Reference file as input for rules to use correct ensembl version.

rules/PreprocessInput.smk

+from Reports import ReportSegmentation
+
 rule firehose:
     """Download copy number profiles for a cancer type of choice from firehose."""
     output:
@@ -44,6 +46,17 @@ rule define_input:
         else:       #split firehose data in tumor and normal files.
             split_normal_tumor(input[0], output.tumor, output.normal)
 
+rule report_seg:
+    """Report information on the input segmentation file."""
+    input:
+        seg="Input/Segments_tumor.txt"
+    output:
+        "Reports/Segments.txt"
+    run:
+        ReportSegmentation(input[0], output[0])
+
+
+
 def split_normal_tumor(all_samples, out_tumor, out_normal):
     """Split a segmentation file from firehose in a segmentation file with only tumor samples and one with only normal samples"""
     with open(all_samples, 'r') as old:

rules/RecurrentRegions.smk

 from Rubic import MarkerFile
-from Reports import ReportTools, ReportSegmentation
+from Reports import ReportTools
 import os.path
 import datetime
 
@@ -69,20 +69,11 @@ rule report_tools:
     params:                                                 #select input files from repository or own input files
         census=os.path.join(workflow.basedir, config["census_genes"]) if config["census_genes"].startswith("input_files") else config["census_genes"],
         known=os.path.join(workflow.basedir, config["prev_found_genes"]) if config["prev_found_genes"].startswith("input_files") else config["prev_found_genes"],
-        gene_info=os.path.join(workflow.basedir, config["biomart_genes"]) if config["biomart_genes"].startswith("input_files") else config["biomart_genes"]
+        gene_info=os.path.join(workflow.basedir, config["biomart_genes"]) if config["biomart_genes"].startswith("input_files") else config["biomart_genes"],
+        ref=config["reference"]
     run:
-        ReportTools(input.gistic, input.rubic, params.census, params.known, params.gene_info,
-                    output.tools,output.venn, output.genes_both, output.genes_gistic, output.genes_rubic, output.overlap, output.swarmplot)
-
-rule report_seg:
-    """Report information on the input segmentation file."""
-    input:
-        seg="Input/Segments_tumor.txt"
-    output:
-        "Reports/Segments.txt"
-    run:
-        ReportSegmentation(input[0], output[0])
-
+        ReportTools(input.gistic, input.rubic, params.census, params.known, params.gene_info, params.ref
+                    output.tools, output.venn, output.genes_both, output.genes_gistic, output.genes_rubic, output.overlap, output.swarmplot)
 
 
 from snakemake.utils import report

rules/SampleSizes.smk

@@ -60,6 +60,7 @@ rule report_sizes:
     params:
         census=config["census_genes"],
         known=config["prev_found_genes"],
-        reps=config["repeats"]
+        reps=config["repeats"],
+        ref=config["reference"]
     run:
-        ReportSizes(input[0], params.census, params.known, params.reps, output.report, output.plots)
+        ReportSizes(input[0], params.census, params.known, params.reps, params.ref, output.report, output.plots)

rules/UseControl.smk

@@ -24,6 +24,7 @@ rule report_control:
         "Reports/Control.txt"
     params:
         census=config["census_genes"],
-        known=config["prev_found_genes"]
+        known=config["prev_found_genes"],
+        ref=config["reference"]
     run:
-        ReportControl(input.control, input.nocontrol, output[0], params.census, params.known)
+        ReportControl(input.control, input.nocontrol, output[0], params.census, params.known, params.ref)

scripts/Reports.py

@@ -87,10 +87,10 @@ class ReportSegmentation:
 
 class ReportTools:
     """Make a report and gene file from the analyses with GISTIC2 and RUBIC"""
-    def __init__(self, gistic_dir, rubic_gain_file, census_genes, known_genes, gene_file, file_tools,
+    def __init__(self, gistic_dir, rubic_gain_file, census_genes, known_genes, gene_file, ref_genome, file_tools,
                  file_venn, file_genes_both, file_genes_GISTIC, file_genes_RUBIC, file_overlap, file_swarmplot):        #double brackets? ((...))
         rubic_dir = rubic_gain_file.split("/gains.txt")[0]   #get diretory with rubic results from gain file path.
-        install_ensembl("hg19")
+        install_ensembl(ref_genome)
         parsed, stats = self.calculate_stats(gistic_dir, rubic_dir, file_tools, census_genes, known_genes)
         self.make_tool_report(file_tools, stats)
 
@@ -253,8 +253,8 @@ class ReportTools:
 
 class ReportSizes:
     """Make a report and plots on analyses with different sample sizes"""
-    def __init__(self, size_results, census_genes, known_genes, reps, report_file, plot_dir):
-        install_ensembl("hg19")
+    def __init__(self, size_results, census_genes, known_genes, reps, ref_genome, report_file, plot_dir):
+        install_ensembl(ref_genome)
         self.make_sizes_report(size_results, census_genes, known_genes, reps, report_file, plot_dir)
 
     def make_sizes_report(self, size_results, census_genes, known_genes, reps, report_file, plot_dir):
@@ -328,8 +328,8 @@ class ReportSizes:
 
 class ReportControl:
     """Make a report on analyses using control samples or without using them."""
-    def __init__(self, control_results, nocontrol_results, report_file, census_genes, known_genes):
-        install_ensembl("hg19")
+    def __init__(self, control_results, nocontrol_results, report_file, census_genes, known_genes, ref_genome):
+        install_ensembl(ref_genome)
         self.make_control_report(control_results, nocontrol_results, report_file, census_genes, known_genes)
 
     def make_control_report(self, control_results, nocontrol_results, report_file, census_genes, known_genes):