@Mette In order to gather matched control (healthy) samples, I am using the following criteria on meta data features: ????((meta_clinical.cogdx==1)&(meta_clinical.dcfdx_lv==1))& #Cognitive impairment scores (meta_clinical.age_at_visit_max>80) & ((meta_clinical.braaksc==0)|(meta_clinical.braaksc.isna())) & #Braak Score for Alzheimer's ((meta_clinical.ceradsc==4)|(meta_clinical.ceradsc.isna())) & #CERAD score for Alzheimer's I am wondering if for samples with no cognitive impairment, if the Braak and Cerad scores are NA, can we assume that these controls would indeed be healthy because it wasn't deemed important enough for these patients to have to go through these tests? If we make the criteria stringent so as to have the best possible value for these scores for Controls (like Braak 0 and CERAD 4), we will have no control samples.

Created by Arpita Joshi ajoshi
Hello @ajoshi, I see that you added this discussion post a while ago and I wanted to see if this is still an open question of if this was resolved outside of the discussion forum. My sincere apologies if your question was missed and never resolved! Please let us know if you still have questions or need support. Best, Jessica

Choosing control samples/patients for differential expression analysis page is loading…