def test_run_gemma_firstrun_set_false(self, mock_parse_loco):
"""add tests for gemma function where first run is set to false"""
dataset = AttributeSetter(
{"group": AttributeSetter({"genofile": "genofile.geno"})})
output_file = "file1"
mock_parse_loco.return_value = []
this_trait = AttributeSetter({"name": "t1"})
result = run_gemma(this_trait=this_trait,
this_dataset=dataset,
samples=[],
vals=[],
covariates="",
use_loco=True,
first_run=False,
output_files=output_file)
expected_results = ([], "file1")
self.assertEqual(expected_results, result)
def test_run_gemma_firstrun_set_true(self, mock_gen_pheno_txt, mock_os,
mock_choice, mock_gen_covar,
mock_flat_files, mock_parse_loco):
"""add tests for run_gemma where first run is set to true"""
this_chromosomes = {}
for i in range(1, 5):
this_chromosomes[f'CH{i}'] = (AttributeSetter({"name": f"CH{i}"}))
chromosomes = AttributeSetter({"chromosomes": this_chromosomes})
dataset_group = MockGroup({"name": "GP1", "genofile": "file_geno"})
dataset = AttributeSetter({
"group":
dataset_group,
"name":
"dataset1_name",
"species":
AttributeSetter({"chromosomes": chromosomes})
})
trait = AttributeSetter({"name": "trait1"})
samples = []
mock_gen_pheno_txt.return_value = None
mock_os.path.isfile.return_value = True
mock_gen_covar.return_value = None
mock_choice.return_value = "R"
mock_flat_files.return_value = "/home/genotype/bimbam"
mock_parse_loco.return_value = []
results = run_gemma(this_trait=trait,
this_dataset=dataset,
samples=[],
vals=[],
covariates="",
use_loco=True)
self.assertEqual(mock_os.system.call_count, 2)
mock_gen_pheno_txt.assert_called_once()
mock_parse_loco.assert_called_once_with(dataset, "GP1_GWA_RRRRRR",
True)
mock_os.path.isfile.assert_called_once_with(
('/home/user/imgfile_output.assoc.txt'))
self.assertEqual(mock_flat_files.call_count, 4)
self.assertEqual(results, ([], "GP1_GWA_RRRRRR"))
def __init__(self, start_vars, temp_uuid):
helper_functions.get_species_dataset_trait(self, start_vars)
self.temp_uuid = temp_uuid #needed to pass temp_uuid to gn1 mapping code (marker_regression_gn1.py)
self.json_data = {}
self.json_data['lodnames'] = ['lod.hk']
self.samples = [] # Want only ones with values
self.vals = []
all_samples_ordered = self.dataset.group.all_samples_ordered()
primary_sample_names = list(all_samples_ordered)
for sample in self.dataset.group.samplelist:
# sample is actually the name of an individual
in_trait_data = False
for item in self.this_trait.data:
if self.this_trait.data[item].name == sample:
value = start_vars['value:' + self.this_trait.data[item].name]
self.samples.append(self.this_trait.data[item].name)
self.vals.append(value)
in_trait_data = True
break
if not in_trait_data:
value = start_vars.get('value:' + sample)
if value:
self.samples.append(sample)
self.vals.append(value)
self.mapping_method = start_vars['method']
if start_vars['manhattan_plot'] == "True":
self.manhattan_plot = True
else:
self.manhattan_plot = False
self.maf = start_vars['maf'] # Minor allele frequency
self.suggestive = ""
self.significant = ""
self.pair_scan = False # Initializing this since it is checked in views to determine which template to use
self.score_type = "LRS" #ZS: LRS or LOD
self.mapping_scale = "physic"
self.num_perm = 0
self.perm_output = []
self.bootstrap_results = []
#ZS: This is passed to GN1 code for single chr mapping
self.selected_chr = -1
if "selected_chr" in start_vars:
if int(start_vars['selected_chr']) != -1: #ZS: Needs to be -1 if showing full map; there's probably a better way to fix this
self.selected_chr = int(start_vars['selected_chr']) + 1
else:
self.selected_chr = int(start_vars['selected_chr'])
if "startMb" in start_vars:
self.startMb = start_vars['startMb']
if "endMb" in start_vars:
self.endMb = start_vars['endMb']
if "graphWidth" in start_vars:
self.graphWidth = start_vars['graphWidth']
if "lrsMax" in start_vars:
self.lrsMax = start_vars['lrsMax']
if "haplotypeAnalystCheck" in start_vars:
self.haplotypeAnalystCheck = start_vars['haplotypeAnalystCheck']
if "startMb" in start_vars: #ZS: This is to ensure showGenes, Legend, etc are checked the first time you open the mapping page, since startMb will only not be set during the first load
if "permCheck" in start_vars:
self.permCheck = "ON"
else:
self.permCheck = False
self.num_perm = int(start_vars['num_perm'])
self.LRSCheck = start_vars['LRSCheck']
if "showSNP" in start_vars:
self.showSNP = start_vars['showSNP']
else:
self.showSNP = False
if "showGenes" in start_vars:
self.showGenes = start_vars['showGenes']
else:
self.showGenes = False
if "viewLegend" in start_vars:
self.viewLegend = start_vars['viewLegend']
else:
self.viewLegend = False
else:
try:
if int(start_vars['num_perm']) > 0:
self.num_perm = int(start_vars['num_perm'])
except:
self.num_perm = 0
if self.num_perm > 0:
self.permCheck = "ON"
else:
self.permCheck = False
self.showSNP = "ON"
self.showGenes = "ON"
self.viewLegend = "ON"
self.dataset.group.get_markers()
if self.mapping_method == "gemma":
self.score_type = "-log(p)"
self.manhattan_plot = True
with Bench("Running GEMMA"):
marker_obs = gemma_mapping.run_gemma(self.dataset, self.samples, self.vals)
results = marker_obs
elif self.mapping_method == "rqtl_plink":
results = self.run_rqtl_plink()
elif self.mapping_method == "rqtl_geno":
self.score_type = "LOD"
self.mapping_scale = "morgan"
self.control_marker = start_vars['control_marker']
self.do_control = start_vars['do_control']
self.dataset.group.genofile = start_vars['genofile']
self.method = start_vars['mapmethod_rqtl_geno']
self.model = start_vars['mapmodel_rqtl_geno']
if start_vars['pair_scan'] == "true":
self.pair_scan = True
if self.permCheck and self.num_perm > 0:
self.perm_output, self.suggestive, self.significant, results = rqtl_mapping.run_rqtl_geno(self.vals, self.dataset, self.method, self.model, self.permCheck, self.num_perm, self.do_control, self.control_marker, self.manhattan_plot, self.pair_scan)
else:
results = rqtl_mapping.run_rqtl_geno(self.vals, self.dataset, self.method, self.model, self.permCheck, self.num_perm, self.do_control, self.control_marker, self.manhattan_plot, self.pair_scan)
elif self.mapping_method == "reaper":
if "startMb" in start_vars: #ZS: Check if first time page loaded, so it can default to ON
if "additiveCheck" in start_vars:
self.additiveCheck = start_vars['additiveCheck']
else:
self.additiveCheck = False
if "bootCheck" in start_vars:
self.bootCheck = "ON"
else:
self.bootCheck = False
self.num_bootstrap = int(start_vars['num_bootstrap'])
else:
self.additiveCheck = "ON"
try:
if int(start_vars['num_bootstrap']) > 0:
self.bootCheck = "ON"
self.num_bootstrap = int(start_vars['num_bootstrap'])
else:
self.bootCheck = False
self.num_bootstrap = 0
except:
self.bootCheck = False
self.num_bootstrap = 0
self.control_marker = start_vars['control_marker']
self.do_control = start_vars['do_control']
self.dataset.group.genofile = start_vars['genofile']
logger.info("Running qtlreaper")
results, self.json_data, self.perm_output, self.suggestive, self.significant, self.bootstrap_results = qtlreaper_mapping.gen_reaper_results(self.this_trait,
self.dataset,
self.samples,
self.json_data,
self.num_perm,
self.bootCheck,
self.num_bootstrap,
self.do_control,
self.control_marker,
self.manhattan_plot)
elif self.mapping_method == "plink":
self.score_type = "-log(p)"
self.manhattan_plot = True
results = plink_mapping.run_plink(self.this_trait, self.dataset, self.species, self.vals, self.maf)
#results = self.run_plink()
elif self.mapping_method == "pylmm":
logger.debug("RUNNING PYLMM")
self.dataset.group.genofile = start_vars['genofile']
if self.num_perm > 0:
self.run_permutations(str(temp_uuid))
results = self.gen_data(str(temp_uuid))
else:
logger.debug("RUNNING NOTHING")
if self.pair_scan == True:
self.qtl_results = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker['chr1'] > 0 or marker['chr1'] == "X" or marker['chr1'] == "X/Y":
if marker['chr1'] > highest_chr or marker['chr1'] == "X" or marker['chr1'] == "X/Y":
highest_chr = marker['chr1']
if 'lod_score' in marker.keys():
self.qtl_results.append(marker)
self.trimmed_markers = results
for qtl in enumerate(self.qtl_results):
self.json_data['chr1'].append(str(qtl['chr1']))
self.json_data['chr2'].append(str(qtl['chr2']))
self.json_data['Mb'].append(qtl['Mb'])
self.json_data['markernames'].append(qtl['name'])
self.js_data = dict(
json_data = self.json_data,
this_trait = self.this_trait.name,
data_set = self.dataset.name,
maf = self.maf,
manhattan_plot = self.manhattan_plot,
mapping_scale = self.mapping_scale,
qtl_results = self.qtl_results
)
else:
self.cutoff = 2
self.qtl_results = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker['chr'] > 0 or marker['chr'] == "X" or marker['chr'] == "X/Y":
if marker['chr'] > highest_chr or marker['chr'] == "X" or marker['chr'] == "X/Y":
highest_chr = marker['chr']
if ('lod_score' in marker.keys()) or ('lrs_value' in marker.keys()):
self.qtl_results.append(marker)
self.trimmed_markers = trim_markers_for_table(results)
if self.mapping_method != "gemma":
self.json_data['chr'] = []
self.json_data['pos'] = []
self.json_data['lod.hk'] = []
self.json_data['markernames'] = []
self.json_data['suggestive'] = self.suggestive
self.json_data['significant'] = self.significant
#Need to convert the QTL objects that qtl reaper returns into a json serializable dictionary
for index, qtl in enumerate(self.qtl_results):
#if index<40:
