arXiv subjects

Philippe Lemey

Publications and source records attributed to Philippe Lemey.

34 records · Page 2Linked to original sources

arXiv2019

Massive parallelization boosts big Bayesian multidimensional scaling

Big Bayes is the computationally intensive co-application of big data and large, expressive Bayesian models for the analysis of complex phenomena in scientific inference and statistical learning. Standing as an example, Bayesian multidimensional scaling (MDS) can help scientists learn viral trajectories through space-time, but its computational burden prevents its wider use. Crucial MDS model calculations scale quadratically in the number of observations. We partially mitigate this limitation through massive parallelization using multi-core central processing units, instruction-level vectorization and graphics processing units (GPUs). Fitting the MDS model using Hamiltonian Monte Carlo, GPUs can deliver more than 100-fold speedups over serial calculations and thus extend Bayesian MDS to a big data setting. To illustrate, we employ Bayesian MDS to infer the rate at which different seasonal influenza virus subtypes use worldwide air traffic to spread around the globe. We examine 5392 viral sequences and their associated 14 million pairwise distances arising from the number of commercial airline seats per year between viral sampling locations. To adjust for shared evolutionary history of the viruses, we implement a phylogenetic extension to the MDS model and learn that subtype H3N2 spreads most effectively, consistent with its epidemic success relative to other seasonal influenza subtypes. Finally, we provide MassiveMDS, an open-source, stand-alone C++ library and rudimentary R package, and discuss program design and high-level implementation with an emphasis on important aspects of computing architecture that become relevant at scale.

stat.CO↗

arXiv2019

Relaxed random walks at scale

Relaxed random walk (RRW) models of trait evolution introduce branch-specific rate multipliers to modulate the variance of a standard Brownian diffusion process along a phylogeny and more accurately model overdispersed biological data. Increased taxonomic sampling challenges inference under RRWs as the number of unknown parameters grows with the number of taxa. To solve this problem, we present a scalable method to efficiently fit RRWs and infer this branch-specific variation in a Bayesian framework. We develop a Hamiltonian Monte Carlo (HMC) sampler to approximate the high-dimensional, correlated posterior that exploits a closed-form evaluation of the gradient of the trait data log-likelihood with respect to all branch-rate multipliers simultaneously. Our gradient calculation achieves computational complexity that scales only linearly with the number of taxa under study. We compare the efficiency of our HMC sampler to the previously standard univariable Metropolis-Hastings approach while studying the spatial emergence of the West Nile virus in North America in the early 2000s. Our method achieves an over 300-fold speed-increase over the univariable approach. Additionally, we demonstrate the scalability of our method by applying the RRW to study the correlation between five mammalian life history traits in a phylogenetic tree with 3650 tips.

q-bio.PE↗

arXiv2019

Markov-modulated continuous-time Markov chains to identify site- and branch-specific evolutionary variation

Markov models of character substitution on phylogenies form the foundation of phylogenetic inference frameworks. Early models made the simplifying assumption that the substitution process is homogeneous over time and across sites in the molecular sequence alignment. While standard practice adopts extensions that accommodate heterogeneity of substitution rates across sites, heterogeneity in the process over time in a site-specific manner remains frequently overlooked. This is problematic, as evolutionary processes that act at the molecular level are highly variable, subjecting different sites to different selective constraints over time, impacting their substitution behaviour. We propose incorporating time variability through Markov-modulated models (MMMs) that allow the substitution process (including relative character exchange rates as well as the overall substitution rate) that models the evolution at an individual site to vary across lineages. We implement a general MMM framework in BEAST, a popular Bayesian phylogenetic inference software package, allowing researchers to compose a wide range of MMMs through flexible XML specification. Using examples from bacterial, viral and plastid genome evolution, we show that MMMs impact phylogenetic tree estimation and can substantially improve model fit compared to standard substitution models. Through simulations, we show that marginal likelihood estimation accurately identifies the generative model and does not systematically prefer the more parameter-rich MMMs. In order to mitigate the increased computational demands associated with MMMs, our implementation exploits recently developed updates to BEAGLE, a high-performance computational library for phylogenetic inference.

