process evolving over time, the relationship between its relevant variables may change. Therefore, it is advantageous to have different inference models for each state of the process. Asymmetric hidden Markov models fulfil this dynamical requirement and provide a framework where the trend of the process can be expressed as a latent variable. In this paper, we modify these recent asymmetric hidden Markov models to have an asymmetric autoregressive component in the case of continuous variables, allowing the model to choose the order of autoregression that maximizes its penalized likelihood for a given training set. Additionally, we show how inference, hidden states decoding and parameter learning must be adapted to fit the proposed model. Finally, we run experiments with synthetic and real data to show the capabilities of this new model.

Hybrid semiparametric Bayesian networks

This paper presents a new class of Bayesian networks called hybrid semiparametric Bayesian networks, which can model hybrid data (discrete and continuous data) by mixing parametric and nonparametric estimation models. The parametric estimation models can represent a conditional linear Gaussian relationship between variables, while the nonparametric estimation model can represent other types of relationships, such as non-Gaussian and nonlinear relationships. This new class of Bayesian networks generalizes the conditional linear Gaussian Bayesian networks, including them as a special case. In addition, we describe a learning procedure for the structure and the parameters of our proposed type of Bayesian network. This learning procedure finds the best combination of parametric and nonparametric models automatically from data. This requires the definition of a cross-validated score. We also detail how new data can be sampled from a hybrid semiparametric Bayesian network, which in turn can be useful to solve other related tasks, such as inference. Furthermore, we intuitively relate our proposal with adaptive kernel density estimation models. The experimental results show that hybrid semiparametric Bayesian networks are a valuable contribution when dealing with data that do not meet the parametric assumptions that are expected for other models, such as conditional linear Gaussian Bayesian networks. We include experiments with synthetic data and real-world data from the UCI repository which demonstrate the good performance and the ability to extract useful information about the relationship between the variables in the model.

Identifying Parkinson's disease subtypes with motor and non-motor symptoms via model-based multi-partition clustering

Identification of Parkinson's disease subtypes may help understand underlying disease mechanisms and provide personalized management. Although clustering methods have been previously used for subtyping, they have reported generic subtypes of limited relevance in real life practice because patients do not always fit into a single category. The aim of this study was to identify new subtypes assuming that patients could be grouped differently according to certain sets of related symptoms. To this purpose, a novel model-based multi-partition clustering method was applied on data from an international, multi-center, cross-sectional study of 402 Parkinson's disease patients. Both motor and non-motor symptoms were considered. As a result, eight sets of related symptoms were identified. Each of them provided a different way to group patients: impulse control issues, overall non-motor symptoms, presence of dyskinesias and pyschosis, fatigue, axial symptoms and motor fluctuations, autonomic dysfunction, depression, and excessive sweating. Each of these groups could be seen as a subtype of the disease. Significant differences between subtypes (P< 0.01) were found in sex, age, age of onset, disease duration, Hoehn & Yahr stage, and treatment. Independent confirmation of these results could have implications for the clinical management of Parkinson's disease patients.

Patient specific prediction of temporal lobe epilepsy surgical outcomes

OBJECTIVE

Drug-resistant temporal lobe epilepsy (TLE) is the most common type of epilepsy for which patients undergo surgery. Despite the best clinical judgment and currently available prediction algorithms, surgical outcomes remain variable. We aimed to build and to evaluate the performance of multidimensional Bayesian network classifiers (MBCs), a type of probabilistic graphical model, at predicting probability of seizure freedom after TLE surgery.

METHODS

Clinical, neurophysiological, and imaging variables were collected from 231 TLE patients who underwent surgery at the University of California, San Francisco (UCSF) or the Montreal Neurological Institute (MNI) over a 15-year period. Postsurgical Engel outcomes at year 1 (Y1), Y2, and Y5 were analyzed as primary end points. We trained an MBC model on combined data sets from both institutions. Bootstrap bias corrected cross-validation (BBC-CV) was used to evaluate the performance of the models.

