Merge pull request #117 from AlexandrovLab/U57

U57

Author: mdbarnesUCSD

SigProfilerAssignment/DecompositionPlots/ExampleSample/COSMIC_CNV48_Signatures.txt

@@ -1,4 +1,4 @@
-MutationType	CN1	CN2	CN9	CN20	CNV48B	CNV48D
+MutationType	CN1	CN2	CN9	CN20	CN3	CN4
 0:homdel:0-100kb	0.00179505	0.00223767	0.00818184	0.00004078	0.00000295	0.00000075
 0:homdel:100kb-1Mb	0.00380710	0.00719579	0.02099919	0.00017845	0.00039561	0.02703219
 0:homdel:>1Mb	0.00243675	0.00383711	0.00900922	0.00011722	0.00122668	0.02703635

SigProfilerAssignment/DecompositionPlots/ExampleSample/COSMIC_SV32_Signatures.txt

@@ -0,0 +1,33 @@
+MutationType	SV1	SV2	SV3	SV4	SV5	SV6	SV7	SV9
+clustered_del_1-10Kb	0.00001830	0.00001750	0.00000831	0.00558463	0.00162643	0.00484161	0.00036457	0.01689226
+clustered_del_10-100Kb	0.00000013	0.00000500	0.00000000	0.00883802	0.00395326	0.00397105	0.00493403	0.02027149
+clustered_del_100Kb-1Mb	0.00004260	0.00000257	0.00000253	0.01342056	0.00264242	0.01567818	0.00175481	0.04559844
+clustered_del_1Mb-10Mb	0.00000003	0.00000557	0.00000000	0.01941683	0.00079517	0.06315566	0.00000135	0.08553749
+clustered_del_>10Mb	0.00000003	0.00000224	0.00000009	0.01212996	0.00000205	0.13914616	0.00000121	0.00004090
+clustered_tds_1-10Kb	0.00000029	0.00002000	0.00233538	0.00119602	0.00044748	0.00173254	0.00023931	0.00407692
+clustered_tds_10-100Kb	0.00080029	0.00003770	0.00137651	0.00406629	0.00092090	0.00411956	0.00031235	0.00799563
+clustered_tds_100Kb-1Mb	0.00217807	0.00001180	0.00018608	0.00927806	0.00180766	0.01359699	0.00000078	0.04878886
+clustered_tds_1Mb-10Mb	0.00006350	0.00000161	0.00000004	0.01669456	0.00008410	0.06927579	0.00000076	0.07607224
+clustered_tds_>10Mb	0.00000003	0.00000314	0.00000010	0.00950079	0.00003530	0.14468700	0.00000187	0.00019873
+clustered_inv_1-10Kb	0.00003510	0.00009050	0.00000000	0.01188233	0.01068676	0.00333129	0.00000116	0.00901519
+clustered_inv_10-100Kb	0.00000113	0.00002000	0.00000415	0.01283746	0.00421801	0.00778075	0.00000166	0.03440818
+clustered_inv_100Kb-1Mb	0.00062736	0.00000829	0.00000000	0.01893295	0.00104651	0.02657075	0.00000052	0.08534106
+clustered_inv_1Mb-10Mb	0.00000637	0.00000052	0.00000003	0.03875053	0.00018613	0.13269196	0.00000021	0.15018331
+clustered_inv_>10Mb	0.00000001	0.00000229	0.00000003	0.02250855	0.00000130	0.26682614	0.00000158	0.00004010
+clustered_trans	0.00005860	0.00000182	0.00000829	0.74672823	0.01500836	0.00000039	0.00000036	0.00000269
+non-clustered_del_1-10Kb	0.00016322	0.01077137	0.03108973	0.00000143	0.27051044	0.00049070	0.11080824	0.03271109
+non-clustered_del_10-100Kb	0.01062551	0.00189318	0.00026854	0.00000001	0.04236249	0.00047467	0.40908574	0.03762502
+non-clustered_del_100Kb-1Mb	0.01031305	0.00009900	0.00051531	0.00110315	0.02494482	0.00056581	0.34862452	0.05284083
+non-clustered_del_1Mb-10Mb	0.00810848	0.04772402	0.00323839	0.00114858	0.04036508	0.00908681	0.02776373	0.04842356
+non-clustered_del_>10Mb	0.00460793	0.05938062	0.00861972	0.00280099	0.02737192	0.01709695	0.01575099	0.01781587
+non-clustered_tds_1-10Kb	0.00000013	0.00004480	0.42763680	0.00000015	0.08730244	0.00000010	0.01078117	0.01195956
+non-clustered_tds_10-100Kb	0.04599380	0.00002460	0.32675443	0.00000001	0.12292716	0.00000054	0.00652355	0.01970948
+non-clustered_tds_100Kb-1Mb	0.45491853	0.01413625	0.01600693	0.00008590	0.06600995	0.00067638	0.00545841	0.01400244
+non-clustered_tds_1Mb-10Mb	0.35062795	0.04029646	0.00096917	0.00013931	0.01815304	0.00881413	0.00239992	0.02583554
+non-clustered_tds_>10Mb	0.00580397	0.03967873	0.00649925	0.00298543	0.01284238	0.01668428	0.00007000	0.00744539
+non-clustered_inv_1-10Kb	0.00440008	0.00000085	0.00411229	0.00018744	0.04638068	0.00022783	0.00000043	0.01662502
+non-clustered_inv_10-100Kb	0.01048762	0.00459802	0.00275734	0.00116401	0.03219888	0.00171438	0.03025930	0.01859228
+non-clustered_inv_100Kb-1Mb	0.01700822	0.05558915	0.00038638	0.00198348	0.03887257	0.00243843	0.01706313	0.03888396
+non-clustered_inv_1Mb-10Mb	0.01075404	0.09313637	0.00482722	0.00198335	0.03699181	0.01686964	0.00553730	0.04676524
+non-clustered_inv_>10Mb	0.01304926	0.09976361	0.01624505	0.01157483	0.04637754	0.02742841	0.00186541	0.02399899
+non-clustered_trans	0.04930641	0.53263233	0.14615191	0.02307616	0.04292692	0.00002510	0.00039160	0.00230227

