Merge pull request #48 from AlexandrovLab/plots1

import decomp plots from assignment, update MANIFEST, fix column name…

Author: mdbarnesUCSD

.gitignore

@@ -51,7 +51,13 @@ coverage.xml
 .hypothesis/
 .pytest_cache/
 cover/
-test/
+SigProfilerAssignment/data/tests/vcf_input/input/
+SigProfilerAssignment/data/tests/vcf_input/logs/
+SigProfilerAssignment/data/tests/vcf_input/output/
+output_vcf_example/
+output_txt_example/
+
+
 
 # Translations
 *.mo
@@ -84,24 +90,6 @@ target/
 profile_default/
 ipython_config.py
 
-# pyenv
-#   For a library or package, you might want to ignore these files since the code is
-#   intended to run in multiple environments; otherwise, check them in:
-# .python-version
-
-# pipenv
-#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
-#   However, in case of collaboration, if having platform-specific dependencies or dependencies
-#   having no cross-platform support, pipenv may install dependencies that don't work, or not
-#   install all needed dependencies.
-#Pipfile.lock
-
-# poetry
-#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
-#   This is especially recommended for binary packages to ensure reproducibility, and is more
-#   commonly ignored for libraries.
-#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
-#poetry.lock
 
 # pdm
 #   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
@@ -154,9 +142,3 @@ dmypy.json
 # Cython debug symbols
 cython_debug/
 
-# PyCharm
-#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
-#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
-#  and can be added to the global gitignore or merged into this file.  For a more nuclear
-#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
-#.idea/

.travis.yml

@@ -4,7 +4,19 @@ language: python
 python:
   - '3.8'
 
+before_install:
+  if ! [ -f ./src/GRCh37.tar.gz ]; then
+    wget ftp://alexandrovlab-ftp.ucsd.edu/pub/tools/SigProfilerMatrixGenerator/GRCh37.tar.gz -P ./src/;
+  fi
+
 install:
   - pip install .
 
+cache:
+  directories:
+    - $TRAVIS_BUILD_DIR/src/
+
+before_script:
+  - python3 install_genome.py $TRAVIS_BUILD_DIR/src/
+
 script: python3 test.py

MANIFEST.in

@@ -7,4 +7,8 @@ include SigProfilerAssignment/data/Reference_Signatures/mm9/*
 include SigProfilerAssignment/data/Reference_Signatures/mm10/*
 include SigProfilerAssignment/data/Reference_Signatures/rn6/*
 include SigProfilerAssignment/DecompositionPlots/reference_files/Fonts/*
-include SigProfilerAssignment/DecompositionPlots/reference_files/*
+include SigProfilerAssignment/DecompositionPlots/reference_files/*
+include SigProfilerAssignment/DecompositionPlots/*
+include SigProfilerAssignment/data/tests/vcf_input/*
+include SigProfilerAssignment/data/tests/txt_input/*
+include SigProfilerAssignment/data/tests/cnv_input/*

README.md

@@ -180,13 +180,17 @@ Analyze.cosmic_fit( samples,
 #import modules
 import SigProfilerAssignment as spa
 from SigProfilerAssignment import Analyzer as Analyze
+import os
 
 #set directories and paths to signatures and samples
-dir_inp     = spa.__path__[0]+'/data/Examples/'
-samples = spa.__path__[0]+'/data/vcftest/' #directory of vcf files
-output      = "output_example/"
-signatures  = dir_inp+"Results_scenario_8/SBS96/All_Solutions/SBS96_3_Signatures/Signatures/SBS96_S3_Signatures.txt"
-sigs        = "COSMIC_v3_SBS_GRCh37_noSBS84-85.txt" #Custom Signature Database
+dir_inp = os.path.join(spa.__path__[0], '/data/Examples/')
+# directory of vcf files
+samples = os.path.join(spa.__path__[0], '/data/tests/vcf_input/')
+output = "output_example/"
+signatures = os.path.join(dir_inp, \
+    "Results_scenario_8/SBS96/All_Solutions/SBS96_3_Signatures/Signatures/" \
+    + "SBS96_S3_Signatures.txt")
+sigs = "COSMIC_v3_SBS_GRCh37_noSBS84-85.txt" #Custom Signature Database
 
