def _create_row_feature(self, feature_mapping, collapsed_view):
if collapsed_view:
new_map_position = feature_mapping.clone()
else:
map_position = MappingResult.get_empty()
new_map_position = map_position.clone()
new_map_position.set_feature(feature_mapping)
return new_map_position
def _create_row_feature(self, feature_mapping, collapsed_view):
if collapsed_view:
new_map_position = feature_mapping.clone()
else:
map_position = MappingResult.get_empty()
new_map_position = map_position.clone()
new_map_position.set_feature(feature_mapping)
return new_map_position
def _parse_index_file_by_id(self, query_ids_dict, index_path, data_path,
map_config, chrom_dict, multiple_param,
dataset_synonyms, test_set):
mapping_results_list = []
## TODO: DO SOMETHING WITH THE SYNONYMS!!!
##
map_name = map_config.get_name()
map_has_cm_pos = map_config.has_cm_pos()
map_has_bp_pos = map_config.has_bp_pos()
map_is_physical = map_config.as_physical()
sys.stderr.write("MappingsParser: loading index " + str(index_path) +
"...\n")
with open(index_path, 'r') as index_f:
index = cPickle.load(index_f)
sys.stderr.write("MappingsParser: loaded index with " +
str(len(index)) + " entries.\n")
sys.stderr.write("MappingsParser: obtaining index of queries...\n")
queries_bytes = []
for query in test_set:
if query in index:
query_bytes = index[query]
query_ids_dict[query] = 1 # found
queries_bytes.append(query_bytes)
with open(data_path, 'r') as data_f:
for query_bytes in queries_bytes:
data_f.seek(query_bytes)
mapping_line = data_f.readline()
mapping_data = mapping_line.strip().split("\t")
mapping_result = MappingResult.init_from_data(
mapping_data, map_name, chrom_dict, map_is_physical,
map_has_cm_pos, map_has_bp_pos)
if mapping_result.has_multiple_pos():
if multiple_param == False:
continue
mapping_results_list.append(mapping_result)
return mapping_results_list
def parse_mapping_file(self, data_path, map_config, chrom_dict):
mapping_results_list = []
map_name = map_config.get_name()
map_has_cm_pos = map_config.has_cm_pos()
map_has_bp_pos = map_config.has_bp_pos()
map_is_physical = map_config.as_physical()
for hit in open(data_path, 'r'):
if hit.startswith(">") or hit.startswith("#"): continue
hit_data = hit.strip().split("\t")
mapping_result = MappingResult.init_from_data(
hit_data, map_name, chrom_dict, map_is_physical,
map_has_cm_pos, map_has_bp_pos)
mapping_results_list.append(mapping_result)
return mapping_results_list
def _createPositions(self, markers_positions, multiple_param, chrom_dict,
map_name):
positions_list = []
for marker_id in markers_positions:
#sys.stderr.write(marker_id+"\n")
positions = markers_positions[marker_id]["positions"]
#sys.stderr.write(str(positions)+"\n")
num_marker_pos = len(positions)
if num_marker_pos == 0: continue # contigs without position
if (not multiple_param) and (num_marker_pos > 1):
if self._verbose:
sys.stderr.write(
"Mappers: discarded multiple pos marker: " +
str(marker_id) + "\n")
continue # Multiple positions
num_contig_no_pos = len(
markers_positions[marker_id]["hits_no_position"])
for pos in positions:
# marker - chr - cm_pos - bp_pos - multiple - has_contigs_with_no_pos - map_name
chr_pos = pos["chr"]
# If the chromosome is not in the genome, skip this alignment result
if not chr_pos in chrom_dict: continue
chrom_order = chrom_dict[
chr_pos] # Numeric value of chromsome (for sorting purposes)
mapping_result = MappingResult(
marker_id, chr_pos, chrom_order, pos["cm_pos"],
pos["cm_end_pos"], pos["bp_pos"], pos["bp_end_pos"],