SigProfilerAssignment/DecompositionPlots/ExampleSample/SV32_De-Novo_Signatures.txt

@@ -0,0 +1,33 @@
+MutationType	SV32A	SV32B	SV32C
+clustered_del_1-10Kb	0.00662723	0.00177044	0.00000007
+clustered_del_10-100Kb	0.00972409	0.00167625	0.00000003
+clustered_del_100Kb-1Mb	0.02672959	0.00069852	0.00000004
+clustered_del_1Mb-10Mb	0.05108407	0.00033823	0.00000009
+clustered_del_>10Mb	0.03969927	0.00000047	0.00000040
+clustered_tds_1-10Kb	0.00166860	0.00057069	0.00011002
+clustered_tds_10-100Kb	0.00695235	0.00099348	0.00000008
+clustered_tds_100Kb-1Mb	0.02396912	0.00136236	0.00000000
+clustered_tds_1Mb-10Mb	0.04294237	0.00000963	0.00000009
+clustered_tds_>10Mb	0.04211685	0.00000072	0.00000016
+clustered_inv_1-10Kb	0.01459632	0.00452079	0.00000000
+clustered_inv_10-100Kb	0.01751233	0.00170391	0.00000000
+clustered_inv_100Kb-1Mb	0.03681359	0.00037369	0.00000012
+clustered_inv_1Mb-10Mb	0.10623513	0.00000047	0.00000013
+clustered_inv_>10Mb	0.07525245	0.00108337	0.00000002
+clustered_trans	0.23293450	0.00000023	0.00000022
+non-clustered_del_1-10Kb	0.00016344	0.11881046	0.04395760
+non-clustered_del_10-100Kb	0.00447025	0.09378964	0.00000052
+non-clustered_del_100Kb-1Mb	0.01586304	0.03619608	0.00000016
+non-clustered_del_1Mb-10Mb	0.02913311	0.02808863	0.00010709
+non-clustered_del_>10Mb	0.01433401	0.03214281	0.00690197
+non-clustered_tds_1-10Kb	0.00022787	0.00000001	0.34090417
+non-clustered_tds_10-100Kb	0.00000028	0.01990271	0.34120828
+non-clustered_tds_100Kb-1Mb	0.00938708	0.10950006	0.06107216
+non-clustered_tds_1Mb-10Mb	0.02086042	0.02455653	0.02874254
+non-clustered_tds_>10Mb	0.01097690	0.02568377	0.01287658
+non-clustered_inv_1-10Kb	0.00947122	0.06278640	0.00888667
+non-clustered_inv_10-100Kb	0.00786742	0.03894469	0.00344490
+non-clustered_inv_100Kb-1Mb	0.01588333	0.02367522	0.00125801
+non-clustered_inv_1Mb-10Mb	0.03501238	0.04148272	0.00366291
+non-clustered_inv_>10Mb	0.03495364	0.06231916	0.01299224
+non-clustered_trans	0.05653771	0.26701783	0.13387279