RESULTS

The MBC was compared with logistic regression and Cox proportional hazards according to the area under the receiver-operating characteristic curve (AUC). The MBC achieved an AUC of 0.67 at Y1, 0.72 at Y2, and 0.67 at Y5, which indicates modest performance yet superior to what has been reported in the state-of-the-art studies to date.

SIGNIFICANCE

The MBC can more precisely encode probabilistic relationships between predictors and class variables (Engel outcomes), achieving promising experimental results compared to other well-known statistical methods. Multisite application of the MBC could further optimize its classification accuracy with prospective data sets. Online access to the MBC is provided, paving the way for its use as an adjunct clinical tool in aiding pre-operative TLE surgical counseling.

Comparing the Electrophysiology and Morphology of Human and Mouse Layer 2/3 Pyramidal Neurons With Bayesian Networks

Pyramidal neurons are the most common neurons in the cerebral cortex. Understanding how they differ between species is a key challenge in neuroscience. We compared human temporal cortex and mouse visual cortex pyramidal neurons from the Allen Cell Types Database in terms of their electrophysiology and dendritic morphology. We found that, among other differences, human pyramidal neurons had a higher action potential threshold voltage, a lower input resistance, and larger dendritic arbors. We learned Gaussian Bayesian networks from the data in order to identify correlations and conditional independencies between the variables and compare them between the species. We found strong correlations between electrophysiological and morphological variables in both species. In human cells, electrophysiological variables were correlated even with morphological variables that are not directly related to dendritic arbor size or diameter, such as mean bifurcation angle and mean branch tortuosity. Cortical depth was correlated with both electrophysiological and morphological variables in both species, and its effect on electrophysiology could not be explained in terms of the morphological variables. For some variables, the effect of cortical depth was opposite in the two species. Overall, the correlations among the variables differed strikingly between human and mouse neurons. Besides identifying correlations and conditional independencies, the learned Bayesian networks might be useful for probabilistic reasoning regarding the morphology and electrophysiology of pyramidal neurons.

A community-based transcriptomics classification and nomenclature of neocortical cell types

To understand the function of cortical circuits, it is necessary to catalog their cellular diversity. Past attempts to do so using anatomical, physiological or molecular features of cortical cells have not resulted in a unified taxonomy of neuronal or glial cell types, partly due to limited data. Single-cell transcriptomics is enabling, for the first time, systematic high-throughput measurements of cortical cells and generation of datasets that hold the promise of being complete, accurate and permanent. Statistical analyses of these data reveal clusters that often correspond to cell types previously defined by morphological or physiological criteria and that appear conserved across cortical areas and species. To capitalize on these new methods, we propose the adoption of a transcriptome-based taxonomy of cell types for mammalian neocortex. This classification should be hierarchical and use a standardized nomenclature. It should be based on a probabilistic definition of a cell type and incorporate data from different approaches, developmental stages and species. A community-based classification and data aggregation model, such as a knowledge graph, could provide a common foundation for the study of cortical circuits. This community-based classification, nomenclature and data aggregation could serve as an example for cell type atlases in other parts of the body.

Classification of GABAergic interneurons by leading neuroscientists

There is currently no unique catalog of cortical GABAergic interneuron types. In 2013, we asked 48 leading neuroscientists to classify 320 interneurons by inspecting images of their morphology. That study was the first to quantify the degree of agreement among neuroscientists in morphology-based interneuron classification, showing high agreement for the chandelier and Martinotti types, yet low agreement for most of the remaining types considered. Here we present the dataset containing the classification choices by the neuroscientists according to interneuron type as well as to five prominent morphological features. These data can be used as crisp or soft training labels for learning supervised machine learning interneuron classifiers, while further analyses can try to pinpoint anatomical characteristics that make an interneuron especially difficult or especially easy to classify.