# logger.debug("lod score is:", qtl['lod_score'])
if qtl['chr'] == highest_chr and highest_chr != "X" and highest_chr != "X/Y":
#logger.debug("changing to X")
self.json_data['chr'].append("X")
else:
self.json_data['chr'].append(str(qtl['chr']))
self.json_data['pos'].append(qtl['Mb'])
if 'lrs_value' in qtl.keys():
self.json_data['lod.hk'].append(str(qtl['lrs_value']))
else:
self.json_data['lod.hk'].append(str(qtl['lod_score']))
self.json_data['markernames'].append(qtl['name'])
#Get chromosome lengths for drawing the interval map plot
chromosome_mb_lengths = {}
self.json_data['chrnames'] = []
for key in self.species.chromosomes.chromosomes.keys():
self.json_data['chrnames'].append([self.species.chromosomes.chromosomes[key].name, self.species.chromosomes.chromosomes[key].mb_length])
chromosome_mb_lengths[key] = self.species.chromosomes.chromosomes[key].mb_length
# logger.debug("json_data:", self.json_data)
self.js_data = dict(
result_score_type = self.score_type,
json_data = self.json_data,
this_trait = self.this_trait.name,
data_set = self.dataset.name,
maf = self.maf,
manhattan_plot = self.manhattan_plot,
mapping_scale = self.mapping_scale,
chromosomes = chromosome_mb_lengths,
qtl_results = self.qtl_results,
num_perm = self.num_perm,
perm_results = self.perm_output,
)
def __init__(self, start_vars, temp_uuid):
helper_functions.get_species_dataset_trait(self, start_vars)
# tempdata = temp_data.TempData(temp_uuid)
self.json_data = {}
self.json_data["lodnames"] = ["lod.hk"]
self.samples = [] # Want only ones with values
self.vals = []
for sample in self.dataset.group.samplelist:
value = start_vars["value:" + sample]
self.samples.append(str(sample))
self.vals.append(value)
self.mapping_method = start_vars["method"]
if start_vars["manhattan_plot"] == "true":
self.manhattan_plot = True
else:
self.manhattan_plot = False
self.maf = start_vars["maf"] # Minor allele frequency
self.suggestive = ""
self.significant = ""
self.pair_scan = False # Initializing this since it is checked in views to determine which template to use
self.score_type = "LRS" # ZS: LRS or LOD
self.dataset.group.get_markers()
if self.mapping_method == "gemma":
self.score_type = "LOD"
included_markers, p_values = gemma_mapping.run_gemma(self.dataset, self.samples, self.vals)
self.dataset.group.get_specified_markers(markers=included_markers)
self.dataset.group.markers.add_pvalues(p_values)
results = self.dataset.group.markers.markers
elif self.mapping_method == "rqtl_plink":
results = self.run_rqtl_plink()
elif self.mapping_method == "rqtl_geno":
self.score_type = "LOD"
if start_vars["num_perm"] == "":
self.num_perm = 0
else:
self.num_perm = start_vars["num_perm"]
self.control = start_vars["control_marker"]
self.do_control = start_vars["do_control"]
print("StartVars:", start_vars)
self.method = start_vars["mapmethod_rqtl_geno"]
self.model = start_vars["mapmodel_rqtl_geno"]
if start_vars["pair_scan"] == "true":
self.pair_scan = True
results = self.run_rqtl_geno()
print("qtl_results:", results)
elif self.mapping_method == "plink":
results = self.run_plink()
# print("qtl_results:", pf(results))
elif self.mapping_method == "pylmm":
print("RUNNING PYLMM")
self.num_perm = start_vars["num_perm"]
if self.num_perm != "":
if int(self.num_perm) > 0:
self.run_permutations(str(temp_uuid))
results = self.gen_data(str(temp_uuid))
else:
print("RUNNING NOTHING")
if self.pair_scan == True:
self.qtl_results = []
highest_chr = 1 # This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker["chr1"] > 0 or marker["chr1"] == "X" or marker["chr1"] == "X/Y":
if marker["chr1"] > highest_chr or marker["chr1"] == "X" or marker["chr1"] == "X/Y":
highest_chr = marker["chr1"]
if "lod_score" in marker:
self.qtl_results.append(marker)
for qtl in enumerate(self.qtl_results):
self.json_data["chr1"].append(str(qtl["chr1"]))
self.json_data["chr2"].append(str(qtl["chr2"]))
self.json_data["Mb"].append(qtl["Mb"])
self.json_data["markernames"].append(qtl["name"])
self.js_data = dict(
json_data=self.json_data,
this_trait=self.this_trait.name,
data_set=self.dataset.name,
maf=self.maf,
manhattan_plot=self.manhattan_plot,
qtl_results=self.qtl_results,
)
else:
self.cutoff = 2
self.qtl_results = []
highest_chr = 1 # This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker["chr"] > 0 or marker["chr"] == "X" or marker["chr"] == "X/Y":
if marker["chr"] > highest_chr or marker["chr"] == "X" or marker["chr"] == "X/Y":
highest_chr = marker["chr"]
if "lod_score" in marker:
self.qtl_results.append(marker)
self.json_data["chr"] = []
self.json_data["pos"] = []
self.json_data["lod.hk"] = []
self.json_data["markernames"] = []
self.json_data["suggestive"] = self.suggestive
self.json_data["significant"] = self.significant
