Protiens are the end effectors of all genetic information. In order to properly understand the attack method of COVID19 we must understand the properties of its protiens. Much of the expirimental data on COVID19 protiens can by found in the protien data base (PDB). For example The main protease of COVID19 can be found (in complex with inhibitor N3) available for download at https://www.rcsb.org/3d-view/6LU7.

Protiens have been the subject of much research even before the the COVID19 outbreak. As such several very developed algorithms have been deployed against COVID19 protiens. Neural Models of protien binding have been used to identify preaproved drugs with the potential to inhibit COVID19 operation. https://www.biorxiv.org/content/10.1101/2020.01.31.929547v1.full

Citizen science initiatives like (https://fold.it/) can use the 3D structure of active COVID19 protiens provided by labratory expirements to develop inhibitor protiens that selectively bind to the active regions of COVID19 protiens, preventing or slowing the effect of the virus.

Most active COVID19 protiens have been images with cryoelectromicroscopy, but some are difficult to issolate, or triaged as low priority. Attempts have been made with purely machine learning models to predict the structure of these protiens from raw sequence information. https://deepmind.com/research/open-source/computational-predictions-of-protein-structures-associated-with-COVID-19 The technique used has been verified against the expiremental results on the COVID19 spike protien.