Windows pipeline support for BLRunner and BLEvaluator

Algorithms/ARBORETO/Dockerfile

@@ -1,6 +1,3 @@
-
-# FROM continuumio/anaconda3:2018.12
-
 FROM continuumio/anaconda3:2024.02-1
 
 LABEL Maintainer="Aditya Pratapa <adyprat@vt.edu>"
@@ -9,8 +6,6 @@ USER root
 
 RUN apt-get update
 
-# RUN conda install -y -c bioconda/label/cf201901 arboreto=0.1.5 pandas=0.24.0
-
 RUN conda install -y -c bioconda arboreto pandas
 
 COPY runArboreto.py /

Algorithms/PIDC/Dockerfile

@@ -1,4 +1,3 @@
-# FROM julia:1.1.0-stretch
 FROM julia:1.6.0-buster
 
 LABEL maintainer="Aditya Pratapa <adyprat@vt.edu>"

Algorithms/SCSGL/Dockerfile

@@ -1,29 +1,29 @@
 #This is the base image upon which necessary libraries are installed
-FROM python:3.8  
+FROM python:3.8
 
 #Additional information
-LABEL Maintainer="Yiqi Su <yiqisu@vt.edu>" 
+LABEL Maintainer="Yiqi Su <yiqisu@vt.edu>"
 
 #Set main user as root to avoid permission issues
-USER root 
+USER root
 
 #Sets current working directory
-WORKDIR / 
+WORKDIR /
 
 #Copy the main Python script which will perform necessary GRN computations
-COPY run_scSGL.py /  
+COPY run_scSGL.py /
 
 #Copy the original scSGL repo files stored in SCSGL folder
-COPY scSGL /scSGL   
+COPY scSGL /scSGL
 
 #Installl time command to compute time taken
-RUN apt-get update && apt-get install -y r-base time  
+RUN apt-get update && apt-get install -y r-base time
 
 #Install the requirments and install R to conda environment
-RUN pip install -r /scSGL/requirements.txt 
+RUN pip install -r /scSGL/requirements.txt
 
 #Install pcaPP to use zero inflated Kendall tau as a kernel
-RUN Rscript -e "install.packages('pcaPP')" 
+RUN Rscript -e "install.packages('pcaPP')"
 
 #Make a directory to mount the folder containing input files
-RUN mkdir data/ 
+RUN mkdir data/

BLEval/computeDGAUC.py

@@ -1,3 +1,4 @@
+import os
 import pandas as pd
 import numpy as np
 import seaborn as sns
@@ -45,7 +46,7 @@ def PRROC(dataDict, inputSettings, directed = True, selfEdges = False, plotFlag
     AUROC = {}
 
     # set-up outDir that stores output directory name
-    outDir = "outputs/"+str(inputSettings.datadir).split("inputs/")[1]+ '/' +dataDict['name']
+    outDir = "outputs/"+"/".join(str(inputSettings.datadir).split("inputs" + os.sep)[1].split(os.sep))+ '/' +dataDict['name']
 
     if directed:
         for algo in tqdm(inputSettings.algorithms, 
@@ -214,6 +215,9 @@ def computeScores(trueEdgesDF, predEdgeDF,
     # to pass it to sklearn
     outDF = pd.DataFrame([TrueEdgeDict,PredEdgeDict]).T
     outDF.columns = ['TrueEdges','PredEdges']
+    utils = importr('utils')
+    utils.chooseCRANmirror(ind=1)
+    # utils.install_packages('PRROC')
     prroc = importr('PRROC')
     prCurve = prroc.pr_curve(scores_class0 = FloatVector(list(outDF['PredEdges'].values)), 
               weights_class0 = FloatVector(list(outDF['TrueEdges'].values)))

BLEval/computeNetMotifs.py

@@ -9,6 +9,7 @@
 from itertools import product, permutations, combinations, combinations_with_replacement
 from tqdm import tqdm
 import networkx as nx
+import os
 
 def Motifs(datasetDict, inputSettings):
     '''
@@ -68,7 +69,8 @@ def Motifs(datasetDict, inputSettings):
         refMI = 1
 
