pygr/apps/leelabdb.py

   1 import MySQLdb
   2 import os
   3 from splicegraph import *
   4
   5
   6 spliceCalcs={'HUMAN_SPLICE_03':
   7              TableGroup(db='HUMAN_SPLICE_03',suffix='JUN03',clusters='cluster_JUN03',
   8                         exons='exon_formJUN03',splices='splice_verification_JUN03',
   9                         genomic='genomic_cluster_JUN03',mrna='mrna_seqJUN03',
  10                         protein='protein_seqJUN03'),
  11              'HUMAN_SPLICE':
  12              TableGroup(db='HUMAN_SPLICE',suffix='jan02',clusters='cluster_jan02',
  13                         exons='HUMAN_ISOFORMS.exon_form_4',splices='splice_verification_jan02',
  14                         genomic='genomic_cluster_jan02',mrna='HUMAN_ISOFORMS.mrna_seq_4',
  15                         protein='HUMAN_ISOFORMS.protein_seq_4'),
  16              'MOUSE_SPLICE':
  17              TableGroup(db='MOUSE_SPLICE',suffix='jan02',clusters='cluster_jan02',
  18                         exons='MOUSE_ISOFORMS.exon_form_2',splices='splice_verification_jan02',
  19                         genomic='genomic_cluster_jan02',mrna='MOUSE_ISOFORMS.mrna_seq_2',
  20                         protein='MOUSE_ISOFORMS.protein_seq_2'),
  21              'MOUSE_SPLICE_03':
  22              TableGroup(db='MOUSE_SPLICE_03',suffix='JUN03',clusters='cluster_JUN03',
  23                         exons='exon_formJUN03',splices='splice_verification_JUN03',
  24                         genomic='genomic_cluster_JUN03',mrna='mrna_seqJUN03',
  25                         protein='protein_seqJUN03')
  26              }
  27
  28
  29 def getUserCursor(db):
  30     'get a cursor as the current user'
  31     db=MySQLdb.connect(db=db,read_default_file=os.environ['HOME']+'/.my.cnf',compress=True)
  32     return db.cursor()
  33
  34
  35 def getSpliceGraphFromDB(dbgroup,loadAll=False):
  36     """load data from MySQL using the designated database table group.
  37     If loadAll true, then load the entire splice graph into memory."""
  38     cursor=getUserCursor(dbgroup.db)
  39     import sys
  40     print >>sys.stderr,'Reading database schema...'
  41     idDict={}
  42     tables=describeDBTables(dbgroup.db,cursor,idDict)
  43     if hasattr(dbgroup,'suffix'):
  44         tables=suffixSubset(tables,dbgroup.suffix) # SET OF TABLES ENDING IN JUN03
  45         idDict=indexIDs(tables) # CREATE AN INDEX OF THEIR PRIMARY KEYS
  46     for t in dbgroup.values():
  47         if t is not None and '.' in t and t not in tables: # THIS TABLE COMES FROM ANOTHER DATABASE...
  48             tables[t]=SQLTable(t,cursor) # SO GET IT FROM OTHER DATABASE
  49
  50     # LOAD DATA & BUILD THE SPLICE GRAPH
  51     return loadSpliceGraph(tables,dbgroup.clusters,dbgroup.exons,dbgroup.splices,
  52                            dbgroup.genomic,dbgroup.mrna,dbgroup.protein,loadAll)
  53
  54
  55 def localCopy(localFile,cpCommand):
  56     'if not already present on local file location, run cpCommand'
  57     if not os.access(localFile,os.R_OK):
  58         cmd=cpCommand % localFile
  59         print 'copying data:',cmd
  60         exit_code=os.system(cmd)
  61         if exit_code!=0:
  62             raise OSError((exit_code,'command failed: %s' % cmd))
  63     return localFile