US 12,254,958 B2
Methods and compositions for governing phenotypic outcomes in plants
Bradley Michael Zamft, Mountain View, CA (US); and Logan Graham, Mountain View, CA (US)
Assigned to HERITABLE AGRICULTURE INC., Mountain View, CA (US)
Filed by HERITABLE AGRICULTURE INC., Mountain View, CA (US)
Filed on Jan. 12, 2024, as Appl. No. 18/412,306.
Application 18/412,306 is a division of application No. 16/870,838, filed on May 8, 2020, granted, now 11,908,547.
Claims priority of provisional application 62/845,276, filed on May 8, 2019.
Prior Publication US 2024/0221865 A1, Jul. 4, 2024
Int. Cl. G06F 17/00 (2019.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01); G16B 20/00 (2019.01); G16B 40/00 (2019.01); G06Q 50/02 (2012.01)
CPC G16B 20/00 (2019.02) [G06N 5/04 (2013.01); G06N 20/00 (2019.01); G16B 40/00 (2019.02); G06Q 50/02 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
obtaining a model input comprising at least one of i) a desired multi-omics profile of a plant, or ii) a desired phenotype of the plant;
processing the model input using a generative machine learning model to obtain a model output comprising one or more of:
a target multi-omics profile,
a target management practice profile, or
a target environment profile,
wherein the generative machine learning model has been configured through training to:
receive a training model input that identifies desired qualities of a training plant, and
process the training model input to generate one or more of i) a target multi-omics profile, ii) a target management practice profile, or iii) a target environment profile that, when used to modify the training plant, causes the training plant to exhibit the desired qualities; and
determining that one or more interventions should be performed to modify the plant according to the model output.