Merge pull request #11 from Zymo-Research/fixLoggingPathBug

Fix logging path bug

Author: michael-weinstein

README.md

@@ -144,8 +144,8 @@ FIGARO will analyze an entire directory of FASTQ files on the system. Run the fo
 
 ```
 
-import figaro
-resultTable, forwardCurve, reverseCurve = figaro.figaro.runAnalysis(sequenceFolder, minCombinedReadLength, forwardPrimerLength, reversePrimerLength, fileNamingStandard, trimParameterDownsample, trimParameterPercentile)
+from figaro import figaro
+resultTable, forwardCurve, reverseCurve = figaro.runAnalysis(sequenceFolder, ampliconLength, forwardPrimerLength, reversePrimerLength, minimumOverlap, fileNamingStandard, trimParameterDownsample, trimParameterPercentile)
 ```
 
 |Parameter        | Type           | Default  | Description |
@@ -154,6 +154,7 @@ sequenceFolder| string | **REQUIRED** | The folder containing the sequences to a
 minimumCombinedReadLength| integer | **REQUIRED** | The length of the amplified sequence target plus overlap **not including primers**. User is required to set this.
 forwardPrimerLength | integer | **REQUIRED** | The length of the forward primer. User is required to set this.
 reversePrimerLength | integer | **REQUIRED** | The length of the reverse primer. User is required to set this.
+minimumOverlap | integer | 20 | The minimum length of overlap desired for read merging
 fileNamingStandard | string | illumina | Naming convention for files. Currently supporting Illumina and Zymo Services (zymo). Others can be added as requested.--outputFileName | -n | string | trimParameters.json | The desired name of the JSON list of trim parameters and their scores
 subsample | integer | *See description* | What fraction of reads to analyze (1/x) from the FASTQ files. Default value will call a function that sets this based upon the size of the fastq files for a sliding scale.
 percentile | integer | 83 | The percentile to target for read filtering.  The default value of 83 will remove reads that are about 1 standard deviation worse than the average read for that direction in that position. You can generally expect a few percentage points below your percentile value of reads to pass the filtering.

figaro.py

@@ -1,5 +1,8 @@
 import logging
-from . import figaroSupport
+try:  #This block is here to handle importing issues that happen when running this as a python package vs. running in a Docker or directly from the commandline.
+    import figaroSupport
+except ImportError:
+    from . import figaroSupport
 
 
 def getApplicationParameters():
@@ -32,10 +35,9 @@ def getApplicationParameters():
     return parameters
 
 
-def getApplicationParametersFromCommandLine():
+def parseArgs():
     import argparse
     import os
-    # parse args
     parser = argparse.ArgumentParser()
     parser.add_argument("-o", "--outputDirectory", help = "Directory for outputs", default = os.getcwd())
     parser.add_argument("-a", "--ampliconLength", help = "Length of amplicon (not including primers)", required=True, type=int)
@@ -47,8 +49,13 @@ def getApplicationParametersFromCommandLine():
     parser.add_argument("-s", "--subsample", help = "Subsampling level (will analyze approximately 1/x reads", default=default.subsample, type=int)
     parser.add_argument("-p", "--percentile", help = "Percentile to use for expected error model", default=default.percentile, type=int)
     parser.add_argument("-F", "--fileNamingStandard", help = "File naming standard to use", default = "illumina")
-    args = parser.parse_args()
-    # validate args
+    parser.add_argument("-l", "--logFile", help = "Log file path", default = None)
+    return parser.parse_args()
+
+
+def getApplicationParametersFromCommandLine():
+    import os
+    args = parseArgs()
     outputFileName = args.outputFileName
     ampliconLength = args.ampliconLength
     if not args.fileNamingStandard.lower() in figaroSupport.fileNamingStandards.aliasList.keys():
@@ -97,8 +104,19 @@ def getApplicationParametersFromCommandLine():
 
 
 def getLoggingParameters():
+    import sys
+    import os
     loggingParameters = figaroSupport.environmentParameterParser.EnvParameters()
-    loggingParameters.addParameter("logFile", str, default=default.logFile, createdFile=True)
+    if len(sys.argv) > 1:
+        args = parseArgs()
+        if args.logFile:
+            loggingParameters.sideLoadParameter("logFile", args.logFile)
+        else:
+            import datetime
+            timestamp = str(datetime.datetime.now().timestamp()).replace(".", "")
+            loggingParameters.sideLoadParameter("logFile", os.path.join(args.outputDirectory, "figaro.%s.log" %timestamp))
+    else:
+        loggingParameters.addParameter("logFile", str, default=default.logFile, createdFile=True)
     loggingParameters.addParameter("logLevel", str, default=default.loggingLevel, logLevel=True)
     loggingParameters.addParameter("streamOff", bool, default=False)
     loggingParameters.addParameter("streamLoglevel", str, default=default.loggingLevel, logLevel=True)
@@ -167,15 +185,15 @@ def saveResultOutput(outputDirectory:str, outputResultTableFileName:str, resultT
     return outputResultTablePath, outputForwardCurvePath, outputReverseCurvePath
 
 
-def runAnalysis(inputDirectory:str, minimumCombinedReadLength:int, forwardPrimerLength:int, reversePrimerLength:int, fileNamingStandard:str="illumina", subsample:int = -1, percentile:int=83):
+def runAnalysis(inputDirectory:str, ampliconLength:int, forwardPrimerLength:int, reversePrimerLength:int, minimumOverlap:int=20, fileNamingStandard:str="illumina", subsample:int = -1, percentile:int=83):
     import os
     if not os.path.isdir(inputDirectory):
         raise NotADirectoryError("Unable to find directory at %s" %inputDirectory)
     if subsample == -1:
         totalFileSize = figaroSupport.fastqAnalysis.getEstimatedFastqSizeSumFromDirectory(inputDirectory, fileNamingStandard)
         fastqGigabytes = totalFileSize / 1000000000
         subsample = round(fastqGigabytes * 10)
-    resultTable, forwardCurve, reverseCurve = figaroSupport.trimParameterPrediction.performAnalysisLite(inputDirectory, minimumCombinedReadLength, subsample=subsample, percentile=percentile, forwardPrimerLength=forwardPrimerLength, reversePrimerLength=reversePrimerLength, namingStandardAlias=fileNamingStandard)
+    resultTable, forwardCurve, reverseCurve = figaroSupport.trimParameterPrediction.performAnalysisLite(inputDirectory, ampliconLength + minimumOverlap, subsample=subsample, percentile=percentile, forwardPrimerLength=forwardPrimerLength, reversePrimerLength=reversePrimerLength, namingStandardAlias=fileNamingStandard)
     return resultTable, forwardCurve, reverseCurve