Training data generation TomoTwin was trained on 123 data classes composed of subvolumes of 120 different proteins, membranes, noise and fiducials from simulated tomograms. To ensure that TomoTwin is trained on the most diverse set of proteins possible, 108 proteins were selected from the PDB with sizes ranging from 30 kDa to 2.7 mDa and the cross-correlation between pairs of 10 Å lowpass filtered maps of each protein was calculated (Extended Data Fig. 2).
Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.
Claim your profile
Get in touch with Markus
Contact Markus, search articles and posts on X, monitor coverage, and track replies from one place.
Learn more about Muck Rack