def create_db_obj_from_txt_GENEINDEXED(txt_fname,
title,
indexColumnNames='gene'):
data_0, colnames = read_delim(txt_fname, return_list=False)
data = read_data_and_add_title_str_to_each_header(data_0, title,
indexColumnNames)
new_colnames = [
'%s_%s' % (title, c.strip()) for c in colnames
if c not in [indexColumnNames]
]
annotation_table = dict()
for d in data:
if d[indexColumnNames] in annotation_table:
print("ERROR: " + indexColumnNames +
" indices must be unique: %s found multiple times in %s!" %
(d[indexColumnNames], txt_fname))
sys.exit(1)
annotation_table[d[indexColumnNames]] = dict()
for field in new_colnames:
idx_key = d[indexColumnNames]
annotation_table[idx_key][field] = d[field].strip()
return annotation_table, new_colnames
def create_db_obj_from_txt_POSINDEXED(txt_fname,
title,
indexColumnNames='chr,start,end'):
data_0, colnames = read_delim(txt_fname, return_list=False)
data = read_data_and_add_title_str_to_each_header(data_0, title,
indexColumnNames)
indexColumnList = indexColumnNames.split(',')
if len(set(colnames).intersection(set(indexColumnList))) <> len(
set(indexColumnList)):
print("ERROR: Invalid header values!")
print("If indexing by genomic position, headers must include '" +
indexColumnNames + "'.")
raise Exception(
"If indexing by genomic position, headers must include '" +
indexColumnNames + "'.")
excludeList = indexColumnList
excludeList.append('build')
new_colnames = [
'%s_%s' % (title, c) for c in colnames if c not in excludeList
]
annotation_table = dict()
chr = indexColumnList[0]
start = indexColumnList[1]
end = indexColumnList[2]
for d in data:
if d[chr].startswith('chr'): d[chr] = d[chr][3:]
d['bin'] = region2bin(int(d[start]), int(d[end]))
d['start'] = d[start]
d['end'] = d[end]
data = sorted(data, key=operator.itemgetter(end))
data = sorted(data, key=operator.itemgetter(start))
data = sorted(data, key=operator.itemgetter(chr))
for d in data:
if d[chr] in annotation_table:
if d['bin'] in annotation_table[d[chr]]:
annotation_table[d[chr]][d['bin']].append(d)
else:
annotation_table[d[chr]][d['bin']] = [d]
else:
annotation_table[d[chr]] = {d['bin']: [d]}
return annotation_table, new_colnames
def create_db_obj_from_txt_GENEINDEXED(txt_fname, title, indexColumnNames='gene'):
data_0, colnames = read_delim(txt_fname, return_list=False)
data = read_data_and_add_title_str_to_each_header(data_0, title, indexColumnNames)
new_colnames = ['%s_%s' % (title, c.strip()) for c in colnames if c not in [indexColumnNames]]
annotation_table = dict()
for d in data:
if d[indexColumnNames] in annotation_table:
print("ERROR: " + indexColumnNames + " indices must be unique: %s found multiple times in %s!" % (d[indexColumnNames], txt_fname))
sys.exit(1)
annotation_table[d[indexColumnNames]] = dict()
for field in new_colnames:
annotation_table[d[indexColumnNames]][field] = d[field].strip()
return annotation_table, new_colnames
def create_db_obj_from_txt_POSINDEXED(txt_fname, title, indexColumnNames='chr,start,end'):
data_0, colnames = read_delim(txt_fname, return_list=False)
data = read_data_and_add_title_str_to_each_header(data_0, title, indexColumnNames)
indexColumnList = indexColumnNames.split(',')
if len(set(colnames).intersection( set(indexColumnList) )) <> len(set(indexColumnList)):
print("ERROR: Invalid header values!")
print("If indexing by genomic position, headers must include '" + indexColumnNames + "'.")
raise Exception("If indexing by genomic position, headers must include '" + indexColumnNames + "'.")
excludeList = indexColumnList
excludeList.append('build')
new_colnames = ['%s_%s' % (title, c) for c in colnames if c not in excludeList]
annotation_table = dict()
chr = indexColumnList[0]
start = indexColumnList[1]
end = indexColumnList[2]
for d in data:
if d[chr].startswith('chr'): d[chr] = d[chr][3:]
d['bin'] = region2bin(int(d[start]), int(d[end]))
d['start'] = d[start]
d['end'] = d[end]
data = sorted(data, key=operator.itemgetter(end))
data = sorted(data, key=operator.itemgetter(start))
data = sorted(data, key=operator.itemgetter(chr))
for d in data:
if d[chr] in annotation_table:
if d['bin'] in annotation_table[d[chr]]:
annotation_table[d[chr]][d['bin']].append(d)
else:
annotation_table[d[chr]][d['bin']] = [d]
else:
annotation_table[d[chr]] = {d['bin']:[d]}
return annotation_table, new_colnames
