Modeling personalized heart rate response to exercise and environmental factors with wearables data.

Heart rate (HR) response to workout intensity reflects fitness and cardiorespiratory health. Physiological models have been developed to describe such heart rate dynamics and characterize cardiorespiratory fitness. However, these models have been limited to small studies in controlled lab environments and are challenging to apply to noisy-but ubiquitous-data from wearables. We propose a hybrid approach that combines a physiological model with flexible neural network components to learn a personalized, multidimensional representation of fitness. The physiological model describes the evolution of heart rate during exercise using ordinary differential equations (ODEs). ODE parameters are dynamically derived via a neural network connecting personalized representations to external environmental factors, from area topography to weather and instantaneous workout intensity. Our approach efficiently fits the hybrid model to a large set of 270,707 workouts collected from wearables of 7465 users from the Apple Heart and Movement Study. The resulting model produces fitness representations that accurately predict full HR response to exercise intensity in future workouts, with a per-workout median error of 6.1 BPM [4.4-8.8 IQR]. We further demonstrate that the learned representations correlate with traditional metrics of cardiorespiratory fitness, such as VO2 max (explained variance 0.81 ± 0.003). Lastly, we illustrate how our model is naturally interpretable and explicitly describes the effects of environmental factors such as temperature and humidity on heart rate, e.g., high temperatures can increase heart rate by 10%. Combining physiological ODEs with flexible neural networks can yield interpretable, robust, and expressive models for health applications.

Multi-scale inference of genetic trait architecture using biologically annotated neural networks.

In this article, we present Biologically Annotated Neural Networks (BANNs), a nonlinear probabilistic framework for association mapping in genome-wide association (GWA) studies. BANNs are feedforward models with partially connected architectures that are based on biological annotations. This setup yields a fully interpretable neural network where the input layer encodes SNP-level effects, and the hidden layer models the aggregated effects among SNP-sets. We treat the weights and connections of the network as random variables with prior distributions that reflect how genetic effects manifest at different genomic scales. The BANNs software uses variational inference to provide posterior summaries which allow researchers to simultaneously perform (i) mapping with SNPs and (ii) enrichment analyses with SNP-sets on complex traits. Through simulations, we show that our method improves upon state-of-the-art association mapping and enrichment approaches across a wide range of genetic architectures. We then further illustrate the benefits of BANNs by analyzing real GWA data assayed in approximately 2,000 heterogenous stock of mice from the Wellcome Trust Centre for Human Genetics and approximately 7,000 individuals from the Framingham Heart Study. Lastly, using a random subset of individuals of European ancestry from the UK Biobank, we show that BANNs is able to replicate known associations in high and low-density lipoprotein cholesterol content.

Joint analysis of expression levels and histological images identifies genes associated with tissue morphology.

Histopathological images are used to characterize complex phenotypes such as tumor stage. Our goal is to associate features of stained tissue images with high-dimensional genomic markers. We use convolutional autoencoders and sparse canonical correlation analysis (CCA) on paired histological images and bulk gene expression to identify subsets of genes whose expression levels in a tissue sample correlate with subsets of morphological features from the corresponding sample image. We apply our approach, ImageCCA, to two TCGA data sets, and find gene sets associated with the structure of the extracellular matrix and cell wall infrastructure, implicating uncharacterized genes in extracellular processes. We find sets of genes associated with specific cell types, including neuronal cells and cells of the immune system. We apply ImageCCA to the GTEx v6 data, and find image features that capture population variation in thyroid and in colon tissues associated with genetic variants (image morphology QTLs, or imQTLs), suggesting that genetic variation regulates population variation in tissue morphological traits.

Sparse multi-output Gaussian processes for online medical time series prediction.

BACKGROUND: For real-time monitoring of hospital patients, high-quality inference of patients' health status using all information available from clinical covariates and lab test results is essential to enable successful medical interventions and improve patient outcomes. Developing a computational framework that can learn from observational large-scale electronic health records (EHRs) and make accurate real-time predictions is a critical step. In this work, we develop and explore a Bayesian nonparametric model based on multi-output Gaussian process (GP) regression for hospital patient monitoring. METHODS: We propose MedGP, a statistical framework that incorporates 24 clinical covariates and supports a rich reference data set from which relationships between observed covariates may be inferred and exploited for high-quality inference of patient state over time. To do this, we develop a highly structured sparse GP kernel to enable tractable computation over tens of thousands of time points while estimating correlations among clinical covariates, patients, and periodicity in patient observations. MedGP has a number of benefits over current methods, including (i) not requiring an alignment of the time series data, (ii) quantifying confidence regions in the predictions, (iii) exploiting a vast and rich database of patients, and (iv) inferring interpretable relationships among clinical covariates. RESULTS: We evaluate and compare results from MedGP on the task of online prediction for three patient subgroups from two medical data sets across 8,043 patients. We find MedGP improves online prediction over baseline and state-of-the-art methods for nearly all covariates across different disease subgroups and hospitals. CONCLUSIONS: The MedGP framework is robust and efficient in estimating the temporal dependencies from sparse and irregularly sampled medical time series data for online prediction. The publicly available code is at https://github.com/bee-hive/MedGP .

Statistical tests for detecting variance effects in quantitative trait studies.

