| dc.description.abstract | Positive-sense single-stranded RNA ((+)RNA) viruses pose significant challenges to public health. Notable families such as Enteroviridae, Flaviviridae, and Coronaviridae have caused epidemics (Dengue, Poliomyelitis, Zika) and pandemics (COVID-19) with recurring occurrences. Despite their diverse genomic features and infection strategies, these viruses exhibit commonalities in their cellular life cycle processes, such as replication, translation, and virus assembly, and interactions with host cell machinery. Given the public health and socio-economic impact of (+)RNA viruses, understanding their lifecycles and developing effective antiviral strategies is of utmost importance. Commonalities in cellular life cycle features motivate the search for broad-acting antivirals to combat existing (+)RNA viruses and anticipate future threats. However, exhaustive experimental investigations of all (+)RNA viruses are impractical due to technical, funding, and resource limitations. To address this challenge, computational models and tools offer valuable support in streamlining the investigation process and maximizing information extraction from limited resources.
This thesis provides an integrative and universal computational approach to designing experiments and understanding RNA virus lifecycle dynamics that can assist in elucidating the virus growth and escape strategies. It introduces three key contributions: (1) the Approximate Bayesian Computation - Fixed Acceptance Rate (ABC-FAR) algorithm, enabling automated and robust parameter estimation using global sampling-rejection criterion, (2) the PARameter SEnsitivity driven Clustering based DoE (PARSEC) algorithm, facilitating Model-Based Design of Experiments (MB-DoE) by identifying informative measurements and handling parameter uncertainty, and (3) a generalized dynamical model capturing common aspects of (+)RNA virus lifecycles, incorporating compartment formation kinetics, and enabling the identification of virus-specific and broad-acting life cycle bottlenecks. Application of stochastic frameworks to study cellular infectivity, the probability of the establishment of productive infection upon viral entry, reveals effective interventions targeting early life cycle properties to limit cellular infectivity. Synergy among these interventions in limiting (+)RNA virus infection as predicted by our model suggests new avenues for inhibiting infections by targeting early life cycle bottlenecks. Finally, the model is extended to simulate a two-strain co-infection system to study how the interactions among these strains influence cellular infectivity and virus growth dynamics.
Overall, the developed viral dynamics model and PARSEC methodology provide a quantitative platform for the analysis of viruses and dynamics of related biological systems. The findings contribute valuable insights into combating (+)RNA viruses and provide a pathway for future advancements in antiviral development and preparedness for potential novel threats. | en_US |