 #Analysis of SP Assignment 
 Analyze.cosmic_fit( samples, 
@@ -212,11 +216,14 @@ Analyze.cosmic_fit( samples,
 #import modules
 import SigProfilerAssignment as spa
 from SigProfilerAssignment import Analyzer as Analyze
+import os
 
 #set directories and paths to signatures and samples
-dir_inp     = spa.__path__[0]+'/data/Examples/'
-samples = spa.__path__[0]+'/data/cnvtest/all.breast.ascat.summary.sample.tsv' # segmentation file
-output      = "output_example/"
+dir_inp = os.path.join(spa.__path__[0], 'data/Examples/')
+# segmentation file
+samples = os.path.join(spa.__path__[0], \
+    '/data/tests/cnv_input/all.breast.ascat.summary.sample.tsv')
+output = "output_example/"
 
 #Analysis of SP Assignment 
 Analyze.cosmic_fit( samples, 

SigProfilerAssignment/DecompositionPlots/PlotDecomposition.py

@@ -202,8 +202,8 @@ def gen_reconstructed_png(denovo_name, basis_mtx, basis_names, weights, output_p
 	elif mtype in ID_CONTEXTS:
 		reconstruction_plot=sigPlt.plotID(reconstruction_mtx, output_path, "reconstruction_" + project, mtype, True)
 	elif mtype in CNV_CONTEXTS:
-		 reconstruction_plot = pltCNV.plotCNV(reconstruction_mtx, output_path, "reconstruction_"+project, plot_type="pdf", \
-		 	percentage=True, aggregate=False, read_from_file=False, write_to_file=False)
+		reconstruction_plot = pltCNV.plotCNV(reconstruction_mtx, output_path, "reconstruction_"+project, plot_type="pdf", \
+			percentage=True, aggregate=False, read_from_file=False, write_to_file=False)
 	else:
 		print("ERROR: mtype is " + mtype + " and is not yet supported.")
 
@@ -323,14 +323,11 @@ def run_PlotDecomposition(denovo_mtx, denovo_name, basis_mtx, basis_names, \
 	denovo_plots_dict,basis_plots_dict=gen_sub_plots(denovo_mtx, basis_mtx, output_path, project, mtype)
 	reconstructed_mtx,reconstruction_plot_dict = gen_reconstructed_png(denovo_name, basis_mtx, basis_names, \
 		weights, output_path, project, mtype)
-	
 	present_sigs=np.array(basis_mtx[basis_names])
 	reconstructed_mtx = np.dot(present_sigs,nonzero_exposures)
-	
 	denovo_plots_dict = open_byte_to_img_dict(denovo_plots_dict)
 	basis_plots_dict = open_byte_to_img_dict(basis_plots_dict)
 	reconstruction_plot_dict = open_byte_to_img_dict(reconstruction_plot_dict)
-	
 	statistics=calculate_similarities(denovo_mtx, denovo_name, reconstructed_mtx)
 	byte_plot = gen_decomposition(denovo_name, basis_names, weights, output_path, project, \
 		mtype, denovo_plots_dict, basis_plots_dict, reconstruction_plot_dict, \

SigProfilerAssignment/DecompositionPlots/PlotDecomposition_CNV48.py

@@ -17,9 +17,8 @@
 from PIL import Image
 # imports for dashed line
 from reportlab.lib.colors import black
-paths = spa_path.__path__[0]
-# Page Formatting
-inch = 72
+PATHS = spa_path.__path__[0]
+
 # USING LETTER PORTRAIT DIMENSIONS
 WIDTH_LETTER = 965
 HEIGHT_LETTER = 755
@@ -44,8 +43,8 @@
 TEXT_X_COORD = GRAPH_X_COORD + WIDTH_GRAPH - 50
 TEXT_Y_COORD = (HEIGHT_LETTER - HEIGHT_GAP) + 55.75
 