Towards a supervised classification of neocortical interneuron morphologies

BACKGROUND

The challenge of classifying cortical interneurons is yet to be solved. Data-driven classification into established morphological types may provide insight and practical value.

RESULTS

We trained models using 217 high-quality morphologies of rat somatosensory neocortex interneurons reconstructed by a single laboratory and pre-classified into eight types. We quantified 103 axonal and dendritic morphometrics, including novel ones that capture features such as arbor orientation, extent in layer one, and dendritic polarity. We trained a one-versus-rest classifier for each type, combining well-known supervised classification algorithms with feature selection and over- and under-sampling. We accurately classified the nest basket, Martinotti, and basket cell types with the Martinotti model outperforming 39 out of 42 leading neuroscientists. We had moderate accuracy for the double bouquet, small and large basket types, and limited accuracy for the chandelier and bitufted types. We characterized the types with interpretable models or with up to ten morphometrics.

CONCLUSION

Except for large basket, 50 high-quality reconstructions sufficed to learn an accurate model of a type. Improving these models may require quantifying complex arborization patterns and finding correlates of bouton-related features. Our study brings attention to practical aspects important for neuron classification and is readily reproducible, with all code and data available online.

3D morphology-based clustering and simulation of human pyramidal cell dendritic spines

The dendritic spines of pyramidal neurons are the targets of most excitatory synapses in the cerebral cortex. They have a wide variety of morphologies, and their morphology appears to be critical from the functional point of view. To further characterize dendritic spine geometry, we used in this paper over 7,000 individually 3D reconstructed dendritic spines from human cortical pyramidal neurons to group dendritic spines using model-based clustering. This approach uncovered six separate groups of human dendritic spines. To better understand the differences between these groups, the discriminative characteristics of each group were identified as a set of rules. Model-based clustering was also useful for simulating accurate 3D virtual representations of spines that matched the morphological definitions of each cluster. This mathematical approach could provide a useful tool for theoretical predictions on the functional features of human pyramidal neurons based on the morphology of dendritic spines.

Dendritic spines of pyramidal neurons are the targets of most excitatory synapses in the cerebral cortex and their morphology appears to be critical from the functional point of view. Thus, characterizing this morphology is necessary to link structural and functional spine data and thus interpret and make them more meaningful. We have used a large database of more than 7,000 individually 3D reconstructed dendritic spines from human cortical pyramidal neurons that is first transformed into a set of 54 quantitative features characterizing spine geometry mathematically. The resulting data set is grouped into spine clusters based on a probabilistic model with Gaussian finite mixtures. We uncover six groups of spines whose discriminative characteristics are identified with machine learning methods as a set of rules. The clustering model allows us to simulate accurate spines from human pyramidal neurons to suggest new hypotheses of the functional organization of these cells.

MultiMap: A Tool to Automatically Extract and Analyse Spatial Microscopic Data From Large Stacks of Confocal Microscopy Images

The development of 3D visualization and reconstruction methods to analyse microscopic structures at different levels of resolutions is of great importance to define brain microorganization and connectivity. MultiMap is a new tool that allows the visualization, 3D segmentation and quantification of fluorescent structures selectively in the neuropil from large stacks of confocal microscopy images. The major contribution of this tool is the posibility to easily navigate and create regions of interest of any shape and size within a large brain area that will be automatically 3D segmented and quantified to determine the density of puncta in the neuropil. As a proof of concept, we focused on the analysis of glutamatergic and GABAergic presynaptic axon terminals in the mouse hippocampal region to demonstrate its use as a tool to provide putative excitatory and inhibitory synaptic maps. The segmentation and quantification method has been validated over expert labeled images of the mouse hippocampus and over two benchmark datasets, obtaining comparable results to the expert detections.

