Predicting cellular pluripotency using contrast images
Inventors
Chen, Matthew • Schiff, Lauren • Cuevas, Alicia • Haston, Kelly • Zeng, Haoyang • SCANDORE, Cody
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12045982-B2
Publication Date
2024-07-23
Expiration Date
Abstract
Embodiments of the disclosure include methods for implementing a predictive model that predicts pluripotency of cells through a cost efficient and non-destructive means. The predictive model analyzes contrast images captured from the cells and outputs predictions of cellular pluripotency at the cellular level. Thus, implementation of the predictive model guides the selection and isolation of cells that are predicted to be pluripotent. Furthermore, the predictive model facilitates retrospective analyses to correlate pluripotency metrics with differentiation success and further enables tracking of cellular pluripotency over time (e.g., to evaluate differentiation of cells).
Core Innovation
The disclosed invention provides a non-destructive pluripotency characterization system for a plurality of cells by obtaining a contrast image of the plurality of cells and applying a predictive model to translate the contrast image into an intermediate mask representation. The predictive model includes a pluripotency model and a cell localization model, and includes one or more skip connections. The intermediate mask representation comprises pluripotency predictions of biomarker intensities for the plurality of cells and supports identification of where the cells are located in the contrast image.
After translation into the intermediate mask representation, pluripotency metrics are generated for the plurality of cells according to the intermediate mask representation and the identified locations of the cells. The pluripotency metrics are described as indicative of pluripotency, including cellular-level metrics generated from the intermediate mask using localized cell regions. In refinements, the pluripotency metrics include determining a proportion of pixels in an intermediate mask representation that indicates the cell is pluripotent at an identified location predicted by the cell localization model.
The invention also supports pluripotency predictions based on sequencing data by using an intermediate mask representation that comprises pluripotency predictions according to sequencing data, and then generating pluripotency metrics from that intermediate representation. The sequencing-derived representations are described as including RNA-seq transcriptional sequencing profiles, and additionally include DNA-seq, ATAC-seq, and epigenetic signatures including histone or DNA methylation statuses.
Claims Coverage
The partial content contains two independent method claims. Both claims center on translating contrast images into an intermediate mask representation that supports generation of pluripotency metrics for a plurality of cells, with one independent claim additionally including cell localization and pluripotency predictions tied to biomarker intensities, and the other independent claim additionally requiring that the intermediate mask representations are based on sequencing data.
Intermediate mask pluripotency prediction from contrast image using skip connections and cell localization
Obtaining a contrast image of the plurality of cells; applying a predictive model to the contrast image, the predictive model comprising a pluripotency model with one or more skip connections and a cell localization model, wherein the pluripotency model translates the contrast image to an intermediate mask representation comprising pluripotency predictions of biomarker intensities for the plurality of cells, and the cell localization model identifies locations of the plurality of cells within the contrast image.
Cell-region pluripotency metrics from intermediate mask using identified locations
Generating at least pluripotency metrics for the plurality of cells according to the intermediate mask representation and using the identified locations of the plurality of cells predicted by the cell localization model, the pluripotency metrics indicative of pluripotency of the plurality of cells.
Sequencing-data-based intermediate mask pluripotency prediction from contrast image
Obtaining a contrast image of the plurality of cells; applying a predictive model to the contrast image, the predictive model configured to translate the contrast image to an intermediate mask representation comprising pluripotency predictions for the plurality of cells, wherein the intermediate mask representation is a representation comprising pluripotency predictions according to sequencing data.
Pluripotency metrics generated from sequencing-data intermediate mask
Generating pluripotency metrics for the plurality of cells according to the intermediate mask representation, the pluripotency metrics indicative of pluripotency of the plurality of cells.
Across the two independent claims, the core claimed structure is translation of contrast images into an intermediate mask representation that supports generation of pluripotency metrics for multiple cells. One independent claim additionally introduces a cell localization model to identify cell locations and to enable location-based pluripotency metrics derived from the intermediate mask.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?