Hello !
I have a very basic question on GIMVI model. This model is designed to “input missing genes on spatial data”, based on a single-cell reference. However, the tutorial resitricts the single-cell reference to genes present in the spatialdata:
seq_data = seq_data[:, spatial_data.var_names].copy()
Thus, the predicted genes are only genes already present in the spatialdata, which does not make sense. Did I miss some point on GIMVI model ?
Thank you very much,
Alexis
Hi,
This line in the tutorial is not mandatory at all. You can still impute other genes.
However, for the sake of the tutorial there is no real reason to impute another 2000 gene which cant be compared to ground truth.
But you can definitely do it in your own analysis.