def getProblems(data, txn=None):
if 'genome' not in data:
data['genome'] = getGenome(data, txn=txn)
problems = db.Problems(data['genome'])
problemsInBounds = problems.getInBounds(data['ref'], data['start'],
data['end'])
if problemsInBounds is None:
problemsPath = os.path.join(cfg.jbrowsePath, cfg.dataPath, 'genomes',
data['genome'], 'problems.bed')
if not os.path.exists(problemsPath):
location = Hubs.generateProblems(data['genome'], problemsPath)
if not location == problemsPath:
raise Exception
problemsDf = pd.read_csv(problemsPath, sep='\t', header=None)
problemsDf.columns = problemColumns
problems.put(problemsDf, txn=txn)
problemsIsInBounds = problemsDf.apply(db.checkInBounds,
axis=1,
args=(data['ref'], data['start'],
data['end']))
return problemsDf[problemsIsInBounds].to_dict('records')
else:
return problemsInBounds.to_dict('records')
def generateProblems(genome, path, txn=None):
genesUrl = "%s%s/database/" % (cfg.geneUrl, genome)
genomePath = os.path.join(path, 'genomes', genome)
outputFile = os.path.join(genomePath, 'problems.bed')
if db.Problems.has_key(genome):
return outputFile
if not os.path.exists(genomePath):
try:
os.makedirs(genomePath)
except OSError:
return
files = []
for file in ['chromInfo', 'gap']:
outputPath = os.path.join(genomePath, file + '.txt')
fileUrl = genesUrl + file + '.txt.gz'
files.append(downloadAndUnpackFile(fileUrl, outputPath))
chromInfo = pd.read_csv(files[0], sep='\t', header=None).iloc[:, 0:2]
chromInfo.columns = ['chrom', 'bases']
gap = pd.read_csv(files[1], sep='\t', header=None).iloc[:, 1:4]
gap.columns = ['chrom', 'gapStart', 'gapEnd']
join = gap.merge(chromInfo, on='chrom', how='outer')
nan = join.isnull()['gapStart' and 'gapEnd']
nonNan = join.notnull()['gapStart' and 'gapEnd']
nanOutput = join[nan].groupby(['chrom']).apply(createNanProblems)
nonNanOutput = join[nonNan].groupby(['chrom']).apply(createProblems)
frames = [nonNanOutput, nanOutput]
output = pd.concat(frames, sort=False)
# Removes all entries with an _, not needed because these are genome "Fixes"
output = output[~output['chrom'].str.contains('_')]
output['chromStart'] = output['chromStart'].astype(int)
output['chromEnd'] = output['chromEnd'].astype(int)
db.Problems(genome).put(output, txn=txn)
output.to_csv(outputFile, sep='\t', index=False, header=False)
return outputFile
def getProblemsForChrom(genome, chrom, txn=None):
problems = db.Problems(genome).get(txn=txn)
return problems[problems['chrom'] == chrom].copy()