US 12,073,638 B1
Utilizing machine learning and digital embedding processes to generate digital maps of biology and user interfaces for evaluating map efficacy
Nathan Henry Lazar, Salt Lake City, UT (US); Conor Austin Forsman Tillinghast, Salt Lake City, UT (US); James Douglas Jensen, Farmington, UT (US); James Benjamin Taylor, Midlothian, VA (US); Berton Allen Earnshaw, Cedar Hills, UT (US); Marta Marie Fay, Salt Lake City, UT (US); Renat Nailevich Khaliullin, Salt Lake City, UT (US); Jacob Carter Cooper, Sandy, UT (US); Imran Saeedul Haque, Salt Lake City, UT (US); Seyhmus Guler, Salt Lake City, UT (US); Kyle Rollins Hansen, Kaysville, UT (US); and Safiye Celik, Sudbury, MA (US)
Assigned to Recursion Pharmaceuticals, Inc., Salt Lake City, UT (US)
Filed by Recursion Pharmaceuticals, Inc., Salt Lake City, UT (US)
Filed on Dec. 21, 2023, as Appl. No. 18/392,989.
Claims priority of provisional application 63/582,702, filed on Sep. 14, 2023.
Int. Cl. G06V 20/69 (2022.01); G06T 7/00 (2017.01); G06T 7/35 (2017.01); G06V 10/74 (2022.01); G06V 10/82 (2022.01); G16B 40/20 (2019.01)
CPC G06V 20/698 (2022.01) [G06T 7/0012 (2013.01); G06T 7/35 (2017.01); G06V 10/761 (2022.01); G06V 10/82 (2022.01); G16B 40/20 (2019.02); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30072 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
receiving perturbation data for a plurality of perturbation experiment units;
generating, utilizing a machine learning model, a plurality of perturbation experiment unit embeddings from the perturbation data;
aligning, utilizing an alignment model, the plurality of perturbation experiment unit embeddings to generate aligned perturbation unit embeddings by aligning a set of perturbation experiment unit embeddings of a single perturbation class from a plurality of different perturbation experiments according to a statistical alignment model;
aggregating the aligned perturbation unit embeddings to generate aggregated embeddings; and
generating perturbation comparisons utilizing the aggregated embeddings.