Skip to content

Run

Jump to bottom
Michele Bortolomeazzi edited this page May 5, 2023 · 20 revisions

Running SIMPLI

SIMPLI is run as a Nextflow pipeline:

nextflow run PATH_TO_SIMPLI/main.nf [COMMAND_LINE_ARGUMENTS] [NEXTFLOW_OPTIONS]

The most recent version of SIMPLI be downloaded and run from directly from this repository with:
nextflow run ciccalab/SIMPLI [COMMAND_LINE_ARGUMENTS] [NEXTFLOW_OPTIONS]

See the Nextflow documentation for more details.

Configuration file

In alternative all command line parameters can be specified through a configuration file: nextflow run main.nf -c CONFIGURATION_FILE

Parameters in the configuration file are specified in this format:

params.PARAMETER_NAME ="VALUE"

SIMPLI Parameters

SIMPLI accepts the following parameters (specified through the command line or a configuration file).

Input

These parameters specify the paths to the metadata files defining the samples and their associated images and channels:

  • sample_metadata_file = Metadata file defining all the samples used in the analysis.
  • tiff_input_metadata_file = Metadata file for TIFF input (single- or multi-channel).
  • raw_metadata_file = Metadata file for in input of Imaging Mass Cytometry data.
  • channel_metadata = Metadata file defining the channels for Imaging Mass Cytometry input data.

See the input page for their fields and formatting.

Selection of analysis steps

Choose which steps of the analysis to skip. Valid values are false and true.

  • skip_conversion = Do not perform the extraction of TIFF images from Imaging Mass Cytometry input data.
  • skip_normalization = Do not perform the 99th percentile normalization of the images by channnel and by sample.
  • skip_preprocessing = Do not perform image preprocessing with CellProfiler4.
  • skip_area = Do not perform the pixel-level analysis.
  • skip_cp_segmentation = Do not perform cell segmentation with CellProfiler4.
  • skip_sd_segmentation = Do not perform cell segmentation with StarDist.
  • skip_cell_type_identification = Do not perform cell masking.
  • skip_cell_clustering = Do not perform cell phenotyping by unsupervised clustering.
  • skip_cell_thresholding = Do not perform cell phenotyping by expression thresholding.
  • skip_homotypic_interactions = Do not perform homotypic interaction analysis.
  • skip_heterotypic_interactions = Do not perform heterotypic interaction analysis.
  • skip_permuted_interactions = Do not perform the permutation test for the heterotypic interaction analysis.
  • skip_visualization = Do not perform any of the following visualization steps.
  • skip_area_visualization = Do not plot the results of the pixel-level analysis.
  • skip_type_visualization = Do not plot the results of the cell masking.
  • skip_cluster_visualization = Do not plot the results of the cell phenotyping by unsupervised clustering.
  • skip_thresholding_visualization = Do not plot the results of the cell phenotyping by expression thresholding.
  • skip_homotypic_visualization = Do not plot the results of the homotypic interaction analysis.
  • skip_heterotypic_visualization = Do not plot the results of the heterotypic interaction analysis.
  • skip_permuted_visualization = Do not plot the results of the permutation test for the heterotypic interaction analysis.

See the analysis page for details on each step of the analysis.

Step Specific metadata files:

  • area_measurements_metadata = Metadata file with the parameters for the pixel-level analysis.
  • cell_clustering_metadata = Metadata file with the parameters for the cell phenotyping by unsupervised clustering.
  • cell_thresholding_metadata = Metadata file with the parameters for the cell phenotyping by expression thresholding.
  • cell_masking_metadata = Metadata file with the parameters for cell masking.
  • homotypic_interactions_metadata = Metadata file with the parameters for the homotypic interaction analysis.
  • heterotypic_interactions_metadata = Metadata file with the parameters for the heterotypic interaction analysis.

See the analysis page for details on the format of these files.

Step Specific input files:

  • normalized_metadata_file = Metadata file for normalized tiff images.
  • preprocessed_metadata_file = Metadata file for preprocessed tiff images.
  • area_measurements = File with area measurements data.
  • single_cell_data_file = File with single cell data.
  • annotated_cell_data_file = File with single cell data annotated with cell identities.
  • clustered_cell_data_file = File with single cell data annotated with cell phenotypes from unsupervised clustering.
  • thresholded_cell_data_file = File with single cell data annotated with cell phenotypes from expression thresholding.
  • homotypic_interactions_file = File with single cell data annotated with spatial cell clusters from the homotypic interaction analysis.
  • heterotypic_interactions_file = File with cell-cell distances from the heterotypic interaction analysis.
  • single_cell_masks_metadata = Metadata file associating each sample to a UINT16 TIFF cell mask.
  • shuffled_interactions_file = File with the permuted cell-cell distances from the heterotypic interaction analysis.

See the analysis page for the required inputs of each step of the analysis and their required fields and formats.

Tool from which the cell data will be analyzed

cell_source = "StarDist" ("StarDist", "CellProfiler", null) not required if only one of the two segmentation tools is used.

CellProfiler4 pipelines

These parameters specify the paths to the CellProfiler4 pipeline files:

  • cp4_preprocessing_cppipe = Path to the CellProfiler4 pipeline file used for image preprocessing.
  • cp4_segmentation_cppipe = Path to the CellProfiler4 pipeline file used for cell segmentation.

See the CellProfiler4 pipelines page for the requirements of a SIMPLI compatible CellProfiler4 pipeline.

StarDist segmentation parameters

  • sd_labels_to_segment = comma separated list of labels to include in the multichannel image used for segmentation. Comma separated liost. the names must match those in the preprocessed_metadata_file
  • sd_model_name = name of the model to use
  • sd_model_path = path to the model to use or "default" for the models included with StarDist
  • sd_prob_thresh = probability threshold used for calling cells: 0 < value < 1 or "default" to use the default valuse saved in the model.
  • sd_nms_thresh = threshold above which Non-Maximum Suppression is performed: 0 < value < 1 or "default" to use the default valuse saved in the model

Heterotypic spatial analysis

  • permutation = Number of permutations (cell label shuffling, recommended > 10000) to perform for the permutation test in the permutation analysis.

Visualization

These parameters specify the colors used to generate the gradients representing different levels of gene expression in UMAPs and Heatmaps:

  • high_color = "'#FF0000'"
  • mid_color = "'#FFFFFF'"
  • low_color = "'#0000FF'"

Accepted values are color names or hexadecimal #RGB or #RGBA format ("#RRGGBB" or "#RRGGBBAA").

Output

  • output_folder = Specifies the path where all output will be collected.
  • tiff_type = single or ome. Output the preprocessed images as multiple single-channel tiff files, or as a single multi-channel .ome.tiff image.

SIMPLI in HPC environments

SIMPLI can be run in High Performance Computing (HPC) environments using Nextflow to automatically manage job submission. The nextflow.config file defines three execution profiles:

  • local = Execute all processes on the local machine.
  • sge = Profile for job submission on (Open Grid Scheduler/Grid Enginer)[http://gridscheduler.sourceforge.net/documentation.html] based clusters.
  • slurm = Profile for job submission on (Slurm Workload Manager)[https://slurm.schedmd.com/documentation.html] based clusters.

These profiles define a minimum set of options which can be further expanded to suit the configuration of specific computing platforms or analysis. Additional profiles can also be defined by the user.

Clone this wiki locally