DeepSecret is a tool created to help in the design of cell engineering strategies to improve the secretion capacity of Chinese Hamster Ovary (CHO) cells for the production of biologics.
DeepSecret leverages the availability of massive amounts of transcriptome data from cells with different secretory phenotypes to train a neural network that captures the internal logic between the transcription profile and the secretion capacity of the cells.
DeepSecret can be used to predict the impact on protein secretion of different cell engineering strategies, multiknock-ins/outs.
DeepSecret was used to compile a list of knock-in genetic targets to increase productivity of CHO cells.