run_pipeline {DrDimont}

Execute all DrDimont pipeline steps sequentially

Description

This wrapper function executes all necessary steps to generate differential integrated drug response scores from the formatted input data. The following input data is required (and detailed below):

* Layers of stratified molecular data.

* Additional connections between the layers.

* Interactions between drugs and nodes in the network.

* Settings for pipeline execution.

As this function runs through all steps of the DrDimont pipeline it can take a long time to complete, especially if the supplied molecular data is rather large. Several prompts will be printed to supply information on how the pipeline is proceeding. Calculation of the interaction score by generate_interaction_score_graphs requires saving large-scale graphs to file and calling a Python script. This handover may take time.

Eventually a data frame is returned containing the supplied drug name and its associated differential drug response score computed by DrDimont.

Usage

run_pipeline(
  layers,
  inter_layer_connections,
  drug_target_interactions,
  settings
)

Arguments

layers

[list] Named list with different network layers containing data and identifiers for both groups. The required input format is a list with names corresponding to the content of the respective layer (e.g., "protein"). Each named element has to contain the molecular data and corresponding identifiers formatted by make_layer.

inter_layer_connections

[list] A list with specified inter-layer connections. This list contains one or more elements defining individual inter-layer connections created by make_connection.

drug_target_interactions

[list] A list specifying drug-target interactions for drug response score computation. The required input format of this list is created by make_drug_target. The drug response score is calculated for all drugs contained in this object.

settings

[list] A named list containing pipeline settings. The settings list has to be initialized by drdimont_settings. Items in the named list can be adjusted as desired.

Value

Data frame containing drug name and associated differential integrated drug response score. If Python is not installed or the interaction score computation fails for some other reason, NULL is returned instead.

Examples

data(drug_gene_interactions)
data(metabolite_protein_interactions)
data(layers_example)

example_inter_layer_connections = list(make_connection(from='mrna', to='protein',
                                           connect_on='gene_name', weight=1),
                                       make_connection(from='protein', to='phosphosite',
                                           connect_on='gene_name', weight=1),
                                       make_connection(from='protein', to='metabolite',
                                           connect_on=metabolite_protein_interactions,
                                           weight='combined_score'))

example_drug_target_interactions <- make_drug_target(target_molecules='protein',
                                        interaction_table=drug_gene_interactions,
                                        match_on='gene_name')

example_settings <- drdimont_settings(
                        handling_missing_data=list(
                            default="pairwise.complete.obs",
                            mrna="all.obs"),
                        reduction_method="pickHardThreshold",
                        r_squared=list(default=0.65, metabolite=0.1),
                        cut_vector=list(default=seq(0.2, 0.65, 0.01)))


run_pipeline(
    layers=layers_example, 
    inter_layer_connections=example_inter_layer_connections, 
    drug_target_interactions=example_drug_target_interactions, 
    settings=example_settings)