Raw data series filter #28
Description
Hello, very interesting work. I have some questions.
How do you select the appropriate array data series to construct the pathway response model? Even the same drug can have very different effects on different cells. I followed your approach to find WNT-related data series and constructed a new model. After building the WNT model, I scored the scRNA-seq of embryonic development by "decoupleR" and found that the pathway activity changes were completely opposite to your model. My advisor told me that your model is correct because the WNT signal gradually decreases during embryonic cell specialization. I am very confused about this. I know that one of your filtering criteria is perturbation data within 48 hours. How did you handle the analysis details and how do you perform quality control? Did you have any additional processing to adjust the model after building it?