Motivation: Identifying variants, both discrete and continuous, that are associated with quantitative traits, or QTs, is the primary focus of quantitative genetics. Most current methods are limited to identifying mean effects, or associations between genotype or covariates and the mean value of a quantitative trait. It is possible, however, that a variant may affect the variance of the quantitative trait in lieu of, or in addition to, affecting the trait mean. Here, we develop a general methodology to identify covariates with variance effects on a quantitative trait using a Bayesian heteroskedastic linear regression model (BTH). We compare BTH with existing methods to detect variance effects across a large range of simulations drawn from scenarios common to the analysis of quantitative traits. Results: We find that BTH and a double generalized linear model (dglm) outperform classical tests used for detecting variance effects in recent genomic studies. We show BTH and dglm are less likely to generate spurious discoveries through simulations and application to identifying methylation variance QTs and expression variance QTs. We identify four variance effects of sex in the Cardiovascular and Pharmacogenetics study. Our work is the first to offer a comprehensive view of variance identifying methodology. We identify shortcomings in previously used methodology and provide a more conservative and robust alternative. We extend variance effect analysis to a wide array of covariates that enables a new statistical dimension in the study of sex and age specific quantitative trait effects. Availability and implementation: https://github.com/b2du/bth. Supplementary information: Supplementary data are available at Bioinformatics online.

Incorporating prior information into association studies.

UNLABELLED: Recent technological developments in measuring genetic variation have ushered in an era of genome-wide association studies which have discovered many genes involved in human disease. Current methods to perform association studies collect genetic information and compare the frequency of variants in individuals with and without the disease. Standard approaches do not take into account any information on whether or not a given variant is likely to have an effect on the disease. We propose a novel method for computing an association statistic which takes into account prior information. Our method improves both power and resolution by 8% and 27%, respectively, over traditional methods for performing association studies when applied to simulations using the HapMap data. Advantages of our method are that it is as simple to apply to association studies as standard methods, the results of the method are interpretable as the method reports p-values, and the method is optimal in its use of prior information in regards to statistical power. AVAILABILITY: The method presented herein is available at http://masa.cs.ucla.edu.

Mechanism of cis-inhibition of polyQ fibrillation by polyP: PPII oligomers and the hydrophobic effect.

PolyQ peptides teeter between polyproline II (PPII) and beta-sheet conformations. In tandem polyQ-polyP peptides, the polyP segment tips the balance toward PPII, increasing the threshold number of Gln residues needed for fibrillation. To investigate the mechanism of cis-inhibition by flanking polyP segments on polyQ fibrillation, we examined short polyQ, polyP, and tandem polyQ-polyP peptides. These polyQ peptides have only three glutamines and cannot form beta-sheet fibrils. We demonstrate that polyQ-polyP peptides form small, soluble oligomers at high concentrations (as shown by size exclusion chromatography and diffusion coefficient measurements) with PPII structure (as shown by circular dichroism spectroscopy and (3)J(HN-C alpha) constants of Gln residues from constant time correlation spectroscopy NMR). Nuclear Overhauser effect spectroscopy and molecular modeling suggest that self-association of these peptides occurs as a result of both hydrophobic and steric effects. Pro side chains present three methylenes to solvent, favoring self-association of polyP through the hydrophobic effect. Gln side chains, with two methylene groups, can adopt a conformation similar to that of Pro side chains, also permitting self-association through the hydrophobic effect. Furthermore, steric clashes between Gln and Pro side chains to the C-terminal side of the polyQ segment favor adoption of the PPII-like structure in the polyQ segment. The conformational adaptability of the polyQ segment permits the cis-inhibitory effect of polyP segments on fibrillation by the polyQ segments in proteins such as huntingtin.

Flanking polyproline sequences inhibit beta-sheet structure in polyglutamine segments by inducing PPII-like helix structure.

Polyglutamine (poly(Q)) expansion is associated with protein aggregation into beta-sheet amyloid fibrils and neuronal cytotoxicity. In the mutant poly(Q) protein huntingtin, associated with Huntington's disease, both aggregation and cytotoxicity may be abrogated by a polyproline (poly(P)) domain flanking the C terminus of the poly(Q) region. To understand structural changes that may occur with the addition of the poly(P) sequence, we synthesized poly(Q) peptides with 3-15 glutamine residues and a corresponding set of poly(Q) peptides flanked on the C terminus by 11 proline residues (poly(Q)-poly(P)), as occurs in the huntingtin sequence. The shorter soluble poly(Q) peptides (three or six glutamine residues) showed polyproline type II-like (PPII)-like helix conformation when examined by circular dichroism spectroscopy and were monomers as judged by size-exclusion chromatography (SEC), while the longer poly(Q) peptides (nine or 15 glutamine residues) showed a beta-sheet conformation by CD and defined oligomers by SEC. Soluble poly(Q)-poly(P) peptides showed PPII-like content but SEC showed poorly defined, overlapping oligomeric peaks, and as judged by CD these peptides retained significant PPII-like structure with increasing poly(Q) length. More importantly, addition of the poly(P) domain increased the threshold for fibril formation to approximately 15 glutamine residues. X-ray diffraction, electron microscopy, and film CD showed that, while poly(Q) peptides with >or=6 glutamine residues formed beta-sheet-rich fibrils, only the longest poly(Q)-poly(P) peptide (15 glutamine residues) did so. From these and other observations, we propose that poly(Q) domains exist in a "tug-of-war" between two conformations, a PPII-like helix and a beta-sheet, while the poly(P) domain is conformationally constrained into a proline type II helix (PPII). Addition of poly(P) to the C terminus of a poly(Q) domain induces a PPII-like structure, which opposes the aggregation-prone beta-sheet. These structural observations may shed light on the threshold phenomenon of poly(Q) aggregation, and support the hypothesized evolution of "protective" poly(P) tracts adjacent to poly(Q) aggregation domains.