Platforms and systems for automated cell culture
Disclosed herein are platforms, systems, and methods including a cell culture system that includes a cell culture container comprising a cell culture, the cell culture receiving input cells, a cell imaging subsystem configured to acquire images of the cell culture, a computing subsystem configured to perform a cell culture process on the cell culture according to the images acquired by the cell imaging subsystem, and a cell editing subsystem configured to edit the cell culture to produce output cell products according to the cell culture process.
1. A method for producing an induced pluripotent stem cell (iPSC) product, comprising:
(a) culturing a plurality of iPSC candidate cells in a closed cell culture container, wherein a plurality of iPSC candidate cell colonies emerge from the plurality of iPSC candidate cells;
(b) acquiring, using an image sensor, time-series image data of the plurality of iPSC candidate cell colonies;
(c) processing, by a computer processor, the acquired time-series image data using one or more trained machine learning models to predict clonal quality of the plurality of iPSC candidate cell colonies;
(d) managing, by a cell editing subsystem, the plurality of iPSC candidate cell colonies based at least in part on the acquired time-series image data, wherein managing the plurality of iPSC candidate cell colonies comprises performing selective removal of at least a portion of an iPSC candidate cell colony of the plurality of iPSC candidate colonies based at least in part on whether the iPSC candidate cell colony will collide with another iPSC candidate cell colony of the plurality of iPSC candidate cell colonies;
(e) selecting, by the computer processor, at least one of the plurality of iPSC candidate cell colonies for expansion based at least in part on the predicted clonal quality of the plurality of iPSC candidate cell colonies;
(f) removing, by the cell editing subsystem, non-selected iPSC candidate cell colonies of the plurality of iPSC candidate cell colonies from the closed cell culture container; and
(g) expanding the selected at least one of the iPSC candidate cell colonies into the iPSC product.
2. The method of claim 1 , further comprising further training the one or more trained machine learning models using iPSC assay data.
3. The method of claim 2 , wherein the iPSC assay data is acquired from a previously expanded iPSC product.
4. The method of claim 1 , further comprising performing one or more additional iterations of (b) to (f).
5. The method of claim 1 , wherein processing the acquired time-series image data using the one or more trained machine learning models to predict the clonal quality of the plurality of iPSC candidate cell colonies further comprises identifying one or more phenotypic features of the plurality of iPSC candidate cell colonies.
6. The method of claim 1 , wherein the closed cell culture container comprises a cell culture chamber having a first semi-transparent surface, the first semi-transparent surface comprising a laser-activated film.
7. The method of claim 6 , wherein the plurality of iPSC candidate cells are adhered to the first semi-transparent surface.
8. The method of claim 6 , wherein the laser-activated film transmits incident light in a first range of wavelengths and absorbs incident light in a second range of wavelengths.
9. The method of claim 8 , further comprising imaging cells within the cell culture chamber using incident light in the first range of wavelengths, and editing cells within the cell culture chamber using incident light in the second range of wavelengths.
10. The method of claim 1 , wherein the iPSC product is suitable for differentiation into a target cell type.
11. The method of claim 1 , wherein the selected at least one of the iPSC candidate cell colonies are clonal cell colonies, and wherein the iPSC product is a monoclonal iPSC product.
12. A system for producing an induced pluripotent stem cell (iPSC) product, comprising:
a closed cell culture container configured to culture a plurality of iPSC candidate cells, wherein a plurality of iPSC candidate cell colonies emerge from the plurality of iPSC candidate cells;
an image sensor configured to acquire time-series image data of the plurality of iPSC candidate cell colonies;
a computer processor programmed to:
process the acquired time-series image data using one or more trained machine learning models to predict clonal quality of the plurality of iPSC candidate cell colonies; and
select at least one of the plurality of iPSC candidate cell colonies for expansion based at least in part on the predicted clonal quality of the plurality of iPSC candidate cell colonies; and
a cell editing subsystem configured to:
manage the plurality of iPSC candidate cell colonies based at least in part on the acquired time-series image data, wherein managing the plurality of iPSC candidate cell colonies comprises performing selective removal of at least a portion of an iPSC candidate cell colony of the plurality of iPSC candidate colonies based at least in part on whether the iPSC candidate cell colony will collide with another iPSC candidate cell colony of the plurality of iPSC candidate cell colonies; and
remove non-selected iPSC candidate cell colonies of the plurality of iPSC candidate cell colonies from the closed cell culture container, wherein the selected at least one of the iPSC candidate cell colonies expand into the iPSC product.
