EvoVGM

• DOI: 10.1145/3535508.3545563
• Publication Date: 2022-07-28T22:26:03Z
• Authors: Amine M. Remita; Abdoulaye Baniré ...
• Keywords: Sequence (biology); Robustness; Generative model

Most evolutionary-oriented deep generative models do not explicitly consider the underlying evolutionary dynamics of biological sequences as it is performed within Bayesian phylogenetic inference framework. In this study, we propose a method for variational model (EvoVGM) that jointly approximates true posterior local parameters and generates sequence alignments. Moreover, instantiated tuned continuous-time Markov chain substitution such JC69, K80 GTR. We train via low-variance stochastic estimator gradient ascent algorithm. Here, analyze consistency effectiveness EvoVGM on synthetic alignments simulated with several scenarios different sizes. Finally, highlight robustness fine-tuned using alignment gene S coronaviruses.