pos["strand"], num_marker_pos > 1, num_contig_no_pos > 0,
map_name)
positions_list.append(mapping_result)
return positions_list
def _parse_mapping_file_by_id(self,
query_ids_dict,
data_path,
map_config,
chrom_dict,
multiple_param,
dataset_synonyms={},
test_set=None):
mapping_results_list = []
map_name = map_config.get_name()
map_has_cm_pos = map_config.has_cm_pos()
map_has_bp_pos = map_config.has_bp_pos()
map_is_physical = map_config.as_physical()
for hit in open(data_path, 'r'):
#sys.stderr.write(" ONE**************************\n")
#sys.stderr.write(str(hit)+"\n")
if hit.startswith(">") or hit.startswith("#"): continue
hit_data = hit.strip().split("\t")
#sys.stderr.write("data\n")
if test_set:
hit_query = hit_data[0]
#if hit_query == "12_30924":
# sys.stderr.write(str(test_set)+"\n")
#sys.stderr.write("CHECK TESTSET\n")
if hit_query in dataset_synonyms:
#if hit_query == "12_30924":
# sys.stderr.write("IS IN SYNONYMS\n")
hit_synonyms = dataset_synonyms[hit_query]
synonyms_found = test_set.intersection(hit_synonyms)
if len(synonyms_found) > 0:
#if hit_query == "12_30924":
# sys.stderr.write("-".join(synonyms_found)+"\n")
mapping_result = MappingResult.init_from_data(
hit_data, map_name, chrom_dict, map_is_physical,
map_has_cm_pos, map_has_bp_pos)
for synonym in synonyms_found: # all found
query_ids_dict[synonym] = 1
if mapping_result.has_multiple_pos():
if multiple_param == False:
continue
else: # just for sake of readability
for synonym in synonyms_found: # all found
if synonym in test_set:
test_set.remove(synonym)
mapping_result.set_marker_id("|".join(synonyms_found))
mapping_results_list.append(mapping_result)
else:
#if hit_query == "12_30924":
# sys.stderr.write("IS NOT IN SYNONYMS\n")
if hit_query in test_set:
#sys.stderr.write("create mapping data\n")
mapping_result = MappingResult.init_from_data(
hit_data, map_name, chrom_dict, map_is_physical,
map_has_cm_pos, map_has_bp_pos)
query_ids_dict[hit_query] = 1 # found
#sys.stderr.write("append\n")
if mapping_result.has_multiple_pos():
if multiple_param == False:
continue
else: # just for sake of readability
test_set.remove(hit_query)
mapping_results_list.append(mapping_result)
else: # retrieve all mapping results
mapping_result = MappingResult.init_from_data(
hit_data, map_name, chrom_dict, map_is_physical,
map_has_cm_pos, map_has_bp_pos)
query_ids_dict[hit_query] = 1 # found
if mapping_result.has_multiple_pos():
if multiple_param == False:
continue
mapping_results_list.append(mapping_result)
#sys.stderr.write("**********NEXT\n")
if len(test_set) == 0: break
return mapping_results_list
def parse_mapping_file_on_pos(self, map_intervals, data_path, chrom_dict,
map_config, map_sort_by, dataset,
dataset_name, feature_type):
map_name = map_config.get_name()
map_is_physical = map_config.as_physical()
map_has_cm_pos = map_config.has_cm_pos()
map_has_bp_pos = map_config.has_bp_pos()
current_interval_pos = 0
featured_current_interval = map_intervals[current_interval_pos]
current_interval = featured_current_interval.get_map_interval()
current_features = featured_current_interval.get_features()
#sys.stderr.write("MappingsParser \n")
#sys.stderr.write("\t"+str(current_interval)+"\n")
#sys.stderr.write("\t"+str(len(current_features))+"\n")
#for feature in current_features:
# sys.stderr.write("\t\t"+str(feature)+"\n")
# Find all the hits for this map
for hit in open(data_path, 'r'):
if hit.startswith(">") or hit.startswith("#"): continue
hit_data = hit.strip().split("\t")