SigProfilerAssignment/DecompositionPlots/PlotDecomposition.py

@@ -26,14 +26,15 @@
 import SigProfilerAssignment
 import SigProfilerAssignment.DecompositionPlots
 from SigProfilerAssignment.DecompositionPlots import SigProfilerPlottingMatrix as mPlt
-from SigProfilerAssignment.DecompositionPlots import PlotDecomposition_SBS96 as spd_96
-from SigProfilerAssignment.DecompositionPlots import PlotDecomposition_SBS288 as spd_288
 from SigProfilerAssignment.DecompositionPlots import (
+    PlotDecomposition_SBS96 as spd_96,
+    PlotDecomposition_SBS288 as spd_288,
     PlotDecomposition_SBS1536 as spd_1536,
+    PlotDecomposition_DBS78 as spd_78,
+    PlotDecomposition_ID83 as spd_83,
+    PlotDecomposition_CNV48 as cnv_48,
+    PlotDecomposition_SV32 as sv_32,
 )
-from SigProfilerAssignment.DecompositionPlots import PlotDecomposition_DBS78 as spd_78
-from SigProfilerAssignment.DecompositionPlots import PlotDecomposition_ID83 as spd_83
-from SigProfilerAssignment.DecompositionPlots import PlotDecomposition_CNV48 as cnv_48
 from SigProfilerAssignment import decompose_subroutines as sub
 
 # imports for working with plots in memory
@@ -43,11 +44,13 @@
 import json
 import base64
 
+
 # Global Variables
 SBS_CONTEXTS = ["6", "24", "96", "288", "384", "1536", "6144"]
 DBS_CONTEXTS = ["78", "186", "1248", "2976"]
 ID_CONTEXTS = ["28", "83", "415"]
 CNV_CONTEXTS = ["48"]
+SV_CONTEXTS = ["32"]
 MTYPE_OPTIONS = [
     "6",
     "24",
@@ -62,14 +65,14 @@
     "186",
     "1248",
     "2976",
+    "32",
 ]
 DECOMPOSITION_PATH = SigProfilerAssignment.DecompositionPlots.__path__[0]
 REFSIG_PATH = os.path.join(
     SigProfilerAssignment.__path__[0], "data/Reference_Signatures"
 )
 TEMPLATE_PATH = os.path.join(DECOMPOSITION_PATH, "CosmicTemplates")
 
-
 # Remove templates so that they can be rebuilt
 def remove_cosmic_templates():
     if not os.path.exists(TEMPLATE_PATH):
@@ -84,10 +87,11 @@ def remove_cosmic_templates():
 def install_cosmic_plots(
     context_type="96", genome_build="GRCh37", cosmic_version="3.4", exome=False
 ):
+
     if not os.path.exists(TEMPLATE_PATH):
         os.mkdir(TEMPLATE_PATH)
 
-    # determine if context is from SBS, ID, DBS, or CNV
+    # determine if context is from SBS, ID, DBS, CNV or SV
     context_type_str = ""
     if context_type in SBS_CONTEXTS:
         context_type_str = "SBS"
@@ -102,6 +106,9 @@ def install_cosmic_plots(
     elif context_type in CNV_CONTEXTS:
         context_type_str = "CNV"
         cosmic_mtype = "48"
+    elif context_type in SV_CONTEXTS:
+        context_type_str = "SV"
+        cosmic_mtype = "32"
     else:
         raise ValueError("ERROR: context", context_type, "not in context lists.")
 
