Emerging infectious diseases are one of the leading threats to human health, food security, and biodiversity. Ecologists rely on estimates of disease processes (e.g., population size, transmission rate, recovery probability, etc.) to develop risk assessments and disease mitigation strategies. In the earliest mathematical formulations to understand disease spread, Kermack & McKendrick and Anderson & May formulated a series of ordinary differential equations− commonly known as SIR models (i.e., susceptible, infected, recovered models). However, the key parameters in these host-pathogen models (e.g., transmission, contacts rates, recovery probabilities) are difficult to estimate using real-world data because the data needed for these mechanistic models are difficult or impossible to collect, and/or the data collected contain observation/sampling error. This session aims to synthesize research on novel statistical and mathematical approaches to estimate host-pathogen disease dynamics, using modeling techniques that take into account biological scales of organization (i.e., individual, population, community), nonlinear stochastic behaviors, host age-structure, and spatial variation. Speakers will present recent research including empirical and theoretical work on a variety of disease systems, including malaria, dengue, avian influenza, carnivore rabies, and other diseases that affect humans, wildlife, or both (i.e., spillover diseases; zoonosis). The speakers come from diverse backgrounds, and the uniting factor among them are their novel modeling approaches to understanding how pathogens affect individual-, population-, and community-level host ecology. There are also several conceptual themes running throughout the session, including the impact of host and spatial heterogeneity in disease dynamics, linking biological scales of organization, and using novel statistical and mathematical approaches to answering age-old questions in disease ecology.