integrate/totalvi

Description

Performs mapping to the reference by totalvi model: https://docs.scvi-tools.org/en/stable/tutorials/notebooks/scarches_scvi_tools.html#Reference-mapping-with-TOTALVI

Type

python_script

Contributors

Launch

Run this component

Run the following command to execute this component with Nextflow:

cat > params.yaml <<'EOM'  
query_modality: [ "rna" ]  
reference_modality: [ "rna" ]  
reference_proteins_modality: [ "prot" ]  
obs_batch: [ "sample_id" ]  
output: "$id.$key.output"  
obsm_output: [ "X_integrated_totalvi" ]  
obsm_normalized_rna_output: [ "X_totalvi_normalized_rna" ]  
obsm_normalized_protein_output: [ "X_totalvi_normalized_protein" ]  
reference_model_path: "$id.$key.reference_model_path"  
query_model_path: "$id.$key.query_model_path"  
max_epochs: [ 400 ]  
max_query_epochs: [ 200 ]  
weight_decay: [ 0 ]  
id: "run"  
publish_dir: "output/"  
EOM

nextflow run https://packages.viash-hub.com/vsh/openpipeline.git \  
  -revision 1.0.1 \  
  -main-script target/nextflow/integrate/totalvi/main.nf \  
  -params-file params.yaml

Inputs

Name	Type & Properties	Description
--input -i	file required	Input h5mu file with query data to integrate with reference.
--reference -r	file required	Input h5mu file with reference data to train the TOTALVI model.
--force_retrain -f	boolean_true	If true, retrain the model and save it to reference_model_path
--query_modality	string
--query_proteins_modality	string	Name of the modality in the input (query) h5mu file containing protein data
--reference_modality	string
--reference_proteins_modality	string	Name of the modality containing proteins in the reference
--input_layer	string	Input layer to use. If None, X is used
--obs_batch	string	Column name discriminating between your batches.
--var_input	string	.var column containing highly variable genes. By default, do not subset genes.

Name

Type & Properties

--input

-i

file

required

--reference

-r

file

required

--force_retrain

-f

boolean_true

--query_modality

string

--query_proteins_modality

string

--reference_modality

string

--reference_proteins_modality

string

--input_layer

string

--obs_batch

string

--var_input

string

Outputs

Name	Type & Properties	Description
--output -o	file required output	Output h5mu file.
--obsm_output	string	In which .obsm slot to store the resulting integrated embedding.
--obsm_normalized_rna_output	string	In which .obsm slot to store the normalized RNA from TOTALVI.
--obsm_normalized_protein_output	string	In which .obsm slot to store the normalized protein data from TOTALVI.
--reference_model_path	file output	Directory with the reference model. If not exists, trained model will be saved there
--query_model_path	file output	Directory, where the query model will be saved

Name

Type & Properties

--output

-o

file

required

output

--obsm_output

string

--obsm_normalized_rna_output

string

--obsm_normalized_protein_output

string

--reference_model_path

file

output

--query_model_path

file

output

Learning parameters

Name	Type & Properties	Description
--max_epochs	integer	Number of passes through the dataset
--max_query_epochs	integer	Number of passes through the dataset, when fine-tuning model for query
--weight_decay	double	Weight decay, when fine-tuning model for query

Name

Type & Properties

--max_epochs

integer

--max_query_epochs

integer

--weight_decay

double