InVitae has filed a patent for a system, method, and computer program product to determine the phenotypic impacts of molecular variants in a biological sample. The patent involves receiving molecular variants, determining molecular scores, signals, and population signals, deriving evidence scores, and determining phenotypic impacts based on functional scores or evidence scores. GlobalData’s report on InVitae gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on InVitae, AI-assisted genome sequencing was a key innovation area identified from patents. InVitae's grant share as of June 2023 was 1%. Grant share is based on the ratio of number of grants to total number of patents.
Determining phenotypic impacts of molecular variants within a biological sample
A recently filed patent (Publication Number: US20230187016A1) describes a method and system for determining the phenotypic impact of a target molecular variant. The method involves receiving a plurality of samples, each containing a variant in a gene. The molecular variants are divided into two groups: a Truth Set comprising variants with known phenotypic impacts, and a Target Set comprising variants with unknown phenotypic impacts, including the target molecular variant.
The method then involves training a machine learning model using the known association between the molecular variants in the Truth Set and their known phenotypic impacts. This known association is based on a functional assay that generates molecular measurements or derivatives of the measurements for each variant in the Truth Set. The machine learning model is trained using these dependent features.
Once the machine learning model is trained, it can be used to determine the phenotypic impact of the target molecular variant. The method can be applied to various types of samples, including single cells, cellular compartments, subcellular compartments, or synthetic compartments. The molecular variants can include coding or non-coding variants within mutational hotspots of functional elements, genes, and pathways associated with clinically valuable genes, Mendelian disorders, known cancer drivers, or variation in drug response.
The molecular measurement used in the method can include locus-specific measurements of gene expression, protein expression, chromatin accessibility, epigenetic modification, regulatory activity, post-transcriptional processing, post-translational modification, mutation status, mutation burden, or mutation rate of molecules within each sample. The machine learning model can be a supervised learning model, and the derivative of the molecular measurement can be generated using a plurality of Artificial Neural Networks (ANNs).
The method and system described in the patent can be used to inform a test subject's lifetime risk of developing cancer if they have the target molecular variant. It can also be used to identify significantly mutated regions and networks by analyzing phenotype-associated mutation density.
In summary, the patent describes a method and system for determining the phenotypic impact of a target molecular variant using machine learning and functional assays. The method can be applied to various types of samples and can provide valuable insights into the pathogenicity, functionality, and relative effect of molecular variants. It has potential applications in cancer risk assessment and identifying mutated regions and networks.