Made names of rules more understandable.

rules/Circos.smk

 from Circos import InputCircos, bed_to_circos
 
-rule circos_input:
+rule make_CIRCOS_input:
     """Make input files for making a circos diagram."""
     input:
         seg="Input/Segments_tumor.txt",
@@ -14,7 +14,7 @@ rule circos_input:
     run:
         InputCircos(input.seg, input.gistic, input.rubic_gains, input.rubic_losses, output.seg, output.gistic, output.rubic)
 
-rule make_circos:
+rule make_CIRCOS_plot:
     """Make circos diagram of recurrent regions in RUBIC and GISTIC2.0"""
     input:
         seg="Reports/Circos/Segments.txt",
@@ -23,13 +23,13 @@ rule make_circos:
     output:
         "Reports/Circos/RecurrentRegions.png"
     params:
-        workflow.basedir + "/scripts/circos/circos.conf"
+        conf=workflow.basedir + "/scripts/circos/circos.conf"
     conda:
         workflow.basedir + "/envs/circos.yaml"
     shell:
-        "circos -conf {params[0]} -outputfile {output[0]} -param cnv_file={input.seg} -param gistic_file={input.gistic} -param rubic_file={input.rubic}"
+        "circos -conf {params.conf} -outputfile {output[0]} -param cnv_file={input.seg} -param gistic_file={input.gistic} -param rubic_file={input.rubic}"
 
-rule make_legend_circos:
+rule make_legend_CIRCOS:
     output:
         "Reports/Circos/legend.png"
     run:
@@ -45,7 +45,7 @@ rule make_legend_circos:
         bbox = legend.get_window_extent().transformed(fig.dpi_scale_trans.inverted())
         fig.savefig(output[0], dpi=400, bbox_inches=bbox)
 
-rule add_legend_circos:
+rule add_legend_CIRCOS:
     input:
         "Reports/Circos/RecurrentRegions.png",
         "Reports/Circos/legend.png"
@@ -61,7 +61,7 @@ rule add_legend_circos:
         circos.paste(legend, offset)
         circos.save(output[0])
 
-rule circos_input_zoom:
+rule circos_input_zoom: #necessary?
     """Make input files for making a circos diagram."""
     input:
         bed="Reports/Overlap_known_genes.bed"
@@ -72,17 +72,22 @@ rule circos_input_zoom:
     run:
         bed_to_circos(input.bed, output.rubic, output.gistic, output.genes)
 
-rule make_circos_zoom:
+def get_list_genes(gene_file):
+    with open(gene_file, 'r') as genes:
+        for line in genes:
+            print(line)
+
+rule make_zoomed_CIRCOS_plot:
     """Compare locations of known genes, recurrent regions from RUBIC and recurrent regions from GISTIC2."""
     input:
         gistic="Reports/Circos/Zoom/GISTIC.txt",
         rubic="Reports/Circos/Zoom/RUBIC.txt",
-        genes="Reports/Circos/Zoom/Genes.txt",
+        genes="Reports/Circos/Zoom/Genes.txt"
     output:
-        plots="Reports/Overlap_plots/12.png"
+        plots=expand("Reports/Overlap_plots/{gene}.png", gene="NR6") #get_list_genes(input.genes))
     params:
         workflow.basedir + "/scripts/circos/circos_zoom.conf",
-        chrom='hs12'
+        chrom='hs12:67-71'
     conda:
         workflow.basedir + "/envs/circos.yaml"
     shell:

rules/ComparisonRegions.smk

@@ -31,7 +31,7 @@ rule report_tools:
                                 output.genes_both, output.genes_gistic, output.genes_rubic,
                                 input.overlap, output.bed_known)
 
-rule bed_intersect:
+rule get_overlap_GISTIC_RUBIC:
     """Intersect the recurrent regions detected by RUBIC and GISTIC2.0."""
     input:
         gistic="GISTIC/regions_track.conf_" + config["gistic_precision"] + ".bed",
@@ -43,7 +43,7 @@ rule bed_intersect:
     shell:
         "bedtools intersect -a {input.gistic} -b {input.rubic} -wo > {output}"
 
