Speaker: Cristian Meza (https://ingemat.uv.cl/instituto/academicos?view=article&id=23&catid=10) from Universitad de Valparaíso
Abstract: The study of the human microbiome based on genetic sequencing techniques produces particular longitudinal data that must be analyzed using complex mixed-effects models that can explain temporal variation, given its asymmetric nature and overabundance of zero values. The standard estimation procedures proposed to analyze this kind of compositional longitudinal data are based on log-likelihood approximation methods as for instance Gauss-Hermite quadrature. It is well known that this kind of numerical methods can produce inconsistent estimates and require a sufficiently large number of quadrature points, implying an often slow convergence that is not very stable. It is here that this work is based, as we seek to improve the quality of statistical inference in longitudinal microbiome data that will allow better decision-making and conclusions. Thus, in this work we propose a maximum likelihood estimation method for one of these models, the zero-inflated beta regression (ZIBR), based on the Stochastic Approximation Expectation Maximization (SAEM) algorithm which has showed to have good performances in complex mixed-effects models. The details of its application and implementation are presented, as well as the results in both simulated and real data. Comparisons with the method based on log-likelihood approximations such as Gauss-Hermite quadrature, in the context of simulation studies, show that the SAEM algorithm produces better results in parameter estimation and hypothesis testing in the realistic scenario of unbalanced data. Finally, the SAEM-based estimation method is used on two real examples of microbiome data (pediatric Inflammatory Bowel Disease patients and vaginal microbiome in pregnant women) and its results prove its usefulness in detecting changes in both the presence and abundance of bacterial taxa.
This work is in collaboration with John Barrera (PhD student, Universidad de Valparaíso).
Seminar series .ics calendar:
https://indico.math.cnrs.fr/category/711/events.ics
Program of the Biostatistics seminars:
https://indico.math.cnrs.fr/category/711/
Subscribe to the seminar mailing list:
https://diff.u-bordeaux.fr/sympa/subscribe/seminaire.biostat.bph
Former e-seminars on our YouTube channel (mostly in French): https://www.youtube.com/channel/UCURp-hEQL7k23UzGfqgEurA/videos
Biostatistics seminar series from the Department of Public Health from the University of Bordeaux and the Bordeaux Population Health UMR 1219 research center
Boris Hejblum
