The popular story of CRISPR IP is a two-party fight over who invented the core system. The patent record from a single year - 2020 - tells a more crowded story, and that crowding is the strategic fact that matters for anyone trying to clear freedom-to-operate.
Start with three grants issued that year. US10563226B2 (Seattle Children's Hospital, February 2020) claims enhancing endonuclease-based gene editing specifically in primary cells - a use-context limitation. US10738290B2 (Novartis AG, August 2020) claims an RNA-guided gene-editing system as machinery. US10704060B2 (Duke University, July 2020) reaches both editing and gene regulation. Three assignees, three different angles on overlapping technology, all live in the same year.
Aggregate by assignee before drawing any conclusion, which is the discipline this kind of map demands. The 2020 facet data on broad CRISPR queries shows the editing space dominated not by therapeutics names but by agricultural and tooling players - Pioneer Hi-Bred, Inscripta, Agrigenetics - alongside the academic anchors (Regents of the University of California, Harvard, the General Hospital Corporation). A therapeutics developer is operating inside a field whose foundational layer was filed largely by others.
The CPC clustering reinforces it: C12N 9/22 (the Cas endonuclease class), C12N 15/907, C12N 2310/20 (guide-RNA targeting) recur across assignees. When the same CPC subclasses light up across unrelated owners, that is the signature of a thicket - many narrow, overlapping claims - rather than a moat, where one owner controls a defensible core.
The strategic implication for 2020-era programs was that a license to one foundational estate was never going to be sufficient. The editing machinery, the delivery context, and the cell-type use limitations were each separately claimed, often by different owners. That fragmentation is why the licensing stacks on later gene-editing therapies became so layered - the groundwork was laid in grants like these.