#sys.stderr.write(hit+"\n")
#sys.stderr.write("\t"+str(current_interval)+"\n")
mapping_result = MappingResult.init_from_data(
hit_data, map_name, chrom_dict, map_is_physical,
map_has_cm_pos, map_has_bp_pos)
chrom_name = mapping_result.get_chrom_name()
if chrom_name != current_interval.get_chrom(): continue
map_end_pos = mapping_result.get_sort_end_pos(map_sort_by)
if float(map_end_pos) < float(current_interval.get_ini_pos()):
continue
marker_id = mapping_result.get_marker_id()
chrom_order = mapping_result.get_chrom_order()
map_pos = mapping_result.get_sort_pos(
map_sort_by) #float(mapping_result.get_sort_pos(map_sort_by))
dataset_interval = MapInterval(chrom_name, map_pos, map_end_pos)
does_overlap = MapInterval.intervals_overlap(
dataset_interval, current_interval)
# This if-else could be unnecessary, but hopefully is useful to read the code
if does_overlap:
next_interval_pos = current_interval_pos
next_interval = current_interval
next_features = current_features
while (does_overlap):
feature = FeaturesFactory.get_feature(
marker_id, dataset, dataset_name, feature_type,
mapping_result)
next_features.append(feature)
next_interval_pos += 1
if next_interval_pos >= len(map_intervals):
break
featured_next_interval = map_intervals[next_interval_pos]
next_interval = featured_next_interval.get_map_interval()
next_features = featured_next_interval.get_features()
does_overlap = MapInterval.intervals_overlap(
dataset_interval, next_interval)
else:
while (float(map_pos) > float(current_interval.get_end_pos())):
current_interval_pos += 1
if current_interval_pos >= len(map_intervals):
break
featured_current_interval = map_intervals[
current_interval_pos]
current_interval = featured_current_interval.get_map_interval(
)
current_features = featured_current_interval.get_features()
#sys.stderr.write("MappingsParser generated intervals\n")
#for featured_map_interval in map_intervals:
# map_interval = featured_map_interval.get_map_interval()
# sys.stderr.write("\tinterval: "+str(map_interval)+"\n")
# features = featured_map_interval.get_features()
# for feature in features:
# sys.stderr.write("\t\tfeature: "+str(feature)+"\n")
return map_intervals
def parse_mapping_file_by_pos(self, map_intervals, data_path, chrom_dict,
map_config, map_sort_by):
mapping_results_list = []
map_name = map_config.get_name()
map_is_physical = map_config.as_physical()
map_has_cm_pos = map_config.has_cm_pos()
map_has_bp_pos = map_config.has_bp_pos()
current_interval_pos = 0
current_interval = map_intervals[current_interval_pos]
# Find all the hits for this map within intervals of interest
# Note: hits/map features are pre-computed & sorted along chroms, chroms are in chrom_dict order
# Note: intervals are pre-sorted as well
for hit in open(data_path, 'r'):
if hit.startswith(">") or hit.startswith("#"): continue
hit_data = hit.strip().split("\t")
#sys.stderr.write(hit+"\n")
#sys.stderr.write("\t"+str(current_interval)+"\n")
mapping_result = MappingResult.init_from_data(
hit_data, map_name, chrom_dict, map_is_physical,
map_has_cm_pos, map_has_bp_pos)
chrom_name = mapping_result.get_chrom_name()
# move to next interval to match chrom (only if previous mappings exist)
# Note: needed when last mapping matched exactly the last gene of a chrom
while len(mapping_results_list) > 0 and \
current_interval_pos < len(map_intervals)-1 and \
int(chrom_dict[chrom_name]) > int(chrom_dict[current_interval.get_chrom()]):
current_interval_pos += 1
current_interval = map_intervals[current_interval_pos]