-rule bed_known_genes:
+rule get_overlap_known_genes:
     """Intersect known genes and recurrent regions detected by RUBIC and GISTIC2.0."""
     input:
         known="Reports/Locations_known_genes.bed",

rules/ComparisonSettings.smk

+def get_settings(nr_settings, all_settings):
+    print(nr_settings)
+    print(all_settings)
+    print(list(range(len(config["comparison_settings"]))))
+
 rule gistic_settings:
     """Run GISTIC2 based on different settings."""
     input:
         gistic_directory=os.path.join(config["gisticdir"], "gistic2"),
-        seg="Input/Segments_tumor.txt"
+        seg="Input/Segments_tumor.txt",
+        lambda wildcards: config["comparison_settings"][wildcards.setting]
     output:
-        expand("Settings/GISTIC_{setting_nr}/all_lesions.conf_" + config["gistic_precision"] + ".txt", setting_nr=range(len(config["comparison_settings"])),
-        expand("Settings/GISTIC_{setting_nr}/regions_track.conf_" + config["gistic_precision"] + ".bed", setting_nr=range(len(config["comparison_settings"]))
+        "Settings/GISTIC_wildcards.setting/all_lesions.conf_" + config["gistic_precision"] + ".txt",
+        "Settings/GISTIC_wildcards.setting/regions_track.conf_" + config["gistic_precision"] + ".bed"
     params:
         cnv="",
         ref=config["reference"],
         ref_file="",
-        extra=config["comparison_settings"][{setting_nr}]
+        extra="wildcards.setting",
+        confidence=config["gistic_precision"]
     wrapper:
         "file:" + workflow.basedir + "/wrappers/GISTIC2"
 

rules/GISTIC2.smk

 import datetime
 import os.path
 
-rule install_gistic:
+rule install_GISTIC:
     """Install GISTIC2 to a directory of choice."""
     output:
         os.path.join(config["gisticdir"], "gistic2")
@@ -10,7 +10,7 @@ rule install_gistic:
     shell:
         "{workflow.basedir}/scripts/install_gistic2.sh {params}"
 
-rule run_gistic:
+rule run_GISTIC:
     """Run GISTIC2 for the tumor segmentation data."""
     input:
         gistic_directory=os.path.join(config["gisticdir"], "gistic2"),
@@ -22,7 +22,8 @@ rule run_gistic:
         cnv="",
         ref=config["reference"],
         ref_file="",
-        extra=config["settings_gistic"]
+        extra=config["settings_gistic"],
+        confidence=config["gistic_precision"]
     benchmark:
         "Benchmarks/GISTIC2." + str(datetime.datetime.now()).replace(" ", "_") + ".txt"
     wrapper:

rules/GenePrioritization.smk

-rule go_analysis:
+rule do_GO_analysis:
     """Perform a gene ontology enrichment analysis using topGO."""
     input:
         gene_list="Reports/Genes_{tool}.txt"
@@ -13,7 +13,7 @@ rule go_analysis:
     wrapper:
         "file:" + workflow.basedir + "/wrappers/topgo"
 
-rule compare_enriched_terms:
+rule compare_enriched_GOs:
     """Compare the top 50 GO terms detected by RUBIC and GISTIC2.0"""
     input:
         go=expand("GO/{tool}.txt", tool=["GISTIC2", "RUBIC"])

rules/PreprocessInput.smk

 import ReportSegments
 
-rule firehose:
+rule download_firehose_data:
     """Download copy number profiles for a cancer type of choice from firehose."""
     output:
         "Input/Segments_all.txt"
@@ -22,11 +22,11 @@ def pipeline_input(wildcards):
     if (provided_input != ""):     #use provided input file
         return provided_input
     elif (provided_input == "") and (config["cancer_type"] != "") and (config["date_data"] != ""):       #use firehose data
-        return rules.firehose.output
+        return rules.download_firehose_data.output
     else:
         raise ValueError("Provide either a segmentation file as input or get firehose data based on the cancer type and date of choice.")
 
