Hello Dear organizers,
1) Will the test/validation dataset have the same file structure as the provided .h5ad training data (including metadata columns such as Age or APOE.Genotype minus ADNC scores and other ranking metrics)? Or will it only have expression ~36k gene features? 2) The test/validation set also contain data from other parts of brain than MTG or A9, right?
Thanks,
Tushar
Created by TUSHAR PATEL tpatel The test and validation datasets may be from different cortical areas and/or have different donors. The areas will have the same cellular taxonomy (e.g. values in "Class", "Subclass", and "Supertype") and have the same features/genes. The actual AnnData file will be formatted identically, except the relevant variables in .obs that are being predicted will be removed.