     # set-up outDir that stores output directory name
-    outDir = "outputs/"+str(inputSettings.datadir).split("inputs/")[1]+ '/' + datasetDict['name']
+    outDir = "outputs/"+str(inputSettings.datadir).split("inputs" + os.sep)[1] + '/' + datasetDict['name']
+    print(f"\n\nThe output directory is: {outDir}\n\n")
     dataDict = {}
     # dataDict['Conn. Comp'] = {}
     dataDict['FFL'] = {}

BLEval/computePathStats.py

@@ -5,6 +5,7 @@
 from pathlib import Path
 import matplotlib.pyplot as plt
 import seaborn as sns
+import os
 sns.set(rc={"lines.linewidth": 2}, palette  = "deep", style = "ticks")
 from sklearn.metrics import precision_recall_curve, roc_curve, auc
 from itertools import product, permutations, combinations, combinations_with_replacement
@@ -40,7 +41,7 @@ def pathAnalysis(dataDict, inputSettings):
 
 
     # set-up outDir that stores output directory name
-    outDir = "outputs/"+str(inputSettings.datadir).split("inputs/")[1]+ '/' +dataDict['name']
+    outDir = "outputs/"+str(inputSettings.datadir).split("inputs" + os.sep)[1]+ '/' +dataDict['name']
     #print(dataDict['name'])
 
     ##################################################

BLEval/parseTime.py

@@ -135,4 +135,8 @@ def parse_time_files(path):
         print("Algorithm running failed, setting time value to -1\n")
         time_val = -1
 
+    except IndexError:
+        print(f"Time output {path} file found but not populated, setting time value to -1\n")
+        time_val = -1
+
     return time_val

BLPlotter.py

@@ -76,8 +76,8 @@ def boxplot(opts, evalConfigs, datasets, randValue, resTypeFile, resTypeName):
         evalConfig = evalConfigs[i]
 
         # Read output file containing AUROC values
-        DF = pd.read_csv(str(evalConfig.output_settings.base_dir) + '/' \
-                           + str(evalConfig.input_settings.datadir).split("inputs")[1] + '/' \
+        DF = pd.read_csv(str(evalConfig.output_settings.base_dir) + "/" \
+                           + "/".join(str(evalConfig.input_settings.datadir).split("inputs" + os.sep)[1].split(os.sep)) + '/' \
                            + str(evalConfig.output_settings.output_prefix) \
                            + '-' + resTypeFile + '.csv', header = 0, index_col = 0)
 
@@ -465,8 +465,8 @@ def main():
             for j, dataset in enumerate(datasets):
                 evalConfig = evalConfigs[j]
 
-                ResDF = pd.read_csv(str(evalConfig.output_settings.base_dir) + '/' \
-                                    + str(evalConfig.input_settings.datadir).split("inputs")[1] + '/' \
+                ResDF = pd.read_csv(str(evalConfig.output_settings.base_dir) \
+                                    + "/".join(str(evalConfig.input_settings.datadir).split("inputs")[1].split(os.sep)) + '/' \
                                     + str(evalConfig.output_settings.output_prefix) \
                                     + '-' + resTypeFileName[i] + '.csv', header = 0, index_col = 0)
 

BLRun/genie3Runner.py

@@ -1,6 +1,7 @@
 import os
 import pandas as pd
 from pathlib import Path
+from BLRun.out_path_generator import get_output_path
 import numpy as np
 
 def generateInputs(RunnerObj):
@@ -30,10 +31,10 @@ def run(RunnerObj):
 
     :param RunnerObj: An instance of the :class:`BLRun`
     '''
-    inputPath = "data" + str(RunnerObj.inputDir).split(str(Path.cwd()))[1] + \
+    inputPath = "data" + "/".join(str(RunnerObj.inputDir).split(str(Path.cwd()))[1].split(os.sep)) + \
                     "/GENIE3/ExpressionData.csv"
     # make output dirs if they do not exist:
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/GENIE3/"
+    outDir = get_output_path(RunnerObj, "/GENIE3/")
     os.makedirs(outDir, exist_ok = True)
 
     outPath = "data/" +  str(outDir) + 'outFile.txt'
@@ -53,7 +54,7 @@ def parseOutput(RunnerObj):
     :param RunnerObj: An instance of the :class:`BLRun`
     '''
     # Quit if output directory does not exist
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/GENIE3/"
+    outDir = get_output_path(RunnerObj, "/GENIE3/")
 