# Need to convert the QTL objects that qtl reaper returns into a json serializable dictionary
for index, qtl in enumerate(self.qtl_results):
if index < 40:
print("lod score is:", qtl["lod_score"])
if qtl["chr"] == highest_chr and highest_chr != "X" and highest_chr != "X/Y":
print("changing to X")
self.json_data["chr"].append("X")
else:
self.json_data["chr"].append(str(qtl["chr"]))
self.json_data["pos"].append(qtl["Mb"])
if "lrs_value" in qtl:
self.json_data["lod.hk"].append(str(qtl["lrs_value"]))
else:
self.json_data["lod.hk"].append(str(qtl["lod_score"]))
self.json_data["markernames"].append(qtl["name"])
# Get chromosome lengths for drawing the interval map plot
chromosome_mb_lengths = {}
self.json_data["chrnames"] = []
for key in self.species.chromosomes.chromosomes.keys():
self.json_data["chrnames"].append(
[
self.species.chromosomes.chromosomes[key].name,
self.species.chromosomes.chromosomes[key].mb_length,
]
)
chromosome_mb_lengths[key] = self.species.chromosomes.chromosomes[key].mb_length
# print("json_data:", self.json_data)
self.js_data = dict(
result_score_type=self.score_type,
json_data=self.json_data,
this_trait=self.this_trait.name,
data_set=self.dataset.name,
maf=self.maf,
manhattan_plot=self.manhattan_plot,
chromosomes=chromosome_mb_lengths,
qtl_results=self.qtl_results,
)
def do_mapping_for_api(start_vars):
assert ('db' in start_vars)
assert ('trait_id' in start_vars)
dataset = data_set.create_dataset(dataset_name=start_vars['db'])
dataset.group.get_markers()
this_trait = GeneralTrait(dataset=dataset, name=start_vars['trait_id'])
this_trait = retrieve_sample_data(this_trait, dataset)
samples = []
vals = []
for sample in dataset.group.samplelist:
in_trait_data = False
for item in this_trait.data:
if this_trait.data[item].name == sample:
value = str(this_trait.data[item].value)
samples.append(item)
vals.append(value)
in_trait_data = True
break
if not in_trait_data:
vals.append("x")
mapping_params = initialize_parameters(start_vars, dataset, this_trait)
covariates = "" #ZS: It seems to take an empty string as default. This should probably be changed.
if mapping_params['mapping_method'] == "gemma":
header_row = ["name", "chr", "Mb", "lod_score", "p_value"]
if mapping_params[
'use_loco'] == "True": #ZS: gemma_mapping returns both results and the filename for LOCO, so need to only grab the former for api
result_markers = gemma_mapping.run_gemma(
this_trait, dataset, samples, vals, covariates,
mapping_params['use_loco'], mapping_params['maf'])[0]
else:
result_markers = gemma_mapping.run_gemma(
this_trait, dataset, samples, vals, covariates,
mapping_params['use_loco'], mapping_params['maf'])
elif mapping_params['mapping_method'] == "rqtl":
header_row = ["name", "chr", "cM", "lod_score"]
if mapping_params['num_perm'] > 0:
_sperm_output, _suggestive, _significant, result_markers = rqtl_mapping.run_rqtl_geno(
vals, dataset, mapping_params['rqtl_method'],
mapping_params['rqtl_model'], mapping_params['perm_check'],
mapping_params['num_perm'], mapping_params['do_control'],
mapping_params['control_marker'],
mapping_params['manhattan_plot'], mapping_params['pair_scan'])
else:
result_markers = rqtl_mapping.run_rqtl_geno(
vals, dataset, mapping_params['rqtl_method'],
mapping_params['rqtl_model'], mapping_params['perm_check'],
mapping_params['num_perm'], mapping_params['do_control'],
mapping_params['control_marker'],
mapping_params['manhattan_plot'], mapping_params['pair_scan'])
if mapping_params['limit_to']:
result_markers = result_markers[:mapping_params['limit_to']]
if mapping_params['format'] == "csv":
output_rows = []
output_rows.append(header_row)
for marker in result_markers:
this_row = [marker[header] for header in header_row]
output_rows.append(this_row)
return output_rows, mapping_params['format']
elif mapping_params['format'] == "json":
return result_markers, mapping_params['format']
else:
return result_markers, None
def __init__(self, start_vars, temp_uuid):
helper_functions.get_species_dataset_trait(self, start_vars)
#tempdata = temp_data.TempData(temp_uuid)
self.temp_uuid = temp_uuid #needed to pass temp_uuid to gn1 mapping code (marker_regression_gn1.py)
self.json_data = {}
self.json_data['lodnames'] = ['lod.hk']
self.samples = [] # Want only ones with values
self.vals = []
for sample in self.dataset.group.samplelist:
value = start_vars['value:' + sample]
self.samples.append(str(sample))
self.vals.append(value)
self.mapping_method = start_vars['method']
if start_vars['manhattan_plot'] == "true":
self.manhattan_plot = True
else:
self.manhattan_plot = False
self.maf = start_vars['maf'] # Minor allele frequency
self.suggestive = ""
self.significant = ""
