Abstract: A machine-learning (ML) model may be trained to receive, as input, epigenetic data associated with a subject and to output a continuous value and/or a classification of a biochemical state and/or medical condition associated with a subject. For example, the biochemical state and/or medical condition may comprise an indication that the subject consumes alcohol and/or nicotine and/or that the subject is diabetic or is likely to become diabetic, to give a small non-limiting example. The epigenetic data may be derived from saliva and/or blood in some examples.

Abstract: A machine learning (ML) architecture may be trained to determine an estimated epigenetic status at a target DNA locus based at least in part on epigenetic data associated with one or more other DNA loci. The ML architecture may additionally or alternatively be used to detect a failed epigenetic assay and/or determine a likelihood that a subject has falsified information provided about the subject.

Abstract: High throughput quality control for epigenetic profiles associated with different subjects may include a pipeline that identifies as faulty epigenetic profile(s) from among a batch of epigenetic profiles and may additionally or alternatively includes a machine-learning (ML) model trained to identify the type of fault or condition that caused an epigenetic profile to be faulty. Such an ML model and techniques for training such an ML model are discussed herein.