US 11,795,495 B1
Machine learned epigenetic status estimator
Randal S. Olson, Minneapolis, MN (US); and Brian H. Chen, Minneapolis, MN (US)
Assigned to FOXO Labs Inc., Minneapolis, MN (US)
Filed by Life Epigenetics, Inc., Minneapolis, MN (US)
Filed on Oct. 2, 2019, as Appl. No. 16/591,296.
Int. Cl. C12Q 1/6825 (2018.01); G16H 10/60 (2018.01); G16H 50/50 (2018.01); G16B 5/20 (2019.01); G16H 50/20 (2018.01); G16H 50/70 (2018.01); G16B 40/10 (2019.01); G16H 50/30 (2018.01); G01N 33/53 (2006.01); G16B 20/00 (2019.01); G16B 40/30 (2019.01); G16B 40/00 (2019.01); C12Q 1/6869 (2018.01); G16B 40/20 (2019.01); C12Q 1/6876 (2018.01)
CPC C12Q 1/6825 (2013.01) [C12Q 1/6869 (2013.01); C12Q 1/6876 (2013.01); G01N 33/53 (2013.01); G16B 5/20 (2019.02); G16B 20/00 (2019.02); G16B 40/00 (2019.02); G16B 40/10 (2019.02); G16B 40/20 (2019.02); G16B 40/30 (2019.02); G16H 10/60 (2018.01); G16H 50/20 (2018.01); G16H 50/30 (2018.01); G16H 50/50 (2018.01); G16H 50/70 (2018.01)] 20 Claims
OG exemplary drawing
1. A method comprising:
training a first machine-learning (ML) model to estimate an epigenetic value associated with a subject, wherein the training comprises altering one or more parameters of the first ML model based at least in part on a difference between an output of the first ML model and a ground truth epigenetic value to obtain a trained first ML model;
receiving a set of epigenetic data associated with a subject, wherein the set of epigenetic data is determined by an assay configured with a plurality of probes to determine an epigenetic status at each locus of a set of deoxyribonucleic acid (DNA) loci of a biological sample;
receiving an indication of a target DNA locus associated with the biological sample;
determining that a probe associated with the target DNA locus failed or that at least one of the probe or the target DNA locus is unavailable on the assay;
determining, based at least in part on the target DNA locus and determining that the probe failed or is unavailable, a subset of the set of epigenetic data; and
determining, by the trained first ML model and based at least in part on the subset of epigenetic data, an estimated epigenetic value associated with the target DNA locus, wherein the estimated epigenetic value comprises an estimated probe intensity, an estimated probe intensity ratio, an estimated beta-value, or an estimated M-value associated with the target DNA locus.