Parkinson's Disease Subtypes Identified from Cluster Analysis of Motor and Non-motor Symptoms

Parkinson's disease is now considered a complex, multi-peptide, central, and peripheral nervous system disorder with considerable clinical heterogeneity. Non-motor symptoms play a key role in the trajectory of Parkinson's disease, from prodromal premotor to end stages. To understand the clinical heterogeneity of Parkinson's disease, this study used cluster analysis to search for subtypes from a large, multi-center, international, and well-characterized cohort of Parkinson's disease patients across all motor stages, using a combination of cardinal motor features (bradykinesia, rigidity, tremor, axial signs) and, for the first time, specific validated rater-based non-motor symptom scales. Two independent international cohort studies were used: (a) the validation study of the Non-Motor Symptoms Scale (n = 411) and (b) baseline data from the global Non-Motor International Longitudinal Study (n = 540). k-means cluster analyses were performed on the non-motor and motor domains (domains clustering) and the 30 individual non-motor symptoms alone (symptoms clustering), and hierarchical agglomerative clustering was performed to group symptoms together. Four clusters are identified from the domains clustering supporting previous studies: mild, non-motor dominant, motor-dominant, and severe. In addition, six new smaller clusters are identified from the symptoms clustering, each characterized by clinically-relevant non-motor symptoms. The clusters identified in this study present statistical confirmation of the increasingly important role of non-motor symptoms (NMS) in Parkinson's disease heterogeneity and take steps toward subtype-specific treatment packages.

Wiring Economy of Pyramidal Cells in the Juvenile Rat Somatosensory Cortex

Ever since Cajal hypothesized that the structure of neurons is designed in such a way as to save space, time and matter, numerous researchers have analyzed wiring properties at different scales of brain organization. Here we test the hypothesis that individual pyramidal cells, the most abundant type of neuron in the cerebral cortex, optimize brain connectivity in terms of wiring length. In this study, we analyze the neuronal wiring of complete basal arborizations of pyramidal neurons in layer II, III, IV, Va, Vb and VI of the hindlimb somatosensory cortical region of postnatal day 14 rats. For each cell, we search for the optimal basal arborization and compare its length with the length of the real dendritic structure. Here the optimal arborization is defined as the arborization that has the shortest total wiring length provided that all neuron bifurcations are respected and the extent of the dendritic arborizations remain unchanged. We use graph theory and evolutionary computation techniques to search for the minimal wiring arborizations. Despite morphological differences between pyramidal neurons located in different cortical layers, we found that the neuronal wiring is near-optimal in all cases (the biggest difference between the shortest synthetic wiring found for a dendritic arborization and the length of its real wiring was less than 5%). We found, however, that the real neuronal wiring was significantly closer to the best solution found in layers II, III and IV. Our studies show that the wiring economy of cortical neurons is related not to the type of neurons or their morphological complexities but to general wiring economy principles.

Dendritic-branching angles of pyramidal neurons of the human cerebral cortex

In this article, we analyze branching angles of the basal dendrites of pyramidal neurons of layers III and V of the human temporal cortex. For this, we use a novel probability directional statistical distribution called truncated von Mises distribution that is able to describe more accurately the dendritic-branching angles than the previous proposals. Then, we perform comparative studies using this statistical method to determine similarities and/or differences between branches and branching angles that belong to different cortical layers and regions. Using this methodology, we found that common design principles exist and govern the patterns found in the different branches that compose the basal dendrites of human pyramidal cells of the temporal cortex. However, particular differences were found between supra and infragranular cells. Furthermore, we compared the branching angles of human layer III pyramidal neurons with data obtained in the previous studies in layer III of both the rat somatosensory cortex and of several cortical areas of the mouse. Finally, we study the branching angle differences between the humans that compose our data.

Data Publications Correlate with Citation Impact

Neuroscience and molecular biology have been generating large datasets over the past years that are reshaping how research is being conducted. In their wake, open data sharing has been singled out as a major challenge for the future of research. We conducted a comparative study of citations of data publications in both fields, showing that the average publication tagged with a data-related term by the NCBI MeSH (Medical Subject Headings) curators achieves a significantly larger citation impact than the average in either field. We introduce a new metric, the data article citation index (DAC-index), to identify the most prolific authors among those data-related publications. The study is fully reproducible from an executable Rmd (R Markdown) script together with all the citation datasets. We hope these results can encourage authors to more openly publish their data.