q-bio.PE↗

arXiv2019

Inferring phenotypic trait evolution on large trees with many incomplete measurements

Comparative biologists are often interested in inferring covariation between multiple biological traits sampled across numerous related taxa. To properly study these relationships, we must control for the shared evolutionary history of the taxa to avoid spurious inference. Existing control techniques almost universally scale poorly as the number of taxa increases. An additional challenge arises as obtaining a full suite of measurements becomes increasingly difficult with increasing taxa. This typically necessitates data imputation or integration that further exacerbates scalability. We propose an inference technique that integrates out missing measurements analytically and scales linearly with the number of taxa by using a post-order traversal algorithm under a multivariate Brownian diffusion (MBD) model to characterize trait evolution. We further exploit this technique to extend the MBD model to account for sampling error or non-heritable residual variance. We test these methods to examine mammalian life history traits, prokaryotic genomic and phenotypic traits, and HIV infection traits. We find computational efficiency increases that top two orders-of-magnitude over current best practices. While we focus on the utility of this algorithm in phylogenetic comparative methods, our approach generalizes to solve long-standing challenges in computing the likelihood for matrix-normal and multivariate normal distributions with missing data at scale.

stat.ME↗

arXiv2019

Gradients do grow on trees: a linear-time ${\cal O}\hspace{-0.2em}\left( N \right)$-dimensional gradient for statistical phylogenetics

Calculation of the log-likelihood stands as the computational bottleneck for many statistical phylogenetic algorithms. Even worse is its gradient evaluation, often used to target regions of high probability. Order ${\cal O}\hspace{-0.2em}\left( N \right)$-dimensional gradient calculations based on the standard pruning algorithm require ${\cal O}\hspace{-0.2em}\left( N^2 \right)$ operations where N is the number of sampled molecular sequences. With the advent of high-throughput sequencing, recent phylogenetic studies have analyzed hundreds to thousands of sequences, with an apparent trend towards even larger data sets as a result of advancing technology. Such large-scale analyses challenge phylogenetic reconstruction by requiring inference on larger sets of process parameters to model the increasing data heterogeneity. To make this tractable, we present a linear-time algorithm for ${\cal O}\hspace{-0.2em}\left( N \right)$-dimensional gradient evaluation and apply it to general continuous-time Markov processes of sequence substitution on a phylogenetic tree without a need to assume either stationarity or reversibility. We apply this approach to learn the branch-specific evolutionary rates of three pathogenic viruses: West Nile virus, Dengue virus and Lassa virus. Our proposed algorithm significantly improves inference efficiency with a 126- to 234-fold increase in maximum-likelihood optimization and a 16- to 33-fold computational performance increase in a Bayesian framework.

stat.CO↗

arXiv2018

Bayesian Best-Arm Identification for Selecting Influenza Mitigation Strategies

Pandemic influenza has the epidemic potential to kill millions of people. While various preventive measures exist (i.a., vaccination and school closures), deciding on strategies that lead to their most effective and efficient use remains challenging. To this end, individual-based epidemiological models are essential to assist decision makers in determining the best strategy to curb epidemic spread. However, individual-based models are computationally intensive and it is therefore pivotal to identify the optimal strategy using a minimal amount of model evaluations. Additionally, as epidemiological modeling experiments need to be planned, a computational budget needs to be specified a priori. Consequently, we present a new sampling technique to optimize the evaluation of preventive strategies using fixed budget best-arm identification algorithms. We use epidemiological modeling theory to derive knowledge about the reward distribution which we exploit using Bayesian best-arm identification algorithms (i.e., Top-two Thompson sampling and BayesGap). We evaluate these algorithms in a realistic experimental setting and demonstrate that it is possible to identify the optimal strategy using only a limited number of model evaluations, i.e., 2-to-3 times faster compared to the uniform sampling method, the predominant technique used for epidemiological decision making in the literature. Finally, we contribute and evaluate a statistic for Top-two Thompson sampling to inform the decision makers about the confidence of an arm recommendation.

