Latent patient profile modelling and applications with mixed-variate restricted Boltzmann machine

Efficient management of chronic diseases is critical in modern health care. We consider diabetes mellitus, and our ongoing goal is to examine how machine learning can deliver information for clinical efficiency. The challenge is to aggregate highly heterogeneous sources including demographics, diagnoses, pathologies and treatments, and extract similar groups so that care plans can be designed. To this end, we extend our recent model, the mixed-variate restricted Boltzmann machine (MV.RBM), as it seamlessly integrates multiple data types for each patient aggregated over time and outputs a homogeneous representation called "latent profile" that can be used for patient clustering, visualisation, disease correlation analysis and prediction. We demonstrate that the method outperforms all baselines on these tasks - the primary characteristics of patients in the same groups are able to be identified and the good result can be achieved for the diagnosis codes prediction.

Performance, personality, and energetics : correlation, causation, and mechanism

The study of phenotypic evolution should be an integrative endeavor that combines different approaches and crosses disciplinary and phylogenetic boundaries to consider complex traits and organisms that historically have been studied in isolation from each other. Analyses of individual variation within populations can act to bridge studies focused at the levels of morphology, physiology, biochemistry, organismal performance, behavior, and life history. For example, the study of individual variation recently facilitated the integration of behavior into the concept of a pace-of-life syndrome and effectively linked the field of energetics with research on animal personality. Here, we illustrate how studies on the pace-of-life syndrome and the energetics of personality can be integrated within a physiology-performance-behavior-fitness paradigm that includes consideration of ecological context. We first introduce key concepts and definitions and then review the rapidly expanding literature on the links between energy metabolism and personality traits commonly studied in nonhuman animals (activity, exploration, boldness, aggressiveness, sociability). We highlight some empirical literature involving mammals and squamates that demonstrates how emerging fields can develop in rather disparate ways because of historical accidents and/or particularities of different kinds of organisms. We then briefly discuss potentially interesting avenues for future conceptual and empirical research in relation to motivation, intraindividual variation, and mechanisms underlying trait correlations. The integration of performance traits within the pace-of-life-syndrome concept has the potential to fill a logical gap between the context dependency of selection and how energetics and personality are expected to interrelate. Studies of how performance abilities and/or aspects of Darwinian fitness relate to both metabolic rate and personality traits are particularly lacking.

Personality and the emergence of the pace-of-life syndrome concept at the population level

The pace-of-life syndrome (POLS) hypothesis specifies that closely related species or populations experiencing different ecological conditions should differ in a suite of metabolic, hormonal and immunity traits that have coevolved with the life-history particularities related to these conditions. Surprisingly, two important dimensions of the POLS concept have been neglected: (i) despite increasing evidence for numerous connections between behavioural, physiological and life-history traits, behaviours have rarely been considered in the POLS yet; (ii) the POLS could easily be applied to the study of covariation among traits between individuals within a population. In this paper, we propose that consistent behavioural differences among individuals, or personality, covary with life history and physiological differences at the within-population, interpopulation and interspecific levels. We discuss how the POLS provides a heuristic framework in which personality studies can be integrated to address how variation in personality traits is maintained within populations.

The pace of life under artificial selection : personality, energy expenditure, and longevity are correlated in domestic dogs

The domestic dog has undergone extensive artificial selection resulting in an extreme diversity in body size, personality, life‐history, and metabolic traits among breeds. Here we tested whether proactive personalities (high levels of activity, boldness, and aggression) are related to a fast “pace of life” (high rates of growth, mortality, and energy expenditure). Data from the literature provide preliminary evidence that artificial selection on dogs (through domestication) generated variations in personality traits that are correlated with life histories and metabolism. We found that obedient (or docile, shy) breeds live longer than disobedient (or bold) ones and that aggressive breeds have higher energy needs than unaggressive ones. These correlations could result from either human preference for particular trait combinations or, more likely, correlated responses to artificial selection on personality. Our results suggest the existence of a general pace‐of‐life syndrome arising from the coevolution of personality, metabolic, and life‐history traits.

