20 Lessons in Team Science: Learning from the Experience of the International Brain Laboratory

The field of neuroscience has been transformed by technological innovations that enable the collection and anal-ysis of large scale datasets. However, these massive opportunities come with significant scientific and engineeringchallenges that require new and more collaborative scientific approaches. Team science offers researchers the op-portunity to tackle scientific questions that no lab could approach individually. However, scaling up the workflowfrom a single laboratory to a multi-laboratory collaboration presents novel challenges. The International BrainLaboratory is an experiment in building a large-scale distributed collaboration committed to open science. Thecollective work of IBL has successfully demonstrated replicability of a complex mouse behavior and Neuropixelselectrophysiological recordings, developed widely used data analysis tools and shared a massive brain-wide elec-trophysiological data set. To achieve these goals, IBL had to address problems faced by collaborations of all sizes:shared decision making, the division of labor, authorship and credit assignment, and supporting career progres-sion. We also developed methods for efficient standardization, built and scaled robust data analysis pipelines andaddressed challenges faced by open science efforts. Two major strategies we implemented to help achieve ourgoals were a shared, non-hierarchical governance structure, and the creation of (and support for) a professionalcore staff who help integrate the collaboration and maintain the IBL infrastructure. In this paper we describesome of the lessons we have learned in creating and executing the vision of the International Brain Laboratorythat we hope other scientists may find applicable to their own efforts.