A Graphical Decision-Theoretic Model for Neonatal Jaundice

Background. Neonatal jaundice is treated daily at all hospitals. However, the routine, urgency, and case load of most doctors stop them from carefully analyzing all the factors that they would like to (and should) take into account. This article develops a complex decision support system for neonatal jaundice management. Methods. The problem is represented by means of an influence diagram, including admission and treatment decisions. The corresponding uncertainty model is built with the aid of both historical data and subjective judgments. Parents and doctors were interviewed to elicit a multiattribute utility function. The decision analysis cycle is completed with sensitivity analyses and explanations of the results. Results. The construction and use of this decision support system for jaundice management have induced a profound change in daily medical practice, avoiding aggressive treatments—there have been no exchange transfusions in the past 3 years—and reducing the lengths of stay at the hospital. More information is now taken into account to decide on treatments. Interestingly, after embarking on this modeling effort, physicians came to view jaundice as a much more difficult problem than they had initially thought. Comparisons between real cases and system proposals revealed that treatments by nonexpert doctors tend to be longer than what expert doctors would administer. Conclusion. The system is especially designed to help neonatologists in situations in which their lack of experience may lead to unnecessary treatments. Different points of view from several expert doctors and, more interestingly, from parents are taken into account. This knowledge gives a broader picture of the medical problem— incorporating new action criteria, new agents to intervene, more uncertainty variables—to get an insight into the suitability of each therapeutic decision for each patient situation. The benefits gained and the usefulness perceived by neonatologists are worth the increased and time-consuming effort of developing this complex system. Although specially designed for a specific hospital and for neonatal jaundice management, it can be easily adapted to other hospitals and problems.

Dendritic and Axonal Wiring Optimization of Cortical GABAergic Interneurons

The way in which a neuronal tree expands plays an important role in its functional and computational characteristics. We aimed to study the existence of an optimal neuronal design for different types of cortical GABAergic neurons. To do this, we hypothesized that both the axonal and dendritic trees of individual neurons optimize brain connectivity in terms of wiring length. We took the branching points of real three-dimensional neuronal reconstructions of the axonal and dendritic trees of different types of cortical interneurons and searched for the minimal wiring arborization structure that respects the branching points. We compared the minimal wiring arborization with real axonal and dendritic trees. We tested this optimization problem using a new approach based on graph theory and evolutionary computation techniques. We concluded that neuronal wiring is near-optimal in most of the tested neurons, although the wiring length of dendritic trees is generally nearer to the optimum. Therefore, wiring economy is related to the way in which neuronal arborizations grow irrespective of the marked differences in the morphology of the examined interneurons.

Dendritic branching angles of pyramidal cells across layers of the juvenile rat somatosensory cortex

The characterization of the structural design of cortical microcircuits is essential for understanding how they contribute to function in both health and disease. Since pyramidal neurons represent the most abundant neuronal type and their dendritic spines constitute the major postsynaptic elements of cortical excitatory synapses, our understanding of the synaptic organization of the neocortex largely depends on the available knowledge regarding the structure of pyramidal cells. Previous studies have identified several apparently common rules in dendritic geometry. We study the dendritic branching angles of pyramidal cells across layers to further shed light on the principles that determine the geometric shapes of these cells. We find that the dendritic branching angles of pyramidal cells from layers II-VI of the juvenile rat somatosensory cortex suggest common design principles, despite the particular morphological and functional features that are characteristic of pyramidal cells in each cortical layer. J. Comp. Neurol. 524:2567-2576, 2016. © 2016 Wiley Periodicals, Inc.