 
     # Read output

BLRun/grisliRunner.py

@@ -2,6 +2,7 @@
 import pandas as pd
 from pathlib import Path
 import numpy as np
+from BLRun.out_path_generator import get_output_path
 
 def generateInputs(RunnerObj):
     '''
@@ -47,15 +48,15 @@ def run(RunnerObj):
     alphaMin = str(RunnerObj.params['alphaMin'])
 
     # make output dirs if they do not exist:
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/GRISLI/"
+    outDir = get_output_path(RunnerObj, "/GRISLI/")
     os.makedirs(outDir, exist_ok = True)
 
     PTData = pd.read_csv(RunnerObj.inputDir.joinpath(RunnerObj.cellData),
                              header = 0, index_col = 0)
 
     colNames = PTData.columns
     for idx in range(len(colNames)):
-        inputPath = "data"+str(RunnerObj.inputDir).split(str(Path.cwd()))[1]+"/GRISLI/"+str(idx)+"/"
+        inputPath = "data"+"/".join(str(RunnerObj.inputDir).split(str(Path.cwd()))[1].split(os.sep))+"/GRISLI/"+str(idx)+"/"
         os.makedirs(outDir+str(idx), exist_ok = True)
 
         outFile = "data/" +  str(outDir) +str(idx)+"/outFile.txt"
@@ -72,7 +73,7 @@ def parseOutput(RunnerObj):
     Function to parse outputs from GRISLI.
     '''
 
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/GRISLI/"
+    outDir = get_output_path(RunnerObj, "/GRISLI/")
 
     PTData = pd.read_csv(RunnerObj.inputDir.joinpath(RunnerObj.cellData),
                              header = 0, index_col = 0)

BLRun/grnboost2Runner.py

@@ -1,6 +1,7 @@
 import os
 import pandas as pd
 from pathlib import Path
+from BLRun.out_path_generator import get_output_path
 import numpy as np
 
 def generateInputs(RunnerObj):
@@ -25,10 +26,10 @@ def run(RunnerObj):
     '''
     Function to run GRNBOOST2 algorithm
     '''
-    inputPath = "data" + str(RunnerObj.inputDir).split(str(Path.cwd()))[1] + \
+    inputPath = "data" + "/".join(str(RunnerObj.inputDir).split(str(Path.cwd()))[1].split(os.sep)) + \
                     "/GRNBOOST2/ExpressionData.csv"
     # make output dirs if they do not exist:
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/GRNBOOST2/"
+    outDir = get_output_path(RunnerObj, "/GRNBOOST2/")
     os.makedirs(outDir, exist_ok = True)
 
 
@@ -46,7 +47,7 @@ def parseOutput(RunnerObj):
     Function to parse outputs from GRNBOOST2.
     '''
     # Quit if output directory does not exist
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/GRNBOOST2/"
+    outDir = get_output_path(RunnerObj, "/GRNBOOST2/")
 
     if not Path(outDir+'outFile.txt').exists():
         print(outDir+'outFile.txt'+'does not exist, skipping...')

BLRun/grnvbemRunner.py

@@ -1,6 +1,7 @@
 import os
 import pandas as pd
 from pathlib import Path
+from BLRun.out_path_generator import get_output_path
 import numpy as np
 
 def generateInputs(RunnerObj):
@@ -46,9 +47,9 @@ def run(RunnerObj):
     Function to run GRN-VBEM algorithm
     '''
 
-    inputPath = "data" + str(RunnerObj.inputDir).split(str(Path.cwd()))[1]
+    inputPath = "data" + "/".join(str(RunnerObj.inputDir).split(str(Path.cwd()))[1].split(os.sep))
     # make output dirs if they do not exist:
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/GRNVBEM/"
+    outDir = get_output_path(RunnerObj, "/GRNVBEM/")
     os.makedirs(outDir, exist_ok = True)
 
     PTData = pd.read_csv(RunnerObj.inputDir.joinpath(RunnerObj.cellData),
@@ -72,7 +73,7 @@ def parseOutput(RunnerObj):
     '''
     Function to parse outputs from GRNVBEM.
     '''
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/GRNVBEM/"
+    outDir = get_output_path(RunnerObj, "/GRNVBEM/")
 
     PTData = pd.read_csv(RunnerObj.inputDir.joinpath(RunnerObj.cellData),
                              header = 0, index_col = 0)

