wrappers/preprocessing/highly_variable_genes

Description

Annotate highly variable genes, running either the GPU (rapids-singlecell) or
the CPU (scanpy) variant of highly_variable_genes, selected with --device_type.

Expects log-normalized data for the default 'seurat' flavor. The GPU variant
exposes additional flavors ('seurat_v3_paper', 'pearson_residuals',
'poisson_gene_selection') that the CPU variant does not support.

Inputs

Name	Type & Properties	Description
--input	file required	Input h5mu file.
--modality	string	Which modality from the input MuData file to process.
--input_layer	string	Input layer to use for expression values. By default, X is used.

Name

Type & Properties

--input

file

required

--modality

string

--input_layer

string

Compute

Name	Type & Properties	Description
--device_type	string	Which implementation to run: the GPU (rapids-singlecell) variant or the CPU (scanpy/squidpy) variant of the component. Selecting `gpu` requires a CUDA-capable GPU; the component errors out if none is available (there is no automatic fallback to CPU).

Name

Type & Properties

--device_type

string

Outputs

Name	Type & Properties	Description
--output	file required output	Output h5mu file.
--var_name_filter	string	In which .var slot to store a boolean array indicating which features are highly variable.
--varm_name	string	In which .varm slot to store the per-gene HVG metrics.
--output_compression	string	Compression format to use for the output AnnData and/or Mudata objects. By default no compression is applied.

Name

Type & Properties

--output

file

required

output

--var_name_filter

string

--varm_name

string

--output_compression

string

Options

Name	Type & Properties	Description
--flavor	string	Choose the flavor for identifying highly variable genes. For the dispersion-based methods in their default workflows, Seurat passes the cutoffs whereas Cell Ranger passes n_top_features. The 'seurat_v3_paper', 'pearson_residuals' and 'poisson_gene_selection' flavors are only available on the GPU (rapids-singlecell) variant.
--n_top_features	integer	Number of highly-variable features to keep. Mandatory if flavor is 'seurat_v3', 'seurat_v3_paper', 'pearson_residuals' or 'poisson_gene_selection'.
--min_mean	double	If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored. Only used for dispersion-based flavors.
--max_mean	double	If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored. Only used for dispersion-based flavors.
--min_disp	double	If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored. Only used for dispersion-based flavors.
--max_disp	double	If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored. Only used for dispersion-based flavors. Default is +inf.
--span	double	The fraction of the data (cells) used when estimating the variance in the loess model fit if flavor is 'seurat_v3' or 'seurat_v3_paper'.
--n_bins	integer	Number of bins for binning the mean gene expression. Normalization is done with respect to each bin. If just a single gene falls into a bin, the normalized dispersion is artificially set to 1.
--obs_batch_key	string	If specified, highly-variable features are selected within each batch separately and merged.

Name

Type & Properties

--flavor

string

--n_top_features

integer

--min_mean

double

--max_mean

double

--min_disp

double

--max_disp

double

--span

double

--n_bins

integer

--obs_batch_key

string

rapids-singlecell options

Name	Type & Properties	Description
--theta	integer	The negative binomial overdispersion parameter for Pearson residuals. Higher values correspond to less overdispersion. Only used if flavor is 'pearson_residuals'.
--clip	double	Determines if and how Pearson residuals are clipped. If unset, residuals are clipped to [-sqrt(n_obs), sqrt(n_obs)]. If a scalar c is given, residuals are clipped to [-c, c]. Only used if flavor is 'pearson_residuals'.
--chunksize	integer	If flavor is 'poisson_gene_selection', this determines how many genes are processed at once. Choosing a smaller value will reduce the required memory.
--n_samples	integer	The number of Binomial samples used to estimate the posterior probability of enrichment of zeros for each gene. Only used if flavor is 'poisson_gene_selection'.
--check_values	boolean	Check if counts in the selected layer are integers. A warning is emitted otherwise. Only used if flavor is 'seurat_v3', 'seurat_v3_paper', 'pearson_residuals' or 'poisson_gene_selection'.

Name

Type & Properties

--theta

integer

--clip

double

--chunksize

integer

--n_samples

integer

--check_values

boolean

Scanpy options

Name	Type & Properties	Description
--var_input	string	If specified, use boolean array in adata.var[var_input] to calculate hvg on subset of vars.
--features_to_exclude	string multiple	User-defined list of feature names to exclude before HVG calculation. These features will be excluded from HVG selection but will remain in the output data.

Name

Type & Properties

--var_input

string

--features_to_exclude

string

multiple

Run this component

Run the following command to execute this component with Nextflow:

cat > params.yaml <<'EOM'  
modality: [ "rna" ]  
device_type: [ "gpu" ]  
output: "$id.$key.output.h5mu"  
var_name_filter: [ "highly_variable" ]  
varm_name: [ "hvg" ]  
flavor: [ "seurat" ]  
min_mean: [ 0.0125 ]  
max_mean: [ 3 ]  
min_disp: [ 0.5 ]  
span: [ 0.3 ]  
n_bins: [ 20 ]  
theta: [ 100 ]  
chunksize: [ 1000 ]  
n_samples: [ 10000 ]  
check_values: [ true ]  
id: "run"  
publish_dir: "output/"  
EOM

nextflow run https://packages.viash-hub.com/vsh/openpipeline_rapids.git \  
  -revision v0.1.3 \  
  -main-script target/nextflow/wrappers/preprocessing/highly_variable_genes/main.nf \  
  -params-file params.yaml

wrappers/preprocessing/highly_variable_genes

Inputs

Compute

Outputs

Options

rapids-singlecell options

Scanpy options

Run this component

Relationships

Used by

Uses