SC2Spa.sva.PrioritizeLPG
- SC2Spa.sva.PrioritizeLPG(adata, Model, sparse=True, polar=True, CT=None, CT_field='MCT', percent=0.5, scale_factor=1000.0, Norm=False)
Prioritize genes’ contribution to location prediction
Run tl.Self_Mapping first to set Norm as True
Importance scores will be saved in adata.obs
Parameters
- adata
Reference anndata object. Gene exprestlon matrix should be the shape of (cell, gene). Spatial information should be stored in adata.obsm[‘spatial’]
- Model
A neural network trained utlng adata.X and adata.obsm[‘spatial’]
- sparse
if gene exprestlon is saved in sparse format
- CT
- The cell type name used to normalize the importance score.
it must be one category in adata.obs[CT_field] This parameter is for normalization
- polar
Transform cartetlan coordinates to polar coordinates if True. This parameter is for normalization
- percent
The percentage of nodes selected for each layer
- scale_factor
The factor used to scale the importance scores
Returns