self.pair_scan = False # Initializing this since it is checked in views to determine which template to use
self.score_type = "LRS" #ZS: LRS or LOD
self.mapping_scale = "physic"
self.bootstrap_results = []
#ZS: This is passed to GN1 code for single chr mapping
self.selected_chr = -1
if "selected_chr" in start_vars:
if int(
start_vars['selected_chr']
) != -1: #ZS: Needs to be -1 if showing full map; there's probably a better way to fix this
self.selected_chr = int(start_vars['selected_chr']) + 1
else:
self.selected_chr = int(start_vars['selected_chr'])
if "startMb" in start_vars:
self.startMb = start_vars['startMb']
if "endMb" in start_vars:
self.endMb = start_vars['endMb']
if "graphWidth" in start_vars:
self.graphWidth = start_vars['graphWidth']
if "lrsMax" in start_vars:
self.lrsMax = start_vars['lrsMax']
if "haplotypeAnalystCheck" in start_vars:
self.haplotypeAnalystCheck = start_vars['haplotypeAnalystCheck']
if "startMb" in start_vars: #ZS: This is to ensure showGenes, Legend, etc are checked the first time you open the mapping page, since startMb will only not be set during the first load
if "permCheck" in start_vars:
self.permCheck = "ON"
else:
self.permCheck = False
self.num_perm = int(start_vars['num_perm'])
self.LRSCheck = start_vars['LRSCheck']
if "showSNP" in start_vars:
self.showSNP = start_vars['showSNP']
else:
self.showSNP = False
if "showGenes" in start_vars:
self.showGenes = start_vars['showGenes']
else:
self.showGenes = False
if "viewLegend" in start_vars:
self.viewLegend = start_vars['viewLegend']
else:
self.viewLegend = False
else:
try:
if int(start_vars['num_perm']) > 0:
self.num_perm = int(start_vars['num_perm'])
else:
self.num_perm = 0
except:
self.num_perm = 0
self.LRSCheck = self.score_type
self.permCheck = "ON"
self.showSNP = "ON"
self.showGenes = "ON"
self.viewLegend = "ON"
self.dataset.group.get_markers()
if self.mapping_method == "gemma":
self.score_type = "LOD"
self.manhattan_plot = True
with Bench("Running GEMMA"):
included_markers, p_values = gemma_mapping.run_gemma(
self.dataset, self.samples, self.vals)
with Bench("Getting markers from csv"):
marker_obs = get_markers_from_csv(included_markers, p_values,
self.dataset.group.name)
results = marker_obs
elif self.mapping_method == "rqtl_plink":
results = self.run_rqtl_plink()
elif self.mapping_method == "rqtl_geno":
self.score_type = "LOD"
self.mapping_scale = "morgan"
self.control_marker = start_vars['control_marker']
self.do_control = start_vars['do_control']
self.method = start_vars['mapmethod_rqtl_geno']
self.model = start_vars['mapmodel_rqtl_geno']
if start_vars['pair_scan'] == "true":
self.pair_scan = True
results = self.run_rqtl_geno()
elif self.mapping_method == "reaper":
if "startMb" in start_vars: #ZS: Check if first time page loaded, so it can default to ON
if "additiveCheck" in start_vars:
self.additiveCheck = start_vars['additiveCheck']
else:
self.additiveCheck = False
if "bootCheck" in start_vars:
self.bootCheck = "ON"
else:
self.bootCheck = False
self.num_bootstrap = int(start_vars['num_bootstrap'])
else:
self.additiveCheck = "ON"
try:
if int(start_vars['num_bootstrap']) > 0:
self.bootCheck = "ON"
self.num_bootstrap = int(start_vars['num_bootstrap'])
else:
self.bootCheck = False
self.num_bootstrap = 0
except:
self.bootCheck = False
self.num_bootstrap = 0
self.control_marker = start_vars['control_marker']
self.do_control = start_vars['do_control']
results = self.gen_reaper_results()
elif self.mapping_method == "plink":
results = self.run_plink()
elif self.mapping_method == "pylmm":
print("RUNNING PYLMM")
if self.num_perm > 0:
self.run_permutations(str(temp_uuid))
results = self.gen_data(str(temp_uuid))
else:
print("RUNNING NOTHING")
if self.pair_scan == True:
self.qtl_results = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker['chr1'] > 0 or marker['chr1'] == "X" or marker[
'chr1'] == "X/Y":
if marker['chr1'] > highest_chr or marker[
'chr1'] == "X" or marker['chr1'] == "X/Y":
highest_chr = marker['chr1']
if 'lod_score' in marker.keys():
self.qtl_results.append(marker)
for qtl in enumerate(self.qtl_results):
self.json_data['chr1'].append(str(qtl['chr1']))
self.json_data['chr2'].append(str(qtl['chr2']))
self.json_data['Mb'].append(qtl['Mb'])
self.json_data['markernames'].append(qtl['name'])
self.js_data = dict(
json_data=self.json_data,
this_trait=self.this_trait.name,
data_set=self.dataset.name,
maf=self.maf,
manhattan_plot=self.manhattan_plot,
mapping_scale=self.mapping_scale,
qtl_results=self.qtl_results,
)
else:
self.cutoff = 2
self.qtl_results = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker['chr'] > 0 or marker['chr'] == "X" or marker[
'chr'] == "X/Y":
if marker['chr'] > highest_chr or marker[
'chr'] == "X" or marker['chr'] == "X/Y":