Laminar Differences in Dendritic Structure of Pyramidal Neurons in the Juvenile Rat Somatosensory Cortex

Pyramidal cell structure varies between different cortical areas and species, indicating that the cortical circuits that these cells participate in are likely to be characterized by different functional capabilities. Structural differences between cortical layers have been traditionally reported using either the Golgi method or intracellular labeling, but the structure of pyramidal cells has not previously been systematically analyzed across all cortical layers at a particular age. In the present study, we investigated the dendritic architecture of complete basal arbors of pyramidal neurons in layers II, III, IV, Va, Vb, and VI of the hindlimb somatosensory cortical region of postnatal day 14 rats. We found that the characteristics of basal dendritic morphologies are statistically different in each cortical layer. The variations in size and branching pattern that exist between pyramidal cells of different cortical layers probably reflect the particular functional properties that are characteristic of the cortical circuit in which they participate. This new set of complete basal dendritic arbors of 3D-reconstructed pyramidal cell morphologies across each cortical layer will provide new insights into interlaminar information processing in the cerebral cortex.

A univocal definition of the neuronal soma morphology using Gaussian mixture models

The definition of the soma is fuzzy, as there is no clear line demarcating the soma of the labeled neurons and the origin of the dendrites and axon. Thus, the morphometric analysis of the neuronal soma is highly subjective. In this paper, we provide a mathematical definition and an automatic segmentation method to delimit the neuronal soma. We applied this method to the characterization of pyramidal cells, which are the most abundant neurons in the cerebral cortex. Since there are no benchmarks with which to compare the proposed procedure, we validated the goodness of this automatic segmentation method against manual segmentation by neuroanatomists to set up a framework for comparison. We concluded that there were no significant differences between automatically and manually segmented somata, i.e., the proposed procedure segments the neurons similarly to how a neuroanatomist does. It also provides univocal, justifiable and objective cutoffs. Thus, this study is a means of characterizing pyramidal neurons in order to objectively compare the morphometry of the somata of these neurons in different cortical areas and species.

The Vallecas Project: A Cohort to Identify Early Markers and Mechanisms of Alzheimer's Disease

Introduction

Alzheimer’s disease (AD) is a major threat for the well-being of an increasingly aged world population. The physiopathological mechanisms of late-onset AD are multiple, possibly heterogeneous, and not well understood. Different combinations of variables from several domains (i.e., clinical, neuropsychological, structural, and biochemical markers) may predict dementia conversion, according to distinct physiopathological pathways, in different groups of subjects.

Methods

We launched the Vallecas Project (VP), a cohort study of non-demented people aged 70–85, to characterize the social, clinical, neuropsychological, structural, and biochemical underpinnings of AD inception. Given the exploratory nature of the VP, multidimensional and machine learning techniques will be applied, in addition to the traditional multivariate statistical methods.

Results

A total of 1169 subjects were recruited between October 2011 and December 2013. Mean age was 74.4 years (SD 3.9), 63.5% of the subjects were women, and 17.9% of the subjects were carriers of at least one ε4 allele of the apolipoprotein E gene. Cognitive diagnoses at inclusion were as follows: normal cognition 93.0% and mild cognitive impairment (MCI) 7.0% (3.1% amnestic MCI, 0.1% non-amnestic MCI, 3.8% mixed MCI). Blood samples were obtained and stored for future determinations in 99.9% of the subjects and 3T magnetic resonance imaging study was conducted in 89.9% of the volunteers. The cohort is being followed up annually for 4 years after the baseline.

Conclusion

We have established a valuable homogeneous single-center cohort which, by identifying groups of variables associated with high risk of MCI or dementia conversion, should help to clarify the early physiopathological mechanisms of AD and should provide avenues for prompt diagnosis and AD prevention.

Classifying GABAergic interneurons with semi-supervised projected model-based clustering

OBJECTIVES

A recently introduced pragmatic scheme promises to be a useful catalog of interneuron names. We sought to automatically classify digitally reconstructed interneuronal morphologies according to this scheme. Simultaneously, we sought to discover possible subtypes of these types that might emerge during automatic classification (clustering). We also investigated which morphometric properties were most relevant for this classification.

