From 0bd693e45000e95483aa80406e0e831e4ffc8373 Mon Sep 17 00:00:00 2001
From: "Gervaise H. Henry" <gervaise.henry@utsouthwestern.edu>
Date: Fri, 15 Nov 2019 10:09:38 -0600
Subject: [PATCH] Generalize file param input for python scripts
---
workflow/scripts/modifyFetch.py | 8 ++++----
workflow/scripts/splitFetch.py | 6 +++---
2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/workflow/scripts/modifyFetch.py b/workflow/scripts/modifyFetch.py
index 82b1d4c..e6accfb 100644
--- a/workflow/scripts/modifyFetch.py
+++ b/workflow/scripts/modifyFetch.py
@@ -6,20 +6,20 @@ import re
def get_args():
parser = argparse.ArgumentParser()
- parser.add_argument('-f', '--fetchFile',help="The fetch file from bdgap.zip.",required=True)
+ parser.add_argument('-f', '--files',help="The fetch file from bdgap.zip.",required=True)
args = parser.parse_args()
return args
def main():
args = get_args()
- fetchFile = pd.read_csv(args.fetchFile+"/fetch.txt",sep="\t",header=None)
- fileFile = pd.read_csv(args.fetchFile+"/data/File.csv",sep=",",header=0)
+ fetchFile = pd.read_csv(args.files+"/fetch.txt",sep="\t",header=None)
+ fileFile = pd.read_csv(args.files+"/data/File.csv",sep=",",header=0)
fileFile_filtered = fileFile[fileFile["File_Type"]=="FastQ"]
fetchFile_filtered = fetchFile[fetchFile[2].str[-9:]==".fastq.gz"]
fetchFile_filtered_renamed = fetchFile_filtered
for i in fileFile_filtered["File_Name"]:
fetchFile_filtered_renamed[2][fetchFile_filtered_renamed[2].str.contains(i,regex=False)] = fetchFile_filtered_renamed[2][fetchFile_filtered_renamed[2].str.contains(i,regex=False)].values[0].replace(re.sub("\.R.\.fastq\.gz","",i),fileFile_filtered["Replicate_RID"][fileFile_filtered["File_Name"]==i].values[0])
- fetchFile_filtered_renamed.to_csv(args.fetchFile+"/fetch.txt",sep="\t",header=False,index=False)
+ fetchFile_filtered_renamed.to_csv(args.files+"/fetch.txt",sep="\t",header=False,index=False)
if __name__ == '__main__':
main()
\ No newline at end of file
diff --git a/workflow/scripts/splitFetch.py b/workflow/scripts/splitFetch.py
index c8f6004..63385c1 100644
--- a/workflow/scripts/splitFetch.py
+++ b/workflow/scripts/splitFetch.py
@@ -6,14 +6,14 @@ import os
def get_args():
parser = argparse.ArgumentParser()
- parser.add_argument('-f', '--fetchFile',help="The fetch file from bdgap.zip.",required=True)
+ parser.add_argument('-f', '--files',help="The fetch file from bdgap.zip.",required=True)
args = parser.parse_args()
return args
def main():
args = get_args()
- fetchFile = pd.read_csv(args.fetchFile+"/fetch.txt",sep="\t",header=None)
- fileFile = pd.read_csv(args.fetchFile+"/data/File.csv",sep=",",header=0)
+ fetchFile = pd.read_csv(args.files+"/fetch.txt",sep="\t",header=None)
+ fileFile = pd.read_csv(args.files+"/data/File.csv",sep=",",header=0)
replicateRID = fileFile.Replicate_RID.unique()
fetchArray = {i:fileFile.URI[(fileFile.Replicate_RID == i) & (fileFile.File_Type == "FastQ")] for i in replicateRID}
for i in replicateRID:
--
GitLab