highest_chr = marker['chr']
if ('lod_score' in marker.keys()) or ('lrs_value'
in marker.keys()):
self.qtl_results.append(marker)
self.trimmed_markers = trim_markers_for_table(results)
self.json_data['chr'] = []
self.json_data['pos'] = []
self.json_data['lod.hk'] = []
self.json_data['markernames'] = []
self.json_data['suggestive'] = self.suggestive
self.json_data['significant'] = self.significant
#Need to convert the QTL objects that qtl reaper returns into a json serializable dictionary
for index, qtl in enumerate(self.qtl_results):
#if index<40:
# print("lod score is:", qtl['lod_score'])
if qtl['chr'] == highest_chr and highest_chr != "X" and highest_chr != "X/Y":
#print("changing to X")
self.json_data['chr'].append("X")
else:
self.json_data['chr'].append(str(qtl['chr']))
self.json_data['pos'].append(qtl['Mb'])
if 'lrs_value' in qtl.keys():
self.json_data['lod.hk'].append(str(qtl['lrs_value']))
else:
self.json_data['lod.hk'].append(str(qtl['lod_score']))
self.json_data['markernames'].append(qtl['name'])
#Get chromosome lengths for drawing the interval map plot
chromosome_mb_lengths = {}
self.json_data['chrnames'] = []
for key in self.species.chromosomes.chromosomes.keys():
self.json_data['chrnames'].append([
self.species.chromosomes.chromosomes[key].name,
self.species.chromosomes.chromosomes[key].mb_length
])
chromosome_mb_lengths[
key] = self.species.chromosomes.chromosomes[key].mb_length
# print("json_data:", self.json_data)
self.js_data = dict(
result_score_type=self.score_type,
json_data=self.json_data,
this_trait=self.this_trait.name,
data_set=self.dataset.name,
maf=self.maf,
manhattan_plot=self.manhattan_plot,
mapping_scale=self.mapping_scale,
chromosomes=chromosome_mb_lengths,
qtl_results=self.qtl_results,
)
def __init__(self, start_vars, temp_uuid):
helper_functions.get_species_dataset_trait(self, start_vars)
self.temp_uuid = temp_uuid #needed to pass temp_uuid to gn1 mapping code (marker_regression_gn1.py)
#ZS: Needed to zoom in or remap temp traits like PCA traits
if "temp_trait" in start_vars and start_vars['temp_trait'] != "False":
self.temp_trait = "True"
self.group = self.dataset.group.name
self.json_data = {}
self.json_data['lodnames'] = ['lod.hk']
#ZS: Sometimes a group may have a genofile that only includes a subset of samples
genofile_samplelist = []
if 'genofile' in start_vars:
if start_vars['genofile'] != "":
self.genofile_string = start_vars['genofile']
self.dataset.group.genofile = self.genofile_string.split(
":")[0]
genofile_samplelist = get_genofile_samplelist(self.dataset)
all_samples_ordered = self.dataset.group.all_samples_ordered()
self.vals = []
self.samples = []
self.sample_vals = start_vars['sample_vals']
sample_val_dict = json.loads(self.sample_vals)
samples = sample_val_dict.keys()
if (len(genofile_samplelist) != 0):
for sample in genofile_samplelist:
self.samples.append(sample)
if sample in samples:
self.vals.append(sample_val_dict[sample])
else:
self.vals.append("x")
else:
for sample in self.dataset.group.samplelist:
if sample in samples:
self.vals.append(sample_val_dict[sample])
self.samples.append(sample)
if 'n_samples' in start_vars:
self.n_samples = start_vars['n_samples']
else:
self.n_samples = len([val for val in self.vals if val != "x"])
#ZS: Check if genotypes exist in the DB in order to create links for markers
self.geno_db_exists = geno_db_exists(self.dataset)
self.mapping_method = start_vars['method']
if "results_path" in start_vars:
self.mapping_results_path = start_vars['results_path']
else:
mapping_results_filename = self.dataset.group.name + "_" + ''.join(
random.choice(string.ascii_uppercase + string.digits)
for _ in range(6))
self.mapping_results_path = "{}{}.csv".format(
webqtlConfig.GENERATED_IMAGE_DIR, mapping_results_filename)
self.manhattan_plot = False
if 'manhattan_plot' in start_vars:
if start_vars['manhattan_plot'].lower() != "false":
self.color_scheme = "alternating"
if "color_scheme" in start_vars:
self.color_scheme = start_vars['color_scheme']
if self.color_scheme == "single":
self.manhattan_single_color = start_vars[
'manhattan_single_color']
self.manhattan_plot = True
self.maf = start_vars['maf'] # Minor allele frequency
if "use_loco" in start_vars:
self.use_loco = start_vars['use_loco']
else:
self.use_loco = None
self.suggestive = ""
self.significant = ""
self.pair_scan = False # Initializing this since it is checked in views to determine which template to use
if 'transform' in start_vars:
self.transform = start_vars['transform']
else:
self.transform = ""
self.score_type = "LRS" #ZS: LRS or LOD
self.mapping_scale = "physic"
if "mapping_scale" in start_vars:
self.mapping_scale = start_vars['mapping_scale']
self.num_perm = 0
self.perm_output = []
