From 09870eec2a9d0e27b14117e8802cbb2c786b6306 Mon Sep 17 00:00:00 2001
From: Martijn Vermaat <martijn@vermaat.name>
Date: Thu, 20 Jan 2011 15:19:38 +0000
Subject: [PATCH] Added batch SNP converter.
src/wsgi.py:
- Updated batch routing to include SNP converter.
mutalyzer.conf:
src/Modules/Config.py:
- Added result file header line for batch SNP converter.
src/Modules/Scheduler.py:
- Added batch SNP converter.
src/Modules/File.py:
- Updated batch CSV input file parsing. Default is now standard Excel format
and only if the CSV sniffer can find another dialect using a predefined set
of delimiter characters without error we use that one.
templates/menu.html:
- Added link for batch SNP converter.
git-svn-id: https://humgenprojects.lumc.nl/svn/mutalyzer/trunk@163 eb6bd6ab-9ccd-42b9-aceb-e2899b4a52f1
---
mutalyzer.conf | 7 +++--
src/Modules/Config.py | 1 +
src/Modules/File.py | 18 +++++++++----
src/Modules/Scheduler.py | 58 ++++++++++++++++++++++++++++++++++++++++
src/wsgi.py | 8 +++---
templates/menu.html | 12 +++++++++
6 files changed, 94 insertions(+), 10 deletions(-)
diff --git a/mutalyzer.conf b/mutalyzer.conf
index 12ff4d0e..2a35880d 100644
--- a/mutalyzer.conf
+++ b/mutalyzer.conf
@@ -120,12 +120,15 @@ PIDfile = "./var/batch.pid"
# The output header for NameChecking
nameCheckOutHeader = "Input", "Errors | Messages", "AccNo", "Genesymbol", "Variant", "Reference Sequence Start Descr.", "Coding DNA Descr.", "Protein Descr.", "GeneSymbol Coding DNA Descr.", "GeneSymbol Protein Descr.", "Genomic Reference", "Coding Reference", "Protein Reference", "Affected Transcripts", "Affected Proteins"
-# The output header for NameChecking
+# The output header for SyntaxChecking
syntaxCheckOutHeader = "Input", "Status"
-# The output header for NameChecking
+# The output header for PositionConverter
positionConverterOutHeader = "Input Variant", "Errors", "Chromosomal Variant", "Coding Variant(s)"
+# The output header for SnpConverter
+snpConverterOutHeader = "Input Variant", "HGVS description(s)", "Errors | Messages"
+
#
# These settings are used by the File module.
diff --git a/src/Modules/Config.py b/src/Modules/Config.py
index 94b9c66f..f0662721 100644
--- a/src/Modules/Config.py
+++ b/src/Modules/Config.py
@@ -204,6 +204,7 @@ class Config() :
self.Scheduler.nameCheckOutHeader = config["nameCheckOutHeader"]
self.Scheduler.syntaxCheckOutHeader= config["syntaxCheckOutHeader"]
self.Scheduler.positionConverterOutHeader= config["positionConverterOutHeader"]
+ self.Scheduler.snpConverterOutHeader= config["snpConverterOutHeader"]
# Set thte variables neede for the Batch module.
self.Batch.PIDfile = config["PIDfile"]
diff --git a/src/Modules/File.py b/src/Modules/File.py
index d1e13a9f..8206da6b 100644
--- a/src/Modules/File.py
+++ b/src/Modules/File.py
@@ -123,16 +123,24 @@ class File() :
handle.seek(0)
buf = handle.read(self.__config.bufSize)
+ # Default dialect
+ dialect = 'excel'
+
+ # The idea is that for new-style batch input files we have only
+ # one column and the sniffer cannot find a delimiter.
+
try :
- dialect = csv.Sniffer().sniff(buf)
+ # Todo: delimiters in config file
+ dialect = csv.Sniffer().sniff(buf, delimiters="\t ;|,")
except csv.Error, e :
- self.__output.addMessage(__file__, 4, "EBPARSE", e)
- return None
+ #self.__output.addMessage(__file__, 4, "EBPARSE", e)
+ #return None
+ pass
#except
#Watch out for : delimiter FIXME and for the . delimiter
- if dialect.delimiter == ":":
- dialect.delimiter = "\t"
+# if dialect.delimiter == ":":
+# dialect.delimiter = "\t"
handle.seek(0)
reader = csv.reader(handle, dialect)
diff --git a/src/Modules/Scheduler.py b/src/Modules/Scheduler.py
index 4aabc806..21503a12 100644
--- a/src/Modules/Scheduler.py
+++ b/src/Modules/Scheduler.py
@@ -27,6 +27,7 @@ from Modules import Config # Config.Config
from Modules import Output # Output.Output
from Modules import Parser # Parser.Nomenclatureparser
from Modules import Mapper # Mapper.Converter
+from Modules import Retriever # Retriever.Retriever
import Mutalyzer # Mutalyzer.process
@@ -333,6 +334,8 @@ class Scheduler() :
self._processSyntaxCheck(inputl, i, flags)
elif jobType == "PositionConverter" :
self._processConversion(inputl, i, arg1, flags)
+ elif jobType == "SnpConverter" :
+ self._processSNP(inputl, i, flags)
else: #unknown jobType
pass #TODO: Scream burning water and remove from Queue
else :
@@ -566,6 +569,61 @@ class Scheduler() :
#_processConversion
+ def _processSNP(self, cmd, i, flags) :
+ """
+ Process an entry from the SNP converter Batch, write the results
+ to the job-file. If an Exception is raised, catch and continue.
