Asimov: A framework for coordinating parameter estimation workflows

Since the first detection in 2015 of gravitational waves from compact binary coalescence (B. P. Abbott & others, 2016a), improvements to the Advanced LIGO and Advanced Virgo detectors have expanded our view into the universe for these signals. Searches of the latest observing run (O3) have increased the number of detected signals to 90, at a rate of approximately 1 per week (The LIGO Scientific Collaboration, the Virgo Collaboration, Abbott, et al., 2021; The LIGO Scientific Collaboration, the Virgo Collaboration, the KAGRA Collaboration, et al., 2021). Future observing runs are expected to increase this even further(B. P. Abbott & others, 2020). Bayesian analysis of the signals can reveal the properties of the coalescing black holes and neutron stars by comparing predicted waveforms to the observed data (B. P. Abbott & others, 2016b). The proliferating number of detected signals, the increasing number of methods that have been deployed (Ashton & others, 2019; Lange et al., 2018; Veitch & others, 2015), and the variety of waveform models (Khan et al., 2020; Ossokine & others, 2020; Pratten & others, 2021) create an ever-expanding number of analyses that can be considered. Asimov is a python package which is designed to simplify and standardise the process of configuring these analyses for a large number of events. It has already been used in developing analyses in three major gravitational wave catalog publications (R. Abbott & others, 2021; The LIGO Scientific Collaboration, the Virgo Collaboration, Abbott, et al., 2021; The LIGO Scientific Collaboration, the Virgo Collaboration, the KAGRA Collaboration, et al., 2021). The source code of Asimov is archived to Zenodo (Williams et al., 2021).