MATERIALS AND METHODS

A set of 118 digitally reconstructed interneuronal morphologies classified into the common basket (CB), horse-tail (HT), large basket (LB), and Martinotti (MA) interneuron types by 42 of the world's leading neuroscientists, quantified by five simple morphometric properties of the axon and four of the dendrites. We labeled each neuron with the type most commonly assigned to it by the experts. We then removed this class information for each type separately, and applied semi-supervised clustering to those cells (keeping the others' cluster membership fixed), to assess separation from other types and look for the formation of new groups (subtypes). We performed this same experiment unlabeling the cells of two types at a time, and of half the cells of a single type at a time. The clustering model is a finite mixture of Gaussians which we adapted for the estimation of local (per-cluster) feature relevance. We performed the described experiments on three different subsets of the data, formed according to how many experts agreed on type membership: at least 18 experts (the full data set), at least 21 (73 neurons), and at least 26 (47 neurons).

RESULTS

Interneurons with more reliable type labels were classified more accurately. We classified HT cells with 100% accuracy, MA cells with 73% accuracy, and CB and LB cells with 56% and 58% accuracy, respectively. We identified three subtypes of the MA type, one subtype of CB and LB types each, and no subtypes of HT (it was a single, homogeneous type). We got maximum (adapted) Silhouette width and ARI values of 1, 0.83, 0.79, and 0.42, when unlabeling the HT, CB, LB, and MA types, respectively, confirming the quality of the formed cluster solutions. The subtypes identified when unlabeling a single type also emerged when unlabeling two types at a time, confirming their validity. Axonal morphometric properties were more relevant that dendritic ones, with the axonal polar histogram length in the [π, 2π) angle interval being particularly useful.

CONCLUSIONS

The applied semi-supervised clustering method can accurately discriminate among CB, HT, LB, and MA interneuron types while discovering potential subtypes, and is therefore useful for neuronal classification. The discovery of potential subtypes suggests that some of these types are more heterogeneous that previously thought. Finally, axonal variables seem to be more relevant than dendritic ones for distinguishing among the CB, HT, LB, and MA interneuron types.

Multi-dimensional classification of GABAergic interneurons with Bayesian network-modeled label uncertainty

Interneuron classification is an important and long-debated topic in neuroscience. A recent study provided a data set of digitally reconstructed interneurons classified by 42 leading neuroscientists according to a pragmatic classification scheme composed of five categorical variables, namely, of the interneuron type and four features of axonal morphology. From this data set we now learned a model which can classify interneurons, on the basis of their axonal morphometric parameters, into these five descriptive variables simultaneously. Because of differences in opinion among the neuroscientists, especially regarding neuronal type, for many interneurons we lacked a unique, agreed-upon classification, which we could use to guide model learning. Instead, we guided model learning with a probability distribution over the neuronal type and the axonal features, obtained, for each interneuron, from the neuroscientists' classification choices. We conveniently encoded such probability distributions with Bayesian networks, calling them label Bayesian networks (LBNs), and developed a method to predict them. This method predicts an LBN by forming a probabilistic consensus among the LBNs of the interneurons most similar to the one being classified. We used 18 axonal morphometric parameters as predictor variables, 13 of which we introduce in this paper as quantitative counterparts to the categorical axonal features. We were able to accurately predict interneuronal LBNs. Furthermore, when extracting crisp (i.e., non-probabilistic) predictions from the predicted LBNs, our method outperformed related work on interneuron classification. Our results indicate that our method is adequate for multi-dimensional classification of interneurons with probabilistic labels. Moreover, the introduced morphometric parameters are good predictors of interneuron type and the four features of axonal morphology and thus may serve as objective counterparts to the subjective, categorical axonal features.

