I'm wanting to run a model that contains an ordered categorical factor (condition A > condition B > condition C) and made some adjustments to the code to allow a designMatrix option to basically overwrite the designMatrix to respect the way it would have been constructed if ordered factors were in the metadata; the columns would contain the levels of the ordered factor from 2 to 4 (.L, .Q, .C, ^4 instead of A,B,C). This is probably not the cleanest way to proceed but i think it sort of works. I'm not exactly sure how to compute the LFC from this but i can make do with the estimates for now.
Hi, thanks for this really great package!
I'm wanting to run a model that contains an ordered categorical factor (condition A > condition B > condition C) and made some adjustments to the code to allow a designMatrix option to basically overwrite the designMatrix to respect the way it would have been constructed if ordered factors were in the metadata; the columns would contain the levels of the ordered factor from 2 to 4 (.L, .Q, .C, ^4 instead of A,B,C). This is probably not the cleanest way to proceed but i think it sort of works. I'm not exactly sure how to compute the LFC from this but i can make do with the estimates for now.
Best regards, Kelvin