self.bootstrap_results = []
self.covariates = start_vars[
'covariates'] if "covariates" in start_vars else ""
self.categorical_vars = []
#ZS: This is passed to GN1 code for single chr mapping
self.selected_chr = -1
if "selected_chr" in start_vars:
if int(
start_vars['selected_chr']
) != -1: #ZS: Needs to be -1 if showing full map; there's probably a better way to fix this
self.selected_chr = int(start_vars['selected_chr']) + 1
else:
self.selected_chr = int(start_vars['selected_chr'])
if "startMb" in start_vars:
self.startMb = start_vars['startMb']
if "endMb" in start_vars:
self.endMb = start_vars['endMb']
if "graphWidth" in start_vars:
self.graphWidth = start_vars['graphWidth']
if "lrsMax" in start_vars:
self.lrsMax = start_vars['lrsMax']
if "haplotypeAnalystCheck" in start_vars:
self.haplotypeAnalystCheck = start_vars['haplotypeAnalystCheck']
if "startMb" in start_vars: #ZS: This is to ensure showGenes, Legend, etc are checked the first time you open the mapping page, since startMb will only not be set during the first load
if "permCheck" in start_vars:
self.permCheck = "ON"
else:
self.permCheck = False
self.num_perm = int(start_vars['num_perm'])
self.LRSCheck = start_vars['LRSCheck']
if "showSNP" in start_vars:
self.showSNP = start_vars['showSNP']
else:
self.showSNP = False
if "showGenes" in start_vars:
self.showGenes = start_vars['showGenes']
else:
self.showGenes = False
if "viewLegend" in start_vars:
self.viewLegend = start_vars['viewLegend']
else:
self.viewLegend = False
else:
try:
if int(start_vars['num_perm']) > 0:
self.num_perm = int(start_vars['num_perm'])
except:
self.num_perm = 0
if self.num_perm > 0:
self.permCheck = "ON"
else:
self.permCheck = False
self.showSNP = "ON"
self.showGenes = "ON"
self.viewLegend = "ON"
#self.dataset.group.get_markers()
if self.mapping_method == "gemma":
self.first_run = True
self.output_files = None
if 'output_files' in start_vars:
self.output_files = start_vars['output_files']
if 'first_run' in start_vars: #ZS: check if first run so existing result files can be used if it isn't (for example zooming on a chromosome, etc)
self.first_run = False
self.score_type = "-logP"
self.manhattan_plot = True
with Bench("Running GEMMA"):
if self.use_loco == "True":
marker_obs, self.output_files = gemma_mapping.run_gemma(
self.this_trait, self.dataset, self.samples, self.vals,
self.covariates, self.use_loco, self.maf,
self.first_run, self.output_files)
else:
marker_obs, self.output_files = gemma_mapping.run_gemma(
self.this_trait, self.dataset, self.samples, self.vals,
self.covariates, self.use_loco, self.maf,
self.first_run, self.output_files)
results = marker_obs
elif self.mapping_method == "rqtl_plink":
results = self.run_rqtl_plink()
elif self.mapping_method == "rqtl_geno":
perm_strata = []
if "perm_strata" in start_vars and "categorical_vars" in start_vars:
self.categorical_vars = start_vars["categorical_vars"].split(
",")
if len(self.categorical_vars
) and start_vars["perm_strata"] == "True":
primary_samples = SampleList(dataset=self.dataset,
sample_names=self.samples,
this_trait=self.this_trait)
perm_strata = get_perm_strata(self.this_trait,
primary_samples,
self.categorical_vars,
self.samples)
self.score_type = "LOD"
self.control_marker = start_vars['control_marker']
self.do_control = start_vars['do_control']
if 'mapmethod_rqtl_geno' in start_vars:
self.method = start_vars['mapmethod_rqtl_geno']
else:
self.method = "em"
self.model = start_vars['mapmodel_rqtl_geno']
#if start_vars['pair_scan'] == "true":
# self.pair_scan = True
if self.permCheck and self.num_perm > 0:
self.perm_output, self.suggestive, self.significant, results = rqtl_mapping.run_rqtl_geno(
self.vals, self.samples, self.dataset, self.mapping_scale,
self.method, self.model, self.permCheck, self.num_perm,
perm_strata, self.do_control, self.control_marker,
self.manhattan_plot, self.pair_scan, self.covariates)
else:
results = rqtl_mapping.run_rqtl_geno(
self.vals, self.samples, self.dataset, self.mapping_scale,
self.method, self.model, self.permCheck, self.num_perm,
perm_strata, self.do_control, self.control_marker,
self.manhattan_plot, self.pair_scan, self.covariates)
elif self.mapping_method == "reaper":
if "startMb" in start_vars: #ZS: Check if first time page loaded, so it can default to ON
if "additiveCheck" in start_vars:
self.additiveCheck = start_vars['additiveCheck']
else:
self.additiveCheck = False
if "bootCheck" in start_vars:
self.bootCheck = "ON"
else:
self.bootCheck = False
self.num_bootstrap = int(start_vars['num_bootstrap'])
else:
self.additiveCheck = "ON"
try:
if int(start_vars['num_bootstrap']) > 0:
self.bootCheck = "ON"
self.num_bootstrap = int(start_vars['num_bootstrap'])
else:
self.bootCheck = False
