(ii) Particle selection. We used two particle-picking algorithms: Xmipp (Abrishami et al., 2013 ▸) and crYOLO (Wagner et al., 2019 ▸). We submitted both results to a picking consensus algorithm using deep learning (Sanchez-Garcia et al., 2018 ▸) and also removed all of the coordinates in contaminations, carbon edges etc. using a deep-learning algorithm (Sanchez-Garcia, Segura et al., 2020 ▸). We then cleaned the set of selected particles using two rounds of cryoSPARC 2D classification (Punjani et al., 2017 ▸; Punjani & Fleet, 2020 ▸) and the consensus of two independent 3D classifications with cryo­SPARC.