I noticed that Pando utilizes regression models like GLM to fit gene expression based on TF expression and chromatin accessibilities without dividing the data into training and test sets. Can I infer that in this scenario, the model is prone to overfitting? Does this strategy provide any benefits for gene expression regulation?
Hi,
I noticed that Pando utilizes regression models like GLM to fit gene expression based on TF expression and chromatin accessibilities without dividing the data into training and test sets. Can I infer that in this scenario, the model is prone to overfitting? Does this strategy provide any benefits for gene expression regulation?