@@ -138,6 +145,20 @@ def install_cosmic_plots(
         genome_build = "GRCh37"
         exome_str = ""
 
+    # CNV signatures exome=False, genome_build=GRCh37
+    if context_type in CNV_CONTEXTS:
+        cosmic_file_name = "COSMIC_v" + str(cosmic_version) + "_CN_GRCh37.txt"
+        json_file_name = "COSMIC_v" + str(cosmic_version) + "_CN_GRCh37.json"
+        genome_build = "GRCh37"
+        exome_str = ""
+
+    # SV signatures exome=False, genome_build=GRCh37
+    if context_type in SV_CONTEXTS:
+        cosmic_file_name = "COSMIC_v" + str(cosmic_version) + "_SV_GRCh38.txt"
+        json_file_name = "COSMIC_v" + str(cosmic_version) + "_SV_GRCh38.json"
+        genome_build = "GRCh38"
+        exome_str = ""
+
     # Load cosmic plots if they exist
     filename = os.path.join(TEMPLATE_PATH, json_file_name)
     if os.path.exists(filename):
@@ -165,7 +186,6 @@ def install_cosmic_plots(
             + exome_str,
             "now...",
         )
-
         # Create the respective plots
         if context_type_str == "SBS":
             cosmic_buff_plots = sigPlt.plotSBS(
@@ -195,6 +215,25 @@ def install_cosmic_plots(
                 percentage=True,
                 savefig_format="PIL_Image",
             )
+        elif context_type_str == "CNV":
+            cosmic_buff_plots = sigPlt.plotCNV(
+                cosmic_file_path,
+                "buffer",
+                "buffer",
+                percentage=True,
+                aggregate=False,
+                read_from_file=False,
+                savefig_format="PIL_Image",
+            )
+        elif context_type_str == "SV":
+            cosmic_buff_plots = sigPlt.plotSV(
+                cosmic_file_path,
+                "buffer",
+                "buffer",
+                percentage=True,
+                aggregate=False,
+                savefig_format="PIL_Image",
+            )
 
         # Process the plots to be stored in JSON file
         cosmic_img_dict = {}
@@ -413,8 +452,33 @@ def genCNV_pngs(denovo_mtx, basis_mtx, output_path, project, mtype):
     return denovo_plots, basis_plots
 
 
+def genSV_pngs(denovo_mtx, basis_mtx, output_path, project, mtype):
+    denovo_plots = dict()
+    basis_plots = dict()
+    denovo_plots = sigPlt.plotSV(
+        denovo_mtx,
+        output_path,
+        project,
+        percentage=True,
+        aggregate=False,
+        savefig_format="PIL_Image",
+    )
+
+    if basis_mtx is not None:
+        basis_plots = sigPlt.plotSV(
+            basis_mtx,
+            output_path,
+            project,
+            percentage=True,
+            aggregate=False,
+            savefig_format="PIL_Image",
+        )
+    return denovo_plots, basis_plots
+
+
 # signames, weights
 def gen_sub_plots(denovo_mtx, basis_mtx, output_path, project, mtype, ss_decomp):
+
     # Make output directory
     if not os.path.exists(output_path):
         os.makedirs(output_path)
@@ -442,6 +506,11 @@ def gen_sub_plots(denovo_mtx, basis_mtx, output_path, project, mtype, ss_decomp)
             denovo_mtx, basis_mtx, output_path, project, mtype
         )
         return denovo_plots, basis_plots
+    elif mtype in SV_CONTEXTS:
+        denovo_plots, basis_plots = genSV_pngs(
+            denovo_mtx, basis_mtx, output_path, project, mtype
+        )
+        return denovo_plots, basis_plots
 
     else:
         print("ERROR: mtype is " + mtype + " and is not yet supported.")
@@ -510,6 +579,15 @@ def gen_reconstructed_png_percent(
             read_from_file=False,
             savefig_format="PIL_Image",
         )
+    elif mtype in SV_CONTEXTS:
+        reconstruction_plot = sigPlt.plotSV(
+            reconstruction_mtx,
+            output_path,
+            "reconstruction_" + project,
+            percentage=True,
+            aggregate=False,
+            savefig_format="PIL_Image",
+        )
     else:
         print("ERROR: mtype is " + mtype + " and is not yet supported.")
 