Energy metabolism and animal personality

In this paper we show how animal personality could explain some of the large inter-individual variation in resting metabolic rate (MR) and explore methodological and functional linkages between personality and energetics. Personality will introduce variability in resting MR measures because individuals consistently differ in their stress response, exploration or activity levels, all of which influence MR measurements made with respirometry and the doubly-labelled water technique. Physiologists try to exclude these behavioural influences from resting MR measurements, but animal personality research indicates that these attempts are unlikely to be successful. For example, because reactive animals “freeze” when submitted to a stress, their MR could be classified as “resting” because of immobility when in fact they are highly stressed with an elevated MR. More importantly, recent research demonstrating that behavioural responses to novel and highly artificial stimuli are correlated with both behaviour and fitness under more natural circumstances calls into question the wisdom of excluding these behavioural influences on MR measurements. The reason that intra-specific variation in resting MR are so weakly correlated with daily energy expenditure (DEE) and fitness, may be that the latter two measures fully incorporate personality while the former partially excludes its influence. Because activity, exploration, boldness and aggressiveness are energetically costly, personality and metabolism should be correlated and physiological constraints may underlie behavioural syndromes. We show how physiological ecologists can better examine behavioural linkages between personality and metabolism, as required to better understand the physiological correlates of personality and the evolutionary consequences of metabolic variability.

Effect of sorghum flour addition on resistant starch content, phenolic profile and antioxidant capacity of durum wheat pasta

Foods containing elevated levels of health functional components such as resistant starch and polyphenolic antioxidants may have beneficial effects on human health. Pasta incorporating either red sorghum flour (RSF) or white sorghum flour (WSF) each at 20%, 30% and 40% substitution of durum wheat semolina (DWS) was prepared and compared to pasta made from 100% DWS (control) for content of starch fractions, phenolic profile and antioxidant capacity, before and after cooking. Total, digestible and resistant starch contents were determined by the AOAC method; individual phenolic acids and anthocyanins by reverse phase-HPLC analysis; total phenolic content by the Folin–Ciocalteu method and antioxidant capacity by the ABTS assay. The addition of both RSF and WSF increased the resistant starch content, bound phenolic acids, total phenolic content and antioxidant capacity at all incorporation levels compared to the control pasta; while free phenolic acids and anthocyanins were higher in the RSF-containing pasta only. Cooking did not change the resistant starch content of any of the pasta formulations. Cooking did however decrease the free phenolic acids, anthocyanins, total phenolic content and antioxidant capacity and increased the bound phenolic acids of the sorghum-containing pastas. The study suggests that these sorghum flours may be very useful for the preparation of pasta with increased levels of resistant starch and polyphenolic antioxidants.

Metabolic profile analysis of zebrafish embryos

A growing goal in the field of metabolism is to determine the impact of genetics on different aspects of mitochondrial function. Understanding these relationships will help to understand the underlying etiology for a range of diseases linked with mitochondrial dysfunction, such as diabetes and obesity. Recent advances in instrumentation, has enabled the monitoring of distinct parameters of mitochondrial function in cell lines or tissue explants. Here we present a method for a rapid and sensitive analysis of mitochondrial function parameters in vivo during zebrafish embryonic development using the Seahorse bioscience XF 24 extracellular flux analyser. This protocol utilizes the Islet Capture microplates where a single embryo is placed in each well, allowing measurement of bioenergetics, including: (i) basal respiration; (ii) basal mitochondrial respiration (iii) mitochondrial respiration due to ATP turnover; (iv) mitochondrial uncoupled respiration or proton leak and (iv) maximum respiration. Using this approach embryonic zebrafish respiration parameters can be compared between wild type and genetically altered embryos (mutant, gene over-expression or gene knockdown) or those manipulated pharmacologically. It is anticipated that dissemination of this protocol will provide researchers with new tools to analyse the genetic basis of metabolic disorders in vivo in this relevant vertebrate animal model.