self.num_bootstrap = 0
except:
self.bootCheck = False
self.num_bootstrap = 0
self.reaper_version = start_vars['reaper_version']
self.control_marker = start_vars['control_marker']
self.do_control = start_vars['do_control']
logger.info("Running qtlreaper")
if self.reaper_version == "new":
self.first_run = True
self.output_files = None
if 'first_run' in start_vars: #ZS: check if first run so existing result files can be used if it isn't (for example zooming on a chromosome, etc)
self.first_run = False
if 'output_files' in start_vars:
self.output_files = start_vars['output_files'].split(
",")
results, self.perm_output, self.suggestive, self.significant, self.bootstrap_results, self.output_files = qtlreaper_mapping.run_reaper(
self.this_trait, self.dataset, self.samples, self.vals,
self.json_data, self.num_perm, self.bootCheck,
self.num_bootstrap, self.do_control, self.control_marker,
self.manhattan_plot, self.first_run, self.output_files)
else:
results, self.json_data, self.perm_output, self.suggestive, self.significant, self.bootstrap_results = qtlreaper_mapping.run_original_reaper(
self.this_trait, self.dataset, self.samples, self.vals,
self.json_data, self.num_perm, self.bootCheck,
self.num_bootstrap, self.do_control, self.control_marker,
self.manhattan_plot)
elif self.mapping_method == "plink":
self.score_type = "-logP"
self.manhattan_plot = True
results = plink_mapping.run_plink(self.this_trait, self.dataset,
self.species, self.vals,
self.maf)
#results = self.run_plink()
else:
logger.debug("RUNNING NOTHING")
self.no_results = False
if len(results) == 0:
self.no_results = True
else:
if self.pair_scan == True:
self.qtl_results = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker['chr1'] > 0 or marker['chr1'] == "X" or marker[
'chr1'] == "X/Y":
if marker['chr1'] > highest_chr or marker[
'chr1'] == "X" or marker['chr1'] == "X/Y":
highest_chr = marker['chr1']
if 'lod_score' in list(marker.keys()):
self.qtl_results.append(marker)
self.trimmed_markers = results
for qtl in enumerate(self.qtl_results):
self.json_data['chr1'].append(str(qtl['chr1']))
self.json_data['chr2'].append(str(qtl['chr2']))
self.json_data['Mb'].append(qtl['Mb'])
self.json_data['markernames'].append(qtl['name'])
self.js_data = dict(json_data=self.json_data,
this_trait=self.this_trait.name,
data_set=self.dataset.name,
maf=self.maf,
manhattan_plot=self.manhattan_plot,
mapping_scale=self.mapping_scale,
qtl_results=self.qtl_results)
else:
self.qtl_results = []
self.results_for_browser = []
self.annotations_for_browser = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if 'Mb' in marker:
this_ps = marker['Mb'] * 1000000
else:
this_ps = marker['cM'] * 1000000
browser_marker = dict(chr=str(marker['chr']),
rs=marker['name'],
ps=this_ps,
url="/show_trait?trait_id=" +
marker['name'] + "&dataset=" +
self.dataset.group.name + "Geno")
if self.geno_db_exists == "True":
annot_marker = dict(name=str(marker['name']),
chr=str(marker['chr']),
rs=marker['name'],
pos=this_ps,
url="/show_trait?trait_id=" +
marker['name'] + "&dataset=" +
self.dataset.group.name + "Geno")
else:
annot_marker = dict(name=str(marker['name']),
chr=str(marker['chr']),
rs=marker['name'],
pos=this_ps)
if 'lrs_value' in marker and marker['lrs_value'] > 0:
browser_marker['p_wald'] = 10**-(marker['lrs_value'] /
4.61)
elif 'lod_score' in marker and marker['lod_score'] > 0:
browser_marker['p_wald'] = 10**-(marker['lod_score'])
else:
browser_marker['p_wald'] = 0
self.results_for_browser.append(browser_marker)
self.annotations_for_browser.append(annot_marker)
if str(marker['chr']) > '0' or str(
marker['chr']) == "X" or str(
marker['chr']) == "X/Y":
if str(marker['chr']) > str(highest_chr) or str(
marker['chr']) == "X" or str(
marker['chr']) == "X/Y":
highest_chr = marker['chr']
if ('lod_score'
in marker.keys()) or ('lrs_value'
in marker.keys()):
if 'Mb' in marker.keys():
marker['display_pos'] = "Chr" + str(
marker['chr']) + ": " + "{:.6f}".format(
marker['Mb'])
elif 'cM' in marker.keys():
marker['display_pos'] = "Chr" + str(
marker['chr']) + ": " + "{:.3f}".format(
marker['cM'])
else:
marker['display_pos'] = "N/A"
self.qtl_results.append(marker)
total_markers = len(self.qtl_results)
with Bench("Exporting Results"):
export_mapping_results(self.dataset, self.this_trait,
self.qtl_results,
self.mapping_results_path,
self.mapping_scale, self.score_type)
with Bench("Trimming Markers for Figure"):
if len(self.qtl_results) > 30000:
self.qtl_results = trim_markers_for_figure(
self.qtl_results)
self.results_for_browser = trim_markers_for_figure(
self.results_for_browser)
filtered_annotations = []
for marker in self.results_for_browser:
for annot_marker in self.annotations_for_browser:
if annot_marker['rs'