VGsim: Scalable viral genealogy simulator for global pandemic

Area of Science:

• Computational Biology
• Epidemiology
• Molecular Evolution

Background:

• Accurate simulation of biological processes is crucial for developing new technologies and inference methods.
• Millions of SARS-CoV-2 genomes have been sequenced, providing rich data for studying pandemic spread and viral evolution.
• Existing frameworks lack the efficiency and flexibility for world-scale pandemic simulation and large-scale genealogy generation.

Purpose of the Study:

• To introduce VGsim, a novel, fast simulator for generating viral genealogies under epidemiological models.
• To address the need for efficient and flexible simulation tools for large-scale pandemic scenarios.
• To enable detailed analysis of viral evolution, including natural selection and variant identification.

Main Methods:

• VGsim employs a two-phase simulation process: a forward run using a hierarchical Gillespie algorithm for population dynamics and a backward run using a coalescent-like approach for genealogy generation.
• The forward run models the pandemic's dynamics by generating a chain of population-level events.
• The backward run constructs a sample genealogy conditioned on the simulated population events.

Main Results:

• VGsim provides an efficient and flexible framework for simulating large-scale viral genealogies.
• The software can model complex population structures, epistasis, and immunity escape.
• Enables the generation of millions of samples for phylogenetic and evolutionary analyses.

Conclusions:

• VGsim is a powerful new tool for simulating viral evolution and pandemic dynamics at an unprecedented scale.
• Facilitates research into viral spread, molecular evolution, and the identification of advantageous mutations.
• Supports the development and validation of inference approaches for large genomic datasets.