wrappers/preprocessing/neighbors

Description

Compute a nearest-neighbor graph of observations, running either the GPU
(rapids-singlecell) or the CPU (scanpy) variant of neighbors, selected
with --device_type.

Inputs

Name	Type & Properties	Description
--input	file required	Input h5mu file.
--modality	string	Which modality from the input MuData file to process.
--obsm_input	string	Which .obsm slot to use as a starting embedding (use_rep).

Name

Type & Properties

--input

file

required

--modality

string

--obsm_input

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 containing the found neighbors.
--uns_output	string	In which .uns slot to store the neighbor graph metadata.
--obsp_distances	string	In which .obsp slot to store the distance matrix between the resulting neighbors.
--obsp_connectivities	string	In which .obsp slot to store the connectivities matrix between the resulting neighbors.
--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

--uns_output

string

--obsp_distances

string

--obsp_connectivities

string

--output_compression

string

Options

Name	Type & Properties	Description
--num_neighbors	integer	The size of the local neighborhood (in terms of number of neighboring data points) used for manifold approximation. Larger values result in more global views of the manifold, while smaller values result in more local data being preserved. In general values should be in the range 2 to 100.
--metric	string	The distance metric to use. Restricted to the metrics supported by both the CPU (scanpy) and GPU (rapids-singlecell) backends, so the value is valid regardless of --device_type. Each backend supports further metrics on its own, but only this common set is exposed here to keep the two variants interchangeable.
--random_state	integer	A random seed.

Name

Type & Properties

--num_neighbors

integer

--metric

string

--random_state

integer

rapids-singlecell options

Name	Type & Properties	Description
--n_pcs	integer	Number of dimensions of the `--obsm_input` representation to use. If not set, all available dimensions are used.
--algorithm	string	The KNN query algorithm to use (provided by cuVS). See https://docs.rapids.ai/api/cuvs/stable/ for details.
--method	string	Method for computing connectivities. 'umap' uses the UMAP fuzzy simplicial set, 'gauss' uses an adaptive Gaussian kernel, 'jaccard' uses the PhenoGraph Jaccard index.

Name

Type & Properties

--n_pcs

integer

--algorithm

string

--method

string

Run this component

Run the following command to execute this component with Nextflow:

cat > params.yaml <<'EOM'  
modality: [ "rna" ]  
obsm_input: [ "X_pca" ]  
device_type: [ "gpu" ]  
output: "$id.$key.output.h5mu"  
uns_output: [ "neighbors" ]  
obsp_distances: [ "distances" ]  
obsp_connectivities: [ "connectivities" ]  
num_neighbors: [ 15 ]  
metric: [ "euclidean" ]  
random_state: [ 0 ]  
algorithm: [ "brute" ]  
method: [ "umap" ]  
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/neighbors/main.nf \  
  -params-file params.yaml

wrappers/preprocessing/neighbors

Inputs

Compute

Outputs

Options

rapids-singlecell options

Run this component

Relationships

Used by

Uses