Bayesian networks in neuroscience: a survey

Bayesian networks are a type of probabilistic graphical models lie at the intersection between statistics and machine learning. They have been shown to be powerful tools to encode dependence relationships among the variables of a domain under uncertainty. Thanks to their generality, Bayesian networks can accommodate continuous and discrete variables, as well as temporal processes. In this paper we review Bayesian networks and how they can be learned automatically from data by means of structure learning algorithms. Also, we examine how a user can take advantage of these networks for reasoning by exact or approximate inference algorithms that propagate the given evidence through the graphical structure. Despite their applicability in many fields, they have been little used in neuroscience, where they have focused on specific problems, like functional connectivity analysis from neuroimaging data. Here we survey key research in neuroscience where Bayesian networks have been used with different aims: discover associations between variables, perform probabilistic reasoning over the model, and classify new observations with and without supervision. The networks are learned from data of any kind-morphological, electrophysiological, -omics and neuroimaging-, thereby broadening the scope-molecular, cellular, structural, functional, cognitive and medical- of the brain aspects to be studied.

Three-dimensional distribution of cortical synapses: a replicated point pattern-based analysis

The biggest problem when analyzing the brain is that its synaptic connections are extremely complex. Generally, the billions of neurons making up the brain exchange information through two types of highly specialized structures: chemical synapses (the vast majority) and so-called gap junctions (a substrate of one class of electrical synapse). Here we are interested in exploring the three-dimensional spatial distribution of chemical synapses in the cerebral cortex. Recent research has showed that the three-dimensional spatial distribution of synapses in layer III of the neocortex can be modeled by a random sequential adsorption (RSA) point process, i.e., synapses are distributed in space almost randomly, with the only constraint that they cannot overlap. In this study we hypothesize that RSA processes can also explain the distribution of synapses in all cortical layers. We also investigate whether there are differences in both the synaptic density and spatial distribution of synapses between layers. Using combined focused ion beam milling and scanning electron microscopy (FIB/SEM), we obtained three-dimensional samples from the six layers of the rat somatosensory cortex and identified and reconstructed the synaptic junctions. A total volume of tissue of approximately 4500μm(3) and around 4000 synapses from three different animals were analyzed. Different samples, layers and/or animals were aggregated and compared using RSA replicated spatial point processes. The results showed no significant differences in the synaptic distribution across the different rats used in the study. We found that RSA processes described the spatial distribution of synapses in all samples of each layer. We also found that the synaptic distribution in layers II to VI conforms to a common underlying RSA process with different densities per layer. Interestingly, the results showed that synapses in layer I had a slightly different spatial distribution from the other layers.

PREDICTING THE EQ-5D FROM THE PARKINSON'S DISEASE QUESTIONNAIRE PDQ-8 USING MULTI-DIMENSIONAL BAYESIAN NETWORK CLASSIFIERS

The impact of the Parkinson's disease and its treatment on the patients' health-related quality of life can be estimated either by means of generic measures such as the european quality of Life-5 Dimensions (EQ-5D) or specific measures such as the 8-item Parkinson's disease questionnaire (PDQ-8). In clinical studies, PDQ-8 could be used in detriment of EQ-5D due to the lack of resources, time or clinical interest in generic measures. Nevertheless, PDQ-8 cannot be applied in cost-effectiveness analyses which require generic measures and quantitative utility scores, such as EQ-5D. To deal with this problem, a commonly used solution is the prediction of EQ-5D from PDQ-8. In this paper, we propose a new probabilistic method to predict EQ-5D from PDQ-8 using multi-dimensional Bayesian network classifiers. Our approach is evaluated using five-fold cross-validation experiments carried out on a Parkinson's data set containing 488 patients, and is compared with two additional Bayesian network-based approaches, two commonly used mapping methods namely, ordinary least squares and censored least absolute deviations, and a deterministic model. Experimental results are promising in terms of predictive performance as well as the identification of dependence relationships among EQ-5D and PDQ-8 items that the mapping approaches are unable to detect.