OVERALL: PROJECT SUMMARY The long-term goal of the Center for Reproducible Biomedical Modeling is to achieve comprehensive predictive models of biological systems, such as whole-cell models, that can guide precision medicine and synthetic biology. In the previous cycle, we focused on model reproducibility. This culminated in the release of the reproducibility portal, biosimulators.org. This was the result of working with over 80 collaborators around the world. For cycle 2, we plan to switch from model reproducibility to model credibility and more formal technologies that deal with model decomposition and composition. Given the recent experience with the perception of poor credibility of COVID models, developing technologies to help create and measure model credibility is critical and above all timely to the biomedical community. Moreover, if biomedical models are ever to reach the clinic and make a substantial impact, model credibility will be of utmost importance. No clinician will use a predictive model without some degree of evidence and trust to substantiate the claims of a model's capabilities. Moreover, for models of national importance that can affect policy decisions affecting millions of people, the ability to automatically assess credibility of a model would seem paramount. To achieve our goals, the three technology and research cores will focus on: (1) Credibility infrastructure, credibility assessment metrics and model management and decomposition; (2) automated annotations, and continuing to curate models for the reproducibility portal; (3) Implementing model composition for building larger multiscale models that will exploit our multi-paradigm simulator farm underlying the reproducibility portal. We will buttress this technology through collaboration with an extremely wide range of collaborators and service projects. Through our outreach and dissemination efforts we will make sure we inform, train and encourage the community to move towards more systematic models approaches. Specifically. we will (1) promote the importance of credible modeling by organizing meetings and publishing perspectives; (2) train researchers to conduct modeling reproducibly and credible by organizing workshops and publishing tutorials; and (3) help researchers and journals build, annotate, simulate, analyze, and verify models; (3) Host small online competitions to encourage younger scientists to take on the baton and peruse more systematic model development and testing. The center was one of the first and still the only large-scale effort to make biomedical modeling reproducible. With the proposed cycle 2 effort, the center will be the first large-scale effort to also work on biomodel credibility. We anticipate that this unique center will accelerate the development of comprehensive predictive models by enhancing the understandability, reusability, reproducibility, and credibility of biomedical modeling.

OVERALL: PROJECT NARRATIVE Our long-term goal is to achieve comprehensive predictive models that can guide precision medicine and synthetic biology. Toward this goal, we will make models understandable, reproducible, credible and reusable by (1) developing tool for reproducibly and credibly building, simulating, and analyzing models; (2) assembling these tools into a user friendly workflow; (3) providing the community services for annotating, simulating, and validating models; (4) promoting the importance of reproducibility; and (5) training researchers to conduct modeling reproducibility. These efforts will enhance the reproducibility of modeling and, in turn, enable comprehensive models.