Transcription is a complex process that occurs in all organs and can be used to identify different cell types. In the liver, current hPSC-derived protocols are unable to achieve sufficient maturity for transplantation so better protocols need to be developed as well as a better way to compare various protocols. Comparing various cellular engineering protocols requires a method to access the extent of differentiation. There are currently several methods to classify maturity of engineered cells. These methods typically score maturity by using a weighted system based on transcriptomic data from adult tissue samples. A combination of known transcription factors (TF) within an adult cell type’s gene regulatory network (GRN) in conjunction with genes associated with differentiation are used to score maturity. However, these current protocols fail to show how incorrect expression of specific TFs leads to dysregulation of other TFs and further downstream differentiation genes. These software packages also don’t consider developmental steps and TF dynamics during differentiation. Many TFs do not increase linearly during development but instead utilize complex fluctuations to achieve maturity. Additionally, cellular processes including migration, signaling pathways, metabolic pathways have been found to be imperative for correct cell differentiation and should be further studied to help classify maturity.
Instead of quantifying how close engineered cell populations resemble target adult cells, we propose instead identifying the developmental cell that most closely resembles the engineered cell population. Therefore, our work proposes a novel method to classify maturity and identify dysregulation of specific TFs in addition to the global effect of these TFs on differentiation and on specific genetic pathways during development. Unlike current methods, we utilize developmental transcriptomic data as well as TF binding site data to identify important TFs and downstream target genes for specific steps during development and to accurately classify cells. We used this method to compare derived hepatic cells but this work has potential to be further expanded for other cell types as well. The use of this more complex scoring system should help guide better cellular engineering protocols.
