Systems, devices, and methods for quality control of cell production using predictive tagging
AssigneeAllen Institute
InventorsGregory JOHNSON, Chawin OUNKOMOL, Forrest COLLMAN, Sharmishtaa SESHAMANI, Michael KRANDA
Life of this patent
Estimating the term…
Abstract
The present invention provides various methods for easily assessing cell quality of a cell production process, suitably using non-invasive visual methods and neural networks for generating predictive fluorescence images of cells to assess quality attributes. Also provided are systems and methods for carrying out such processes.
Claims28
A method for assessing cell quality during a cellular production process, the method comprising:
- a. obtaining a 3D microscopy image that is a transmitted light image of one or more structures of a cell during the cellular production process, wherein no fluorescence labeling is included in the 3D microscopy image;
- b. generating, with a neural network, a predicted 3D fluorescence image that includes predicted fluorescence labeling of the one or more structures of the cell;
- c. identifying a quality attribute of the cell based on the predicted 3D fluorescence image;
- and
- d. assessing the quality attribute against a comparison set for the cell as a measure of cell quality.
A method for assessing cell quality during a cellular production process, the method comprising:
- a. obtaining a 3D microscopy image that is a transmitted light image of one or more structures of a cell in real-time during the cellular production process, wherein no fluorescence labeling is included in the 3D microscopy image, and wherein the cell is not removed from the cellular production process during the obtaining of the 3D microscopy image;
- b. generating, with a neural network, a predicted 3D fluorescence image that includes predicted fluorescence labeling of the one or more structures of the cell;
- c. identifying a quality attribute of the cell based on the predicted 3D fluorescence images, wherein the quality attribute is one or more of an average or median cell size, a cell viability, a cell confluency, a cell count, a cell morphology, a cell cycle phase, a cell growth rate, a cell integrity, a membrane integrity, an organelle integrity, a mitochondrial activation level, the presence or absence of a mycoplasmal, bacterial, fungal or viral contaminant, a fertilization event, a differentiation state, a batch to batch variability, or a quantity of transgenic protein produced;
- and
- d. assessing the quality attribute against a comparison set for the cell as a measure of cell quality.
A system for assessing cell quality during a cellular production process, comprising:
- a. a bioreactor for carrying out the cellular production process;
- b. a microscope configured to obtain a 3D microscopy, transmitted light image of a structure of a cell in the bioreactor;
- c. a communication interface configured to receive the 3D microscopy, transmitted light image;
- d. a processor;
- and
- e. a non-transitory computer-readable medium communicatively coupled to the communication interface and to the processor, and storing computer-executable instructions that, when executed by the processor, causes the processor to:
- i. generate, with a neural network, a predicted 3D fluorescence image that includes predicted fluorescence labeling of the one or more structures of the cell;
- ii. identify a quality attribute of the cell based on the predicted 3D fluorescence image;
- and
- iii. assess the quality attribute against a comparison set for the cell as a measure of cell quality.
Dependencies are inferred from the claim text.