The Odor2Action network consists of 16 investigators from 16 research institutions in the United States, the United Kingdom, and Canada. The composition and scientific goals of the effort are designed to leverage prior investments in neurotechnologies funded by the BRAIN Initiative, other domestic agencies and international partners. The Odor2Action network will address a central question of neuroscience: How do animals use information from odor stimuli in their environment to guide natural behaviors? ​We approach this problem in the context of olfactory-guided behavior as an instance of a much more general problem of many complex brain systems: How are high-dimensional, discrete, and combinatorial variables that are not simply ordered along easily discernible axes represented in brain circuits and mapped to actions?

The compact olfactory circuit architecture offers unique opportunities to achieve an end-to-end understanding of the core computational logic by which brains organize and read out high-dimensional, discrete variables to generate adaptive behaviors. We will study olfactory systems of mammals and insects, which have independently evolved common structural elements at successive levels of olfactory processing in their central nervous systems. These common elements possibly reflect convergent evolution towards a set of similar solutions to shared olfactory problems.

Key research resources being developed and disseminated

To answer these questions, the Odor2Action network has created three interdisciplinary research groups (IRGs) that are designed around specific elements of an end-to-end investigation of olfaction. IRG1 aims to understand the first stages of how neural representations of odor are generated, and how they are progressively reformatted across successive circuit layers to support meaningful behaviors. IRG2 aims to understand how neural circuits translate odor signals into dynamic and adaptive behaviors, a critical component of our overall network goal of understanding how natural odors trigger natural behaviors. IRG3 will investigate the physical structure of odor environments and how animal motion and sensory capabilities interact with those environments to detect, discriminate and localize odor objects. Each IRG integrates theory and experimental approaches in two or more species in ways that produce complementary, synergistic interactions across levels of biological analysis. Collectively, the network will determine how neural representations of odor are generated, how they are progressively reformatted across successive circuit layers, and how they support useful behaviors.

Achieving our goals will require application and development of new technologies in measurement and control of stimuli, quantification of behavior, and recording of neural activity and circuit tracing. We envision the network as a mechanism for creating and disseminating expertise in these areas both within and outside of the network. The Odor2Action network plans to develop a next-generation platform for sharing and analyzing data within the team and across the scientific community. We have designed a systematic, centralized, and expertly supported approach to meeting data management and cyberinfrastructure needs, including a PetaLibrary storage system, multiple hardware and cloud-based computing platforms, and a shared computing portal that utilizes the Jupyter web-based interactive development environment.

The Odor2Action network sees the integration of research and education/training as essential to the success of the overall project. We aim to train a cohort of students to find working on large-scale, team-science projects to be a natural and expected part of their careers.