-FONTS_DIR = os.path.join(paths+'decomposition_plots/reference_files/Fonts/')
-BRACKET_PATH=os.path.join(paths+"/decomposition_plots/reference_files/Accolade_fermante.png")
+FONTS_DIR = os.path.join(PATHS, "DecompositionPlots/reference_files/Fonts/")
+BRACKET_PATH=os.path.join(PATHS, "DecompositionPlots/reference_files/Accolade_fermante.png")
 reportlab.rl_config.TTFSearchPath.append(FONTS_DIR)
 pdfmetrics.registerFont(TTFont('Arial-Bold', 'Arial Bold.ttf'))
 
@@ -217,8 +216,6 @@ def draw_statistics(c_draw, statistics, sig_version, custom_text):
 #	num_bases 	- (Integer) The number of bases to be plotted.
 #	c_draw 		- (Canvas) The canvas to draw the graph decomposition on.
 def draw_bracket(num_bases, c_draw):
-    paths = cosmic.__path__[0]
-
     num_plts = num_bases - 1
 
     if(num_bases >= 5):

SigProfilerAssignment/DecompositionPlots/PlotDecomposition_DBS78.py

@@ -17,12 +17,12 @@
 from PIL import Image
 # imports for dashed line
 from reportlab.lib.colors import black
-paths = spa_path.__path__[0]
+PATHS = spa_path.__path__[0]
 # Page Formatting
-inch = 72
+INCH = 72
 # USING LETTER LANDSCAPE DIMENSIONS
-WIDTH_LETTER = 11 * inch
-HEIGHT_LETTER = 8.5 * inch
+WIDTH_LETTER = 11 * INCH
+HEIGHT_LETTER = 8.5 * INCH
 MID_WIDTH_LETTER = 396
 MID_HEIGHT_LETTER = HEIGHT_LETTER/2
 
@@ -41,8 +41,8 @@
 TEXT_X_COORD = GRAPH_X_COORD + WIDTH_GRAPH - 50
 TEXT_Y_COORD = (HEIGHT_LETTER - HEIGHT_GAP) + 63.75
 
-FONTS_DIR = os.path.join(paths+'decomposition_plots/reference_files/Fonts/')
-BRACKET_PATH=os.path.join(paths+"/decomposition_plots/reference_files/Accolade_fermante.png")
+FONTS_DIR = os.path.join(PATHS, "DecompositionPlots/reference_files/Fonts/")
+BRACKET_PATH=os.path.join(PATHS, "DecompositionPlots/reference_files/Accolade_fermante.png")
 reportlab.rl_config.TTFSearchPath.append(FONTS_DIR)
 pdfmetrics.registerFont(TTFont('Arial-Bold', 'Arial Bold.ttf'))
 

SigProfilerAssignment/DecompositionPlots/PlotDecomposition_ID83.py

@@ -17,12 +17,12 @@
 from PIL import Image
 # imports for dashed line
 from reportlab.lib.colors import black
-paths = spa_path.__path__[0]
+PATHS = spa_path.__path__[0]
 # Page Formatting
-inch = 72
+INCH = 72
 # USING LETTER LANDSCAPE DIMENSIONS
-WIDTH_LETTER = 11 * inch
-HEIGHT_LETTER = 8.5 * inch
+WIDTH_LETTER = 11 * INCH
+HEIGHT_LETTER = 8.5 * INCH
 MID_WIDTH_LETTER = 396
 MID_HEIGHT_LETTER = HEIGHT_LETTER/2
 
@@ -41,8 +41,8 @@
 TEXT_X_COORD = GRAPH_X_COORD + WIDTH_GRAPH - 50
 TEXT_Y_COORD = (HEIGHT_LETTER - HEIGHT_GAP) + 40.75
 
-FONTS_DIR = os.path.join(paths+'decomposition_plots/reference_files/Fonts/')
-BRACKET_PATH=os.path.join(paths+"/decomposition_plots/reference_files/Accolade_fermante.png")
+FONTS_DIR = os.path.join(PATHS, "DecompositionPlots/reference_files/Fonts/")
+BRACKET_PATH=os.path.join(PATHS, "DecompositionPlots/reference_files/Accolade_fermante.png")
 reportlab.rl_config.TTFSearchPath.append(FONTS_DIR)
 pdfmetrics.registerFont(TTFont('Arial-Bold', 'Arial Bold.ttf'))
 