if chrom_name != current_interval.get_chrom(): continue
map_end_pos = mapping_result.get_sort_end_pos(map_sort_by)
if float(map_end_pos) < float(current_interval.get_ini_pos()):
continue
marker_id = mapping_result.get_marker_id()
chrom_order = mapping_result.get_chrom_order()
map_pos = mapping_result.get_sort_pos(
map_sort_by) #float(mapping_result.get_sort_pos(map_sort_by))
while (float(map_pos) > float(current_interval.get_end_pos())):
current_interval_pos += 1
if current_interval_pos >= len(map_intervals):
break
current_interval = map_intervals[current_interval_pos]
if current_interval.get_chrom() != chrom_name:
break
if current_interval_pos >= len(map_intervals): break
if chrom_name != current_interval.get_chrom(): continue
if float(map_end_pos) < float(current_interval.get_ini_pos()):
continue
dataset_interval = MapInterval(chrom_name, map_pos, map_end_pos)
#sys.stderr.write("MappingsParser: by_pos "+str(dataset_interval)+" - "+str(current_interval)+"\n")
does_overlap = MapInterval.intervals_overlap(
dataset_interval, current_interval)
# Check if alignment overlaps with some mapping interval
if does_overlap:
mapping_results_list.append(mapping_result)
return mapping_results_list
def get_empty():
return GeneMapping("-", "-", "-", "-", MappingResult.get_empty(), FeatureMapping.ROW_TYPE_MAPPING_RESULT, empty = True, annots = [])
def get_empty():
return MarkerMapping("-", "-", "-", "-", MappingResult.get_empty(), FeatureMapping.ROW_TYPE_MAPPING_RESULT, empty = True)
def parse_mapping_file_by_pos(self, map_intervals, data_path, chrom_dict,
map_config, map_sort_by):
mapping_results_list = []
map_name = map_config.get_name()
map_is_physical = map_config.as_physical()
map_has_cm_pos = map_config.has_cm_pos()
map_has_bp_pos = map_config.has_bp_pos()
current_interval_pos = 0
current_interval = map_intervals[current_interval_pos]
# Find all the hits for this map
for hit in open(data_path, 'r'):
if hit.startswith(">") or hit.startswith("#"): continue
hit_data = hit.strip().split("\t")
#sys.stderr.write(hit+"\n")
#sys.stderr.write("\t"+str(current_interval)+"\n")
mapping_result = MappingResult.init_from_data(
hit_data, map_name, chrom_dict, map_is_physical,
map_has_cm_pos, map_has_bp_pos)
chrom_name = mapping_result.get_chrom_name()
if chrom_name != current_interval.get_chrom(): continue
map_end_pos = mapping_result.get_sort_end_pos(map_sort_by)
if float(map_end_pos) < float(current_interval.get_ini_pos()):
continue
marker_id = mapping_result.get_marker_id()
chrom_order = mapping_result.get_chrom_order()
map_pos = mapping_result.get_sort_pos(
map_sort_by) #float(mapping_result.get_sort_pos(map_sort_by))
while (float(map_pos) > float(current_interval.get_end_pos())):
current_interval_pos += 1
if current_interval_pos >= len(map_intervals):
break
current_interval = map_intervals[current_interval_pos]
if current_interval.get_chrom() != chrom_name:
break
if current_interval_pos >= len(map_intervals): break
if chrom_name != current_interval.get_chrom(): continue
if float(map_end_pos) < float(current_interval.get_ini_pos()):
continue
dataset_interval = MapInterval(chrom_name, map_pos, map_end_pos)
#sys.stderr.write("MappingsParser: by_pos "+str(dataset_interval)+" - "+str(current_interval)+"\n")
does_overlap = MapInterval.intervals_overlap(
dataset_interval, current_interval)
# Check if alignment overlaps with some mapping interval
if does_overlap:
mapping_results_list.append(mapping_result)
return mapping_results_list