BLRun/jump3Runner.py

@@ -2,6 +2,7 @@
 import pandas as pd
 from pathlib import Path
 import numpy as np
+from BLRun.out_path_generator import get_output_path
 from sklearn import preprocessing
 
 def generateInputs(RunnerObj):
@@ -42,11 +43,11 @@ def run(RunnerObj):
     '''
     Function to run GRN-VBEM algorithm
     '''
-    inputPath = "data" + str(RunnerObj.inputDir).split(str(Path.cwd()))[1] + \
+    inputPath = "data" + "/".join(str(RunnerObj.inputDir).split(str(Path.cwd()))[1].split(os.sep)) + \
                     "/JUMP3/ExpressionData.csv"
 
     # make output dirs if they do not exist:
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/JUMP3/"
+    outDir = get_output_path(RunnerObj, "/JUMP3/")
     os.makedirs(outDir, exist_ok = True)
 
     outPath = "data/" +  str(outDir) + 'outFile.txt'
@@ -62,7 +63,7 @@ def parseOutput(RunnerObj):
     Function to parse outputs from JUMP3.
     '''
     # Quit if output directory does not exist
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/JUMP3/"
+    outDir = get_output_path(RunnerObj, "/JUMP3/")
     if not Path(outDir+'outFile.txt').exists():
         print(outDir+'outFile.txt'+'does not exist, skipping...')
         return

BLRun/leapRunner.py

@@ -2,6 +2,7 @@
 import pandas as pd
 from pathlib import Path
 import numpy as np
+from BLRun.out_path_generator import get_output_path
 
 def generateInputs(RunnerObj):
     '''
@@ -44,12 +45,12 @@ def run(RunnerObj):
     Requires the maxLag parameter
     '''
 
-    inputPath = "data" + str(RunnerObj.inputDir).split(str(Path.cwd()))[1]
+    inputPath = "data" + "/".join(str(RunnerObj.inputDir).split(str(Path.cwd()))[1].split(os.sep))
 
     maxLag = str(RunnerObj.params['maxLag'])
 
     # make output dirs if they do not exist:
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/LEAP/"
+    outDir = get_output_path(RunnerObj, "/LEAP/")
     os.makedirs(outDir, exist_ok = True)
 
     PTData = pd.read_csv(RunnerObj.inputDir.joinpath(RunnerObj.cellData),
@@ -74,7 +75,7 @@ def parseOutput(RunnerObj):
     '''
     Function to parse outputs from LEAP.
     '''
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/LEAP/"
+    outDir = get_output_path(RunnerObj, "/LEAP/")
 
     PTData = pd.read_csv(RunnerObj.inputDir.joinpath(RunnerObj.cellData),
                              header = 0, index_col = 0)

BLRun/out_path_generator.py

@@ -0,0 +1,5 @@
+import os
+
+
+def get_output_path(runner_obj, algo_name):
+    return "outputs/"+"/".join(str(runner_obj.inputDir).split("inputs" + os.sep)[1].split(os.sep))+algo_name

BLRun/pidcRunner.py

@@ -1,6 +1,7 @@
 import os
 import pandas as pd
 from pathlib import Path
+from BLRun.out_path_generator import get_output_path
 import numpy as np
 
 def generateInputs(RunnerObj):
@@ -23,11 +24,11 @@ def run(RunnerObj):
     '''
     Function to run PIDC algorithm
     '''
-    inputPath = "data" + str(RunnerObj.inputDir).split(str(Path.cwd()))[1] + \
+    inputPath = "data" + "/".join(str(RunnerObj.inputDir).split(str(Path.cwd()))[1].split(os.sep)) + \
                     "/PIDC/ExpressionData.csv"
 
     # make output dirs if they do not exist:
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/PIDC/"
+    outDir = get_output_path(RunnerObj, "/PIDC/")
     os.makedirs(outDir, exist_ok = True)
 
     outPath = 'data/'+ str(outDir) + 'outFile.txt'
@@ -43,7 +44,7 @@ def parseOutput(RunnerObj):
     Function to parse outputs from SCODE.
     '''
     # Quit if output directory does not exist
-    outDir = "outputs/"+str(RunnerObj.inputDir).split("inputs/")[1]+"/PIDC/"
+    outDir = get_output_path(RunnerObj, "/PIDC/")
     if not Path(outDir+'outFile.txt').exists():
         print(outDir+'outFile.txt'+'does not exist, skipping...')
         return