Step 2: latent_to_gene¶
usage: gsmap run_latent_to_gene [-h] --workdir WORKDIR --sample_name
SAMPLE_NAME
[--input_hdf5_path INPUT_HDF5_PATH]
[--no_expression_fraction]
[--latent_representation LATENT_REPRESENTATION]
[--num_neighbour NUM_NEIGHBOUR]
[--num_neighbour_spatial NUM_NEIGHBOUR_SPATIAL]
[--homolog_file HOMOLOG_FILE]
[--gM_slices GM_SLICES]
[--annotation ANNOTATION]
Named Arguments¶
- --workdir
Path to the working directory.
- --sample_name
Name of the sample.
- --input_hdf5_path
Path to the input HDF5 file with latent representations, if –latent_representation is specified.
- --no_expression_fraction
Skip expression fraction filtering.
Default:
False
- --latent_representation
Type of latent representation. This should exist in the h5ad obsm.
- --num_neighbour
Number of neighbors.
Default:
21
- --num_neighbour_spatial
Number of spatial neighbors.
Default:
101
- --homolog_file
Path to homologous gene conversion file (optional).
- --gM_slices
Path to the slice mean file (optional).
- --annotation
Name of the annotation in adata.obs to use (optional).