-rule define_input:
+rule define_input_pipeline:
     """Create a tumor and normal segmentation file based on the provided input."""
     input:
         pipeline_input
@@ -46,7 +46,7 @@ rule define_input:
         else:       #split firehose data in tumor and normal files.
             split_normal_tumor(input[0], output.tumor, output.normal)
 
-rule report_seg:
+rule report_segmentation_file:
     """Report information on the input segmentation file."""
     input:
         seg="Input/Segments_tumor.txt"

rules/Rubic.smk

 from Rubic import MarkerFile, BedFile
 
-rule markers_rubic:
+rule make_marker_file_RUBIC:
     """Make marker file to use as input for RUBIC based on segmentation file (start, center and end positions of each segment)."""
     input:
         "Input/Segments_tumor.txt"
@@ -9,7 +9,7 @@ rule markers_rubic:
     run:
         MarkerFile(input[0], output[0])
 
-rule run_rubic:
+rule run_RUBIC:
     """Run RUBIC for the tumor segmentation data."""
     input:
         seg="Input/Segments_tumor.txt",
@@ -26,7 +26,7 @@ rule run_rubic:
     wrapper:
         "file:" + workflow.basedir +"/wrappers/rubic"
 
-rule make_bed_rubic:
+rule make_bed_file_RUBIC:
     input:
         gains="RUBIC/gains.txt",
         losses="RUBIC/losses.txt"

rules/SampleSizes.smk

@@ -2,8 +2,9 @@ import ReportSizes
 from SampleSizes import SegFile
 import os.path
 import datetime
+from AUC import ROC_curve
 
-rule seg_subsets:
+rule get_segmentation_files_subsets:
     """Create segmentation files with different numbers of samples (randomly chosen) for a number of times."""
     input:
         "Input/Segments_tumor.txt"
@@ -12,7 +13,7 @@ rule seg_subsets:
     run:
         SegFile(input[0], output)
 
-rule run_gistic_subsets:
+rule run_GISTIC_subsets:
     """Run GISTIC2 for the segmentation files with different subsets."""
     input:
         gistic_directory=os.path.join(config["gisticdir"], "gistic2"),
@@ -24,11 +25,12 @@ rule run_gistic_subsets:
         cnv="",
         ref=config["reference"],
         ref_file="",
-        extra=config["settings_gistic"]
+        extra=config["settings_gistic"],
+        confidence=config["gistic_precision"]
     wrapper:
         "file:" + workflow.basedir + "/wrappers/GISTIC2"
 
-rule run_rubic_subsets:
+rule run_RUBIC_subsets:
     """Run RUBIC for the segmentation files with different subsets."""
     input:
         seg="Samplesize/Input/Size{rand_nr}.Rep{rep_nr}.txt",
@@ -48,8 +50,8 @@ rule run_rubic_subsets:
         #"68ec08a/bio/rubic"
         "file:" + workflow.basedir +"/wrappers/rubic"
 