SigProfilerAssignment/DecompositionPlots/PlotDecomposition_SBS1536.py

@@ -17,12 +17,12 @@
 from PIL import Image
 # imports for dashed line
 from reportlab.lib.colors import black
-paths = spa_path.__path__[0]
+PATHS = spa_path.__path__[0]
 # Page Formatting
-inch = 72
+INCH = 72
 # USING LETTER LANDSCAPE DIMENSIONS
-WIDTH_LETTER = 11 * inch
-HEIGHT_LETTER = 8.5 * inch
+WIDTH_LETTER = 11 * INCH
+HEIGHT_LETTER = 8.5 * INCH
 MID_WIDTH_LETTER = 396
 MID_HEIGHT_LETTER = HEIGHT_LETTER/2
 
@@ -41,8 +41,8 @@
 TEXT_X_COORD = GRAPH_X_COORD + WIDTH_GRAPH - 50
 TEXT_Y_COORD = (HEIGHT_LETTER - HEIGHT_GAP) + 63.75
 
-FONTS_DIR = os.path.join(paths+'decomposition_plots/reference_files/Fonts/')
-BRACKET_PATH=os.path.join(paths+"/decomposition_plots/reference_files/Accolade_fermante.png")
+FONTS_DIR = os.path.join(PATHS, "DecompositionPlots/reference_files/Fonts/")
+BRACKET_PATH=os.path.join(PATHS, "DecompositionPlots/reference_files/Accolade_fermante.png")
 reportlab.rl_config.TTFSearchPath.append(FONTS_DIR)
 pdfmetrics.registerFont(TTFont('Arial-Bold', 'Arial Bold.ttf'))
 

SigProfilerAssignment/DecompositionPlots/PlotDecomposition_SBS288.py

@@ -17,12 +17,12 @@
 from PIL import Image
 # imports for dashed line
 from reportlab.lib.colors import black
-paths = spa_path.__path__[0]
+PATHS = spa_path.__path__[0]
 # Page Formatting
-inch = 72
+INCH = 72
 # USING LETTER LANDSCAPE DIMENSIONS
-WIDTH_LETTER = 11 * inch
-HEIGHT_LETTER = 8.5 * inch
+WIDTH_LETTER = 11 * INCH
+HEIGHT_LETTER = 8.5 * INCH
 MID_WIDTH_LETTER = 396
 MID_HEIGHT_LETTER = HEIGHT_LETTER/2
 
@@ -41,8 +41,8 @@
 TEXT_X_COORD = GRAPH_X_COORD + WIDTH_GRAPH - 50
 TEXT_Y_COORD = (HEIGHT_LETTER - HEIGHT_GAP) + 63.75
 
-FONTS_DIR = os.path.join(paths+'decomposition_plots/reference_files/Fonts/')
-BRACKET_PATH=os.path.join(paths+"/decomposition_plots/reference_files/Accolade_fermante.png")
+FONTS_DIR = os.path.join(PATHS, "DecompositionPlots/reference_files/Fonts/")
+BRACKET_PATH=os.path.join(PATHS, "DecompositionPlots/reference_files/Accolade_fermante.png")
 reportlab.rl_config.TTFSearchPath.append(FONTS_DIR)
 pdfmetrics.registerFont(TTFont('Arial-Bold', 'Arial Bold.ttf'))
 #print(pdfmetrics.sorted_list())
@@ -201,8 +201,6 @@ def draw_statistics(c_draw, statistics, sig_version, custom_text):
 #	num_bases 	- (Integer) The number of bases to be plotted.
 #	c_draw 		- (Canvas) The canvas to draw the graph decomposition on.
 def draw_bracket(num_bases, c_draw):
-    paths = cosmic.__path__[0]
-
     num_plts = num_bases - 1
 
     if(num_bases >= 5):