Workshop 6: Institute-Wide Coordination and Next Steps

Organizers: Fefferman, Blum, Bourouiba, Candan, Ryan, Strand

Topic: The purpose of this workshop was to present the intermediate work products of the ongoing research projects to the broader Institute community to get feedback and recruit additional expertise.

# Participants: 10

Report:

Each of the ongoing project teams shared their progress thus far and solicited feedback and additional collaboration from the other Institute participants. Each team also shared their first target “deliverable”:

• Team Zoonotic Pathogen Risk: To produce a publication outlining a new perspective on what “evidence” means for evaluating risk of emergent zoonoses “tipping” into outbreaks in city landscapes. This new perspective arose out of team collaboration and uses a medical cohort-control inspiration to unify strategies for discovery across disciplines.

• Team Microbes in Built Environments: To produce a multidisciplinary literature review of influences and highlight knowledge gaps, using an explanatory framework informed by a generalizable model of microbial transport and growth using an ecological metapopulation structure as a framework. This will integrate both functional bioinformatic and microbial ecology perspectives to understand when the system tips from extinction or endemicity of a pathogen into an outbreak scenario.

• Team Spruce Budworm: This team had a dual set of initial products (1) To design and deploy a Hidden Markov Model on multiple, multidisciplinary datasets and (2) To publish long time-series test datasets for use in general AI research. These efforts will provide a foundation for integrating among studies developed in distinct disciplines (economics and conservation ecology) to determine whether a multidisciplinary approach will improve identification and understanding of system tipping points.

• Team Ensemble Model Complexity: To produce a taxonomy of “integrated models” starting with multiple datasets and building all the way up to coupled model systems that can enable better communication across disciplines that employ diverse jargon and methodologies for similar (but not identical) protocols, and hence help integrate and streamline research among those fields.

• Team Disease Network Predictability: To produce an exploratory simulation of an outbreak incorporating realistic model components from each of the following fields: epidemiology, etiology, public health and medical interventions, social networks, social psychology, and individual behavior. Together, this product will serve as a “sand box” simulation, available to the broader research community to help support multidisciplinary exploration move beyond the current state of the art in which each discipline produces a nuanced model of their component, but a coarse or black-box model of components from other disciplines.