-rule report_sizes:
-    """Report the difference when using different sample sizes."""
+rule report_size_differences:
+    """Report the differences when using different sample sizes."""
     input:
         gistic=expand("Samplesize/GISTIC/Size{rand_nr}.Rep{rep_nr}/all_lesions.conf_" + config["gistic_precision"] + ".txt", rand_nr=config["sizes"], rep_nr=config["repeats"]),
         rubic_gains=expand("Samplesize/RUBIC/Size{rand_nr}.Rep{rep_nr}/gains.txt", rand_nr=config["sizes"], rep_nr=config["repeats"]),
@@ -65,3 +67,55 @@ rule report_sizes:
         ref=config["reference"]
     run:
         ReportSizes.make_report(input.gistic, input.rubic_gains, input.rubic_losses, params.census, params.known, params.reps, params.sizes, params.ref, output.report, output.plots)
+
+rule make_bed_RUBIC_subsets:
+    """Make a bed file with the regions detected using RUBIC with a subset of the total sample."""
+    input:
+        gains="Samplesize/RUBIC/Size{rand_nr}.Rep{rep_nr}/gains.txt",
+        losses="Samplesize/RUBIC/Size{rand_nr}.Rep{rep_nr}/losses.txt"
+    output:
+        bed="Samplesize/RUBIC/Size{rand_nr}.Rep{rep_nr}/regions_track.bed"
+    run:
+        BedFile(input.gains, input.losses, output.bed)
+
+rule compare_subset_truth_GISTIC:
+    """Do bedtools intersect between the GISTIC bed file with subset and bed file all samples."""
+    input:
+        bed_subsets="Samplesize/GISTIC/Size{rand_nr}.Rep{rep_nr}/regions_track.conf_" + config["gistic_precision"] + ".bed",
+        bed_truth="GISTIC/regions_track.conf_" + config["gistic_precision"] + ".bed"
+    output:
+        subset="Samplesize/GISTIC/Size{rand_nr}.Rep{rep_nr}/Overlap_subset_truth.bed",
+        truth="Samplesize/GISTIC/Size{rand_nr}.Rep{rep_nr}/Overlap_truth_subset.bed",
+    conda:
+        workflow.basedir + "/envs/bedtools.yaml"
+    shell:
+        "bedtools intersect -a {input.bed_subsets} -b {input.bed_truth} -wao > {output.subset} && \
+        bedtools intersect -a {input.bed_truth} -b {input.bed_subsets} -wao > {output.truth}"
+
+rule compare_subset_truth_RUBIC:
+    """Do bedtools intersect between the RUBIC bed file with subset and bed file all samples."""
+    input:
+        bed_subsets="Samplesize/RUBIC/Size{rand_nr}.Rep{rep_nr}/regions_track.bed",
+        bed_truth="RUBIC/regions_track.bed"
+    output:
+        subset="Samplesize/RUBIC/Size{rand_nr}.Rep{rep_nr}/Overlap_subset_truth.bed",
+        truth="Samplesize/RUBIC/Size{rand_nr}.Rep{rep_nr}/Overlap_truth_subset.bed",
+    conda:
+        workflow.basedir + "/envs/bedtools.yaml"
+    shell:
+        "bedtools intersect -a {input.bed_subsets} -b {input.bed_truth} -wao > {output.subset} && \
+        bedtools intersect -a {input.bed_truth} -b {input.bed_subsets} -wao > {output.truth}"
+
+rule make_ROC_plot:
+    """Make ROC plot on the precision and recall from the subsets using GISTIC and RUBIC."""
+    input:
+        gistic_subset=expand("Samplesize/GISTIC/Size{rand_nr}.Rep{rep_nr}/Overlap_subset_truth.bed", rand_nr=config["sizes"], rep_nr=config["repeats"]),
+        gistic_truth=expand("Samplesize/GISTIC/Size{rand_nr}.Rep{rep_nr}/Overlap_truth_subset.bed", rand_nr=config["sizes"], rep_nr=config["repeats"]),
+        rubic_subset=expand("Samplesize/RUBIC/Size{rand_nr}.Rep{rep_nr}/Overlap_subset_truth.bed", rand_nr=config["sizes"], rep_nr=config["repeats"]),
+        rubic_truth=expand("Samplesize/RUBIC/Size{rand_nr}.Rep{rep_nr}/Overlap_truth_subset.bed", rand_nr=config["sizes"], rep_nr=config["repeats"]),
+    output:
+        AUC="Samplesize/Precision_recall.png"
+    params:
+        sizes=config["sizes"]
+    run:
+        ROC_curve(input.gistic_subset, input.gistic_truth, input.rubic_subset, input.rubic_truth, output.AUC, params.sizes)

rules/UseControl.smk

 import ReportControl
 
-rule run_gistic_control:
+rule run_GISTIC_control:
     """Run GISTIC2 for the tumor segmentation data with data from control samples included."""
     input:
+        gistic_directory=os.path.join(config["gisticdir"], "gistic2"),
         seg="Input/Segments_tumor.txt"
     output:
-        "Control/"
+        "Control/all_lesions.conf_" + config["gistic_precision"] + ".txt",
+        "Control/regions_track.conf_" + config["gistic_precision"] + ".bed"
     params:
         cnv="Input/Segments_normal.txt",
-        gistic_directory=config["gisticdir"],
         ref=config["reference"],
         ref_file="",
-        extra=""
+        extra="",
+        confidence=config["gistic_precision"]
     wrapper:
-        "file:" + os.path.join(workflow.basedir, "wrappers/GISTIC2")
+        "file:" + workflow.basedir + "/wrappers/GISTIC2"
 
 rule report_control:
     """Report the differences between using a control and without using a control."""