+
+ Side-effect:
+ - Output written to outputfile.
+
+ @arg cmd: The SNP converter input
+ @type cmd:
+ @arg i: The JobID
+ @type i:
+ @arg flags: Flags of the current entry
+ @type flags:
+ """
+
+ C = Config.Config()
+ O = Output.Output(__file__, C.Output)
+ O.addMessage(__file__, -1, "INFO",
+ "Received SNP converter batch rs" + cmd)
+
+ #Read out the flags
+ # Todo: Do something with the flags?
+ skip = self.__processFlags(O, flags)
+
+ if not skip :
+ R = Retriever.Retriever(C.Retriever, O, None)
+ R.snpConvert(cmd)
+
+ # Todo: Is output ok?
+ outputline = "%s\t" % cmd
+ outputline += "%s\t" % "|".join(O.getOutput('snp'))
+ outputline += "%s\t" % "|".join(O.getBatchMessages(3))
+
+ outputline += "\n"
+
+ #Output
+ filename = "%s/Results_%s.txt" % (self.__config.resultsDir, i)
+ if not os.path.exists(filename) :
+ # If the file does not yet exist, create it with the correct
+ # header above it. The header is read from the config file as
+ # a list. We need a tab delimited string.
+ header = self.__config.snpConverterOutHeader
+ handle = open(filename, 'a')
+ handle.write("%s\n" % "\t".join(header))
+ #if
+ else :
+ handle = open(filename, 'a')
+
+ handle.write(outputline)
+ handle.close()
+ O.addMessage(__file__, -1, "INFO",
+ "Finished SNP converter batch rs%s" % cmd)
+ #_processSNP
+
def addJob(self, outputFilter, eMail, queue, fromHost, jobType, Arg1) :
"""
diff --git a/src/wsgi.py b/src/wsgi.py
index 9449f0fe..46c251c7 100644
--- a/src/wsgi.py
+++ b/src/wsgi.py
@@ -768,7 +768,7 @@ class BatchChecker:
- arg1: Additional argument. Currently only used if batchType is
'PositionConverter', denoting the human genome build.
- batchType: Type of batch job to run. One of 'NameChecker' (default),
- 'SyntaxChecker', or 'PositionChecker'.
+ 'SyntaxChecker', 'PositionConverter', or 'SnpConverter'.
"""
i = web.input(batchEmail=None, batchFile={}, arg1='',
batchType=None)
@@ -786,7 +786,8 @@ class BatchChecker:
@kwarg arg1: Additional argument. Currently only used if batchType is
'PositionConverter', denoting the human genome build.
@kwarg batchType: Type of batch job to run. One of 'NameChecker'
- (default), 'SyntaxChecker', or 'PositionChecker'.
+ (default), 'SyntaxChecker', 'PositionConverter', or
+ 'SnpConverter'.
"""
O = Output.Output(__file__, C.Output)
@@ -795,7 +796,8 @@ class BatchChecker:
"debug" : [],
"batchTypes" : ["NameChecker",
"SyntaxChecker",
- "PositionConverter"],
+ "PositionConverter",
+ "SnpConverter"],
"hideTypes" : batchType and 'none' or '',
"selected" : "0",
"batchType" : batchType or "",
diff --git a/templates/menu.html b/templates/menu.html
index 0e6f7f9e..4b48bc7b 100644
--- a/templates/menu.html
+++ b/templates/menu.html
@@ -274,6 +274,18 @@
</td>
</tr>
+ <tr>
+ <td></td>
+ <td valign="baseline" width="10">
+ <img src="base/images/bullitmiddel.gif" id="b_batchSnpConverter">
+ </td>
+ <td colspan="2">
+ <a id="page_batchSnpConverter"
+ href="batchSnpConverter"
+ class="vertnavsub">SNP Converter</a>
+ </td>
+ </tr>
+
<tr>
<td valign="top" width="20">
<img src="base/images/bullitdonker.gif" id="b_upload">
--
GitLab