nf-core/sopa

Path to comma-separated file containing information about the samples in the experiment.

The output directory where the results will be saved. You have to use absolute paths to storage on Cloud infrastructure.

Email address for completion summary.

Location of Space Ranger probeset file.

Location of Space Ranger reference directory. May be packed as tar.gz file.

Display the help message.

Display the full detailed help message.

Display hidden parameters in the help message (only works when —help or —help_full are provided).

Technology used for the spatial data, e.g., ‘xenium’, ‘merscope’, …

Optional page for the imageio reader

Width (and height) of each patch in pixels

Number of overlapping pixels between the patches. We advise to choose approximately twice the diameter of a cell

Width (and height) of each patch in microns

Number of overlapping microns between the patches. We advise to choose approximately twice the diameter of a cell

Optional name of the boundaries element to use as a segmentation prior. Either a column name for the transcript dataframe, or a key of sdata containing the shapes names. If combining cellpose with proseg/baysor/comseg, it will be set automatically to 'cellpose_boundaries'.

If prior_shapes_key is provided, this is the value given to transcripts that are not inside any cell (if it’s already 0, don’t provide this argument)

Whether to run tissue segmentation

Level of the image pyramid to use for tissue segmentation

Mode for the tissue segmentation: ‘staining’ or ‘saturation’ (for H&E images).

Additional tissue segmentation parameters as a python dict string

Cells with an area less than this value will be filtered. The unit is in pixels^2, and used by Stardist/Cellpose.

Cells with an area less than this value will be filtered. The unit is in microns^2, and used by Baysor/Comseg. Not used by Proseg.

Cells with less transcripts than this value will be filtered.

Cells whose mean channel intensity is less than min_intensity_ratio * quantile_90 will be filtered.

Whether to aggregate the genes (counts) inside each cell

Whether to aggregate the channels (intensity) inside each cell

Cells polygons will be expanded by expand_radius_ratio * mean_radius for channels averaging only. This help better aggregate boundary stainings

Whether to run scanpy preprocessing

Resolution parameter for the leiden clustering

Whether to check that adata.X contains counts

Whether to compute highly variable genes before computing the UMAP and clustering

Number of microns in a pixel. Invalid value can lead to inconsistent scales in the Explorer.

If True, will not load the full images in memory (except if the image memory is below ram_threshold_gb)

Threshold (in gygabytes) from which image can be loaded in memory.

Parameter for scipy gaussian_filter (applied before running the segmentation method)

Parameter for skimage.exposure.equalize_adapthist (applied before running the segmentation method)

Parameter for skimage.exposure.equalize_adapthist (applied before running the segmentation method)

Whether to run proseg segmentation

String suffix to add to the proseg command line. This can be used to add extra parameters to the proseg command line.

Whether to infer the proseg presets based on the columns of the transcripts dataframe.

Only for Visium HD data. Key of sdata containing the prior cell boundaries. If 'auto', use the latest performed segmentation (e.g., stardist or the 10X Genomics segmentation). If combining stardist with proseg, it will be set automatically to 'stardist_boundaries'.

Whether to run comseg segmentation

Comseg mean_cell_diameter parameter

Comseg max_cell_radius parameter

Comseg alpha parameter

Comseg min_rna_per_cell parameter

Comseg allow_disconnected_polygon parameter

Comseg norm_vector parameter

Whether to run baysor segmentation

Baysor scale parameter

Baysor scale_std parameter

Baysor prior_segmentation_confidence parameter

Baysor min_molecules_per_cell parameter

Baysor min_molecules_per_gene parameter

Baysor min_molecules_per_segment parameter

Baysor confidence_nn_id parameter

Baysor force_2d parameter

Whether to run cellpose segmentation

Cellpose diameter parameter

Channel name(s) to use for cellpose segmentation. If multiple, separate by space, comma or pipe characters.

Cellpose flow_threshold parameter

Cellpose cellprob_threshold parameter

Cellpose model type to use

Cellpose pretrained_model parameter

Whether to use GPU for Cellpose segmentation

Additional cellpose parameters as a python dict string

Whether to run stardist segmentation

Name of stardist model to use

Stardist prob_thresh parameter

Stardist nms_thresh parameter

Optional channel name(s) to use for stardist segmentation. If multiple, separate by space, comma or pipe characters.

Additional stardist parameters as a python dict string

Whether to run tangram cell-type annotation

Path to the scRNAseq annotated reference, as a .h5ad file

Key of adata_ref.obs containing the cell-types

Preprocessing method applied to the reference. Either None (raw counts), or normalized (sc.pp.normalize_total) or log1p (sc.pp.normalize_total and sc.pp.log1p)

Number of cells in each bag of the spatial table. Low values will decrease the memory usage

Maximum samples to be considered in the reference for tangram. Low values will decrease the memory usage

Whether to run cell-type annotation based on a marker-to-cell dictionary

Key of sdata.obs containing the cell-types

Dictionary mapping whose keys are marker channel names and values are the cell types associated to each marker. Should be provided as a string representation of a python dictionary.