cs.LG↗

arXiv2017

Phylogenetic Factor Analysis

Phylogenetic comparative methods explore the relationships between quantitative traits adjusting for shared evolutionary history. This adjustment often occurs through a Brownian diffusion process along the branches of the phylogeny that generates model residuals or the traits themselves. For high-dimensional traits, inferring all pair-wise correlations within the multivariate diffusion is limiting. To circumvent this problem, we propose phylogenetic factor analysis (PFA) that assumes a small unknown number of independent evolutionary factors arise along the phylogeny and these factors generate clusters of dependent traits. Set in a Bayesian framework, PFA provides measures of uncertainty on the factor number and groupings, combines both continuous and discrete traits, integrates over missing measurements and incorporates phylogenetic uncertainty with the help of molecular sequences. We develop Gibbs samplers based on dynamic programming to estimate the PFA posterior distribution, over three-fold faster than for multivariate diffusion and a further order-of-magnitude more efficiently in the presence of latent traits. We further propose a novel marginal likelihood estimator for previously impractical models with discrete data and find that PFA also provides a better fit than multivariate diffusion in evolutionary questions in columbine flower development, placental reproduction transitions and triggerfish fin morphometry.

stat.ME↗

arXiv2016

Understanding Past Population Dynamics: Bayesian Coalescent-Based Modeling with Covariates

Effective population size characterizes the genetic variability in a population and is a parameter of paramount importance in population genetics. Kingman's coalescent process enables inference of past population dynamics directly from molecular sequence data, and researchers have developed a number of flexible coalescent-based models for Bayesian nonparametric estimation of the effective population size as a function of time. A major goal of demographic reconstruction is understanding the association between the effective population size and potential explanatory factors. Building upon Bayesian nonparametric coalescent-based approaches, we introduce a flexible framework that incorporates time-varying covariates through Gaussian Markov random fields. To approximate the posterior distribution, we adapt efficient Markov chain Monte Carlo algorithms designed for highly structured Gaussian models. Incorporating covariates into the demographic inference framework enables the modeling of associations between the effective population size and covariates while accounting for uncertainty in population histories. Furthermore, it can lead to more precise estimates of population dynamics. We apply our model to four examples. We reconstruct the demographic history of raccoon rabies in North America and find a significant association with the spatiotemporal spread of the outbreak. Next, we examine the effective population size trajectory of the DENV-4 virus in Puerto Rico along with viral isolate count data and find similar cyclic patterns. We compare the population history of the HIV-1 CRF02_AG clade in Cameroon with HIV incidence and prevalence data and find that the effective population size is more reflective of incidence rate. Finally, we explore the hypothesis that the population dynamics of musk ox during the Late Quaternary period were related to climate change.

stat.ME↗

arXiv2015

A Relaxed Drift Diffusion Model for Phylogenetic Trait Evolution

Understanding the processes that give rise to quantitative measurements associated with molecular sequence data remains an important issue in statistical phylogenetics. Examples of such measurements include geographic coordinates in the context of phylogeography and phenotypic traits in the context of comparative studies. A popular approach is to model the evolution of continuously varying traits as a Brownian diffusion process. However, standard Brownian diffusion is quite restrictive and may not accurately characterize certain trait evolutionary processes. Here, we relax one of the major restrictions of standard Brownian diffusion by incorporating a nontrivial estimable drift into the process. We introduce a relaxed drift diffusion model for the evolution of multivariate continuously varying traits along a phylogenetic tree via Brownian diffusion with drift. Notably, the relaxed drift model accommodates branch-specific variation of drift rates while preserving model identifiability. We implement the relaxed drift model in a Bayesian inference framework to simultaneously reconstruct the evolutionary histories of molecular sequence data and associated multivariate continuous trait data, and provide tools to visualize evolutionary reconstructions. We illustrate our approach in three viral examples. In the first two, we examine the spatiotemporal spread of HIV-1 in central Africa and West Nile virus in North America and show that a relaxed drift approach uncovers a clearer, more detailed picture of the dynamics of viral dispersal than standard Brownian diffusion. Finally, we study antigenic evolution in the context of HIV-1 resistance to three broadly neutralizing antibodies. Our analysis reveals evidence of a continuous drift at the HIV-1 population level towards enhanced resistance to neutralization by the VRC01 monoclonal antibody over the course of the epidemic.

q-bio.PE↗

arXiv2015

Assessing phenotypic correlation through the multivariate phylogenetic latent liability model

Understanding which phenotypic traits are consistently correlated throughout evolution is a highly pertinent problem in modern evolutionary biology. Here, we propose a multivariate phylogenetic latent liability model for assessing the correlation between multiple types of data, while simultaneously controlling for their unknown shared evolutionary history informed through molecular sequences. The latent formulation enables us to consider in a single model combinations of continuous traits, discrete binary traits and discrete traits with multiple ordered and unordered states. Previous approaches have entertained a single data type generally along a fixed history, precluding estimation of correlation between traits and ignoring uncertainty in the history. We implement our model in a Bayesian phylogenetic framework, and discuss inference techniques for hypothesis testing. Finally, we showcase the method through applications to columbine flower morphology, antibiotic resistance in Salmonella and epitope evolution in influenza.

q-bio.PE↗

arXiv2015

Spatio-temporal Dynamics of Foot-and-Mouth Disease Virus in South America

Although foot-and-mouth disease virus (FMDV) incidence has decreased in South America over the last years, the pathogen still circulates in the region and the risk of re-emergence in previously FMDV-free areas is a veterinary public health concern. In this paper we merge environmental, epidemiological and genetic data to reconstruct spatiotemporal patterns and determinants of FMDV serotypes A and O dispersal in South America. Our dating analysis suggests that serotype A emerged in South America around 1930, while serotype O emerged around 1990. The rate of evolution for serotype A was significantly higher compared to serotype O. Phylogeographic inference identified two well-connected sub networks of viral flow, one including Venezuela, Colombia and Ecuador; another including Brazil, Uruguay and Argentina. The spread of serotype A was best described by geographic distances, while trade of live cattle was the predictor that best explained serotype O spread. Our findings show that the two serotypes have different underlying evolutionary and spatial dynamics and may pose different threats to control programmes. Key-words: Phylogeography, foot-and-mouth disease virus, South America, animal trade.

q-bio.PE↗

arXiv2014

Synonymous and Nonsynonymous Distances Help Untangle Convergent Evolution and Recombination

When estimating a phylogeny from a multiple sequence alignment, researchers often assume the absence of recombination. However, if recombination is present, then tree estimation and all downstream analyses will be impacted, because different segments of the sequence alignment support different phylogenies. Similarly, convergent selective pressures at the molecular level can also lead to phylogenetic tree incongruence across the sequence alignment. Current methods for detection of phylogenetic incongruence are not equipped to distinguish between these two different mechanisms and assume that the incongruence is a result of recombination or other horizontal transfer of genetic information. We propose a new recombination detection method that can make this distinction, based on synonymous codon substitution distances. Although some power is lost by discarding the information contained in the nonsynonymous substitutions, our new method has lower false positive probabilities than the comparable recombination detection method when the phylogenetic incongruence signal is due to convergent evolution. We apply our method to three empirical examples, where we analyze: 1) sequences from a transmission network of the human immunodeficiency virus, 2) tlpB gene sequences from a geographically diverse set of 38 Helicobacter pylori strains, and 3) Hepatitis C virus sequences sampled longitudinally from one patient.

stat.ME↗

arXiv2013

Integrating influenza antigenic dynamics with molecular evolution

Influenza viruses undergo continual antigenic evolution allowing mutant viruses to evade host immunity acquired to previous virus strains. Antigenic phenotype is often assessed through pairwise measurement of cross-reactivity between influenza strains using the hemagglutination inhibition (HI) assay. Here, we extend previous approaches to antigenic cartography, and simultaneously characterize antigenic and genetic evolution by modeling the diffusion of antigenic phenotype over a shared virus phylogeny. Using HI data from influenza lineages A/H3N2, A/H1N1, B/Victoria and B/Yamagata, we determine patterns of antigenic drift across viral lineages, showing that A/H3N2 evolves faster and in a more punctuated fashion than other influenza lineages. We also show that year-to-year antigenic drift appears to drive incidence patterns within each influenza lineage. This work makes possible substantial future advances in investigating the dynamics of influenza and other antigenically-variable pathogens by providing a model that intimately combines molecular and antigenic evolution.

