Literature DB >> 25328261

Estimation for general birth-death processes.

Forrest W Crawford1, Vladimir N Minin2, Marc A Suchard3.   

Abstract

Birth-death processes (BDPs) are continuous-time Markov chains that track the number of "particles" in a system over time. While widely used in population biology, genetics and ecology, statistical inference of the instantaneous particle birth and death rates remains largely limited to restrictive linear BDPs in which per-particle birth and death rates are constant. Researchers often observe the number of particles at discrete times, necessitating data augmentation procedures such as expectation-maximization (EM) to find maximum likelihood estimates. For BDPs on finite state-spaces, there are powerful matrix methods for computing the conditional expectations needed for the E-step of the EM algorithm. For BDPs on infinite state-spaces, closed-form solutions for the E-step are available for some linear models, but most previous work has resorted to time-consuming simulation. Remarkably, we show that the E-step conditional expectations can be expressed as convolutions of computable transition probabilities for any general BDP with arbitrary rates. This important observation, along with a convenient continued fraction representation of the Laplace transforms of the transition probabilities, allows for novel and efficient computation of the conditional expectations for all BDPs, eliminating the need for truncation of the state-space or costly simulation. We use this insight to derive EM algorithms that yield maximum likelihood estimation for general BDPs characterized by various rate models, including generalized linear models. We show that our Laplace convolution technique outperforms competing methods when they are available and demonstrate a technique to accelerate EM algorithm convergence. We validate our approach using synthetic data and then apply our methods to cancer cell growth and estimation of mutation parameters in microsatellite evolution.

Entities:  

Keywords:  Birth-death process; EM algorithm; MM algorithm; continuous-time Markov chain; maximum likelihood estimation; microsatellite evolution

Year:  2014        PMID: 25328261      PMCID: PMC4196218          DOI: 10.1080/01621459.2013.866565

Source DB:  PubMed          Journal:  J Am Stat Assoc        ISSN: 0162-1459            Impact factor:   5.033


  33 in total

1.  An expectation maximization algorithm for training hidden substitution models.

Authors:  I Holmes; G M Rubin
Journal:  J Mol Biol       Date:  2002-04-12       Impact factor: 5.469

2.  Evolutionary HMMs: a Bayesian approach to multiple alignment.

Authors:  I Holmes; W J Bruno
Journal:  Bioinformatics       Date:  2001-09       Impact factor: 6.937

3.  Mutation biases and mutation rate variation around very short human microsatellites revealed by human-chimpanzee-orangutan genomic sequence alignments.

Authors:  William Amos
Journal:  J Mol Evol       Date:  2010-08-11       Impact factor: 2.395

Review 4.  Biological applications of the theory of birth-and-death processes.

Authors:  Artem S Novozhilov; Georgy P Karev; Eugene V Koonin
Journal:  Brief Bioinform       Date:  2006-03       Impact factor: 11.622

5.  Counting labeled transitions in continuous-time Markov models of evolution.

Authors:  Vladimir N Minin; Marc A Suchard
Journal:  J Math Biol       Date:  2007-09-14       Impact factor: 2.259

6.  Antilymphoma effects of anti-HLA-DR and CD20 monoclonal antibodies (Lym-1 and Rituximab) on human lymphoma cells.

Authors:  Cathy Liu; Gerald DeNardo; Evan Tobin; Sally DeNardo
Journal:  Cancer Biother Radiopharm       Date:  2004-10       Impact factor: 3.099

7.  The reconstructed evolutionary process.

Authors:  S Nee; R M May; P H Harvey
Journal:  Philos Trans R Soc Lond B Biol Sci       Date:  1994-05-28       Impact factor: 6.237

Review 8.  Evolutionary dynamics of microsatellite DNA.

Authors:  C Schlötterer
Journal:  Chromosoma       Date:  2000-09       Impact factor: 4.316

9.  Equilibrium distributions of microsatellite repeat length resulting from a balance between slippage events and point mutations.

Authors:  S Kruglyak; R T Durrett; M D Schug; C F Aquadro
Journal:  Proc Natl Acad Sci U S A       Date:  1998-09-01       Impact factor: 11.205

10.  Microsatellite evolution inferred from human-chimpanzee genomic sequence alignments.

Authors:  Matthew T Webster; Nick G C Smith; Hans Ellegren
Journal:  Proc Natl Acad Sci U S A       Date:  2002-06-17       Impact factor: 11.205

View more
  12 in total

1.  Model-Based Detection of Whole-Genome Duplications in a Phylogeny.

Authors:  Arthur Zwaenepoel; Yves Van de Peer
Journal:  Mol Biol Evol       Date:  2020-09-01       Impact factor: 16.240

2.  Social interactions among grazing reef fish drive material flux in a coral reef ecosystem.

Authors:  Michael A Gil; Andrew M Hein
Journal:  Proc Natl Acad Sci U S A       Date:  2017-04-10       Impact factor: 11.205

3.  Computational methods for birth-death processes.

Authors:  Forrest W Crawford; Lam Si Tung Ho; Marc A Suchard
Journal:  Wiley Interdiscip Rev Comput Stat       Date:  2018-01-02

4.  Probabilistic Modeling of Reprogramming to Induced Pluripotent Stem Cells.

Authors:  Lin L Liu; Justin Brumbaugh; Ori Bar-Nur; Zachary Smith; Matthias Stadtfeld; Alexander Meissner; Konrad Hochedlinger; Franziska Michor
Journal:  Cell Rep       Date:  2016-12-20       Impact factor: 9.423

5.  Likelihood-based inference for discretely observed birth-death-shift processes, with applications to evolution of mobile genetic elements.

Authors:  Jason Xu; Peter Guttorp; Midori Kato-Maeda; Vladimir N Minin
Journal:  Biometrics       Date:  2015-07-06       Impact factor: 2.571

6.  Coupling bounds for approximating birth-death processes by truncation.

Authors:  Forrest W Crawford; Timothy C Stutz; Kenneth Lange
Journal:  Stat Probab Lett       Date:  2016-02-01       Impact factor: 0.870

7.  Birth/birth-death processes and their computable transition probabilities with biological applications.

Authors:  Lam Si Tung Ho; Jason Xu; Forrest W Crawford; Vladimir N Minin; Marc A Suchard
Journal:  J Math Biol       Date:  2017-07-24       Impact factor: 2.259

8.  Efficient Transition Probability Computation for Continuous-Time Branching Processes via Compressed Sensing.

Authors:  Jason Xu; Vladimir N Minin
Journal:  Uncertain Artif Intell       Date:  2015-07

9.  SEX, LIES AND SELF-REPORTED COUNTS: BAYESIAN MIXTURE MODELS FOR HEAPING IN LONGITUDINAL COUNT DATA VIA BIRTH-DEATH PROCESSES.

Authors:  Forrest W Crawford; Robert E Weiss; Marc A Suchard
Journal:  Ann Appl Stat       Date:  2015       Impact factor: 2.083

10.  Estimating dose-specific cell division and apoptosis rates from chemo-sensitivity experiments.

Authors:  Yiyi Liu; Forrest W Crawford
Journal:  Sci Rep       Date:  2018-02-09       Impact factor: 4.379

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.