@@ -591,6 +669,16 @@ def gen_reconstructed_png_numerical(
             read_from_file=False,
             savefig_format="PIL_Image",
         )
+    elif mtype in SV_CONTEXTS:
+        reconstruction_plot = sigPlt.plotSV(
+            reconstruction_mtx,
+            output_path,
+            "reconstruction_" + project,
+            percentage=True,
+            aggregate=False,
+            read_from_file=False,
+            savefig_format="PIL_Image",
+        )
     else:
         print("ERROR: mtype is " + mtype + " and is not yet supported.")
 
@@ -612,6 +700,7 @@ def gen_decomposition(
     cosmic_version=None,
     custom_text=None,
 ):
+
     """
     Generate the correct plot based on mtype.
 
@@ -744,6 +833,22 @@ def gen_decomposition(
             custom_text,
         )
         return byte_plot
+    elif mtype == "32":
+        byte_plot = sv_32.gen_decomposition(
+            denovo_name,
+            basis_names,
+            weights,
+            output_path,
+            project,
+            denovo_plots_dict,
+            basis_plots_dict,
+            reconstruction_plot_dict,
+            reconstruction,
+            statistics,
+            cosmic_version,
+            custom_text,
+        )
+        return byte_plot
 
 
 def run_PlotDecomposition(
@@ -773,8 +878,7 @@ def run_PlotDecomposition(
 
     basis_mtx: Pandas Dataframe. This format represents the catalog of mutations seperated by tab.
 
-    basis_names: List of Strings. The names of the samples in denovo_mtx that
-    the denovo_name sample from denovo_mtx is decomposed into.
+    basis_names: List of Strings. The names of the samples in denovo_mtx that the denovo_name sample from denovo_mtx is decomposed into.
     ie. basis_names=["SBS1", "SBS5", "SBS15", "SBS20"]
 
     weights: List of Strings. The percentile weight corresponding to each basis
@@ -796,17 +900,9 @@ def run_PlotDecomposition(
     None.
     """
     # Create the denovo plots and load basis plots
-    if mtype != "48":
-        denovo_plots_dict = gen_sub_plots(
-            denovo_mtx, None, output_path, project, mtype, ss_decomp=False
-        )
-        denovo_plots_dict = denovo_plots_dict[0]
-    else:
-        # cnv basis plots need to be generated and not loaded
-        denovo_plots_dict, basis_plots_dict = gen_sub_plots(
-            denovo_mtx, basis_mtx, output_path, project, mtype, ss_decomp=False
-        )
-    # Create the matrix and plot for the reconstructed matrix
+    denovo_plots_dict, basis_plots_dict = gen_sub_plots(
+        denovo_mtx, basis_mtx, output_path, project, mtype, ss_decomp=False
+    )
     reconstructed_mtx, reconstruction_plot_dict = gen_reconstructed_png_percent(
         denovo_name, basis_mtx, basis_names, weights, output_path, project, mtype
     )
@@ -816,14 +912,13 @@ def run_PlotDecomposition(
     # Convert dictionary of bytes to dictionary of images
     denovo_plots_dict = convert_to_imgReaderDict(denovo_plots_dict)
     # Load in the COSMIC plots
-    if mtype != "48":
-        basis_plots_dict = install_cosmic_plots(
-            context_type=mtype,
-            genome_build=genome_build,
-            cosmic_version=cosmic_version,
-            exome=exome,
-        )
-        basis_plots_dict = {key: basis_plots_dict[key] for key in basis_names}
+    basis_plots_dict = install_cosmic_plots(
+        context_type=mtype,
+        genome_build=genome_build,
+        cosmic_version=cosmic_version,
+        exome=exome,
+    )
+    basis_plots_dict = {key: basis_plots_dict[key] for key in basis_names}
     basis_plots_dict = convert_to_imgReaderDict(basis_plots_dict)
     # Generate the reconstruction plot
     reconstruction_plot_dict = convert_to_imgReaderDict(reconstruction_plot_dict)