q-bio.PE↗

arXiv2013

piBUSS: a parallel BEAST/BEAGLE utility for sequence simulation under complex evolutionary scenarios

Background: Simulated nucleotide or amino acid sequences are frequently used to assess the performance of phylogenetic reconstruction methods. BEAST, a Bayesian statistical framework that focuses on reconstructing time-calibrated molecular evolutionary processes, supports a wide array of evolutionary models, but lacked matching machinery for simulation of character evolution along phylogenies. Results: We present a flexible Monte Carlo simulation tool, called piBUSS, that employs the BEAGLE high performance library for phylogenetic computations within BEAST to rapidly generate large sequence alignments under complex evolutionary models. piBUSS sports a user-friendly graphical user interface (GUI) that allows combining a rich array of models across an arbitrary number of partitions. A command-line interface mirrors the options available through the GUI and facilitates scripting in large-scale simulation studies. Analogous to BEAST model and analysis setup, more advanced simulation options are supported through an extensible markup language (XML) specification, which in addition to generating sequence output, also allows users to combine simulation and analysis in a single BEAST run. Conclusions: piBUSS offers a unique combination of flexibility and ease-of-use for sequence simulation under realistic evolutionary scenarios. Through different interfaces, piBUSS supports simulation studies ranging from modest endeavors for illustrative purposes to complex and large-scale assessments of evolutionary inference procedures. The software aims at implementing new models and data types that are continuously being developed as part of BEAST/BEAGLE.

q-bio.PE↗

arXiv2013

Inferring Heterogeneous Evolutionary Processes Through Time: from sequence substitution to phylogeography

Molecular phylogenetic and phylogeographic reconstructions generally assume time-homogeneous substitution processes. Motivated by computational convenience, this assumption sacrifices biological realism and offers little opportunity to uncover the temporal dynamics in evolutionary histories. Here, we extend and generalize an evolutionary approach that relaxes the time-homogeneous process assumption by allowing the specification of different infinitesimal substitution rate matrices across different time intervals, called epochs, along the evolutionary history. We focus on an epoch model implementation in a Bayesian inference framework that offers great modeling flexibility in drawing inference about any discrete data type characterized as a continuous-time Markov chain, including phylogeographic traits. To alleviate the computational burden that the additional temporal heterogeneity imposes, we adopt a massively parallel approach that achieves both fine- and coarse-grain parallelization of the computations across branches that accommodate epoch transitions, making extensive use of graphics processing units. Through synthetic examples, we assess model performance in recovering evolutionary parameters from data generated according to different evolutionary scenarios that comprise different numbers of epochs for both nucleotide and codon substitution processes. We illustrate the usefulness of our inference framework in two different applications to empirical data sets: the selection dynamics on within-host HIV populations throughout infection and the seasonality of global influenza circulation. In both cases, our epoch model captures key features of temporal heterogeneity that remained difficult to test using ad hoc procedures.

q-bio.PE↗

arXiv2012

The seasonal flight of influenza: a unified framework for spatiotemporal hypothesis testing

Global mobility flow data are at the heart of spatial epidemiological models used to predict infectious disease behavior but this wealth of data on human mobility has been largely neglected by reconstructions of pathogen evolutionary dynamics using viral genetic data. Although stochastic models of viral evolution may potentially be informed by such data, a major challenge lies in deciding which mobility processes are critical and to what extent they contribute to shaping contemporaneous distributions of pathogen diversity. Here, we develop a framework to integrate predictors of viral diffusion with phylogeographic inference and estimate human influenza H3N2 migration history while simultaneously testing and quantifying the factors that underly it. We provide evidence for air travel governing the global dynamics of human influenza whereas other processes act at a more local scale.

q-bio.PE↗