13. The system of claim 12 , wherein the one or more trained machine learning models are further trained based on iPSC assay data.
14. The system of claim 12 , wherein processing the acquired time-series image data using the one or more trained machine learning models to predict the clonal quality of the plurality of iPSC candidate cell colonies further comprises identifying one or more phenotypic features of the plurality of iPSC candidate cell colonies.
15. The system of claim 12 , wherein the closed cell culture container comprises a cell culture chamber having a first semi-transparent surface, the first semi-transparent surface comprising a laser-activated film.
16. The system of claim 15 , wherein the plurality of iPSC candidate cells are adhered to the first semi-transparent surface.
17. The system of claim 15 , wherein the laser-activated film is configured to transmit incident light in a first range of wavelengths and absorb incident light in a second range of wavelengths.
18. The system of claim 17 , wherein the first range of wavelengths enables imaging cells within the cell culture chamber, and wherein the second range of wavelengths enables editing cells within the cell culture chamber.
19. The system of claim 12 , wherein the iPSC product is suitable for differentiation into a target cell type.
20. The system of claim 12 , wherein the selected at least one of the iPSC candidate cell colonies are clonal cell colonies, and wherein the iPSC product is a monoclonal iPSC product.
21. The method of claim 1 , wherein managing the plurality of iPSC candidate cell colonies further comprises at least one of fragmenting a first iPSC candidate cell colony into a plurality of sub-colonies, and removing at least a portion of the first iPSC candidate cell colony to reduce its confluence or density.
22. The method of claim 21 , wherein managing the plurality of iPSC candidate cell colonies further comprises at least one of:
(i) removing the first iPSC candidate cell colony based at least in part on whether the first iPSC candidate cell colony is growing outside of a designated growth region of the closed cell culture container,
(ii) removing non-induced pluripotent stem cells from the closed cell culture container, and
(iii) removing neighboring cells in a zone around the first iPSC candidate cell colony.
23. The method of claim 1 , wherein processing the acquired time-series image data using the one or more trained machine learning models to predict clonal quality of the plurality of iPSC candidate cell colonies further comprises generating an outcome score for each of the iPSC candidate cell colonies that represents a likelihood that each iPSC candidate cell colony produces a desired output cell product.
24. The system of claim 12 , wherein managing the plurality of iPSC candidate cell colonies further comprises at least one of fragmenting a first iPSC candidate cell colony into a plurality of sub-colonies, and removing at least a portion of the first iPSC candidate cell colony to reduce its confluence or density.
25. The system of claim 24 , wherein managing the plurality of iPSC candidate cell colonies further comprises at least one of:
(i) removing the first iPSC candidate cell colony based at least in part on whether the first iPSC candidate cell colony is growing outside of a designated growth region of the closed cell culture container,
(ii) removing non-induced pluripotent stem cells from the closed cell culture container, and
(iii) removing neighboring cells in a zone around the first iPSC candidate cell colony.
26. The system of claim 12 , wherein processing the acquired time-series image data using the one or more trained machine learning models to predict clonal quality of the plurality of iPSC candidate cell colonies further comprises generating an outcome score for each of the iPSC candidate cell colonies that represents a likelihood that each iPSC candidate cell colony produces a desired output cell product.
27. The method of claim 1 , wherein the selective removal of the cells in the plurality of iPSC candidate cell colonies comprises lysing the cells.
28. The system of claim 12 , wherein the selective removal of the cells in the plurality of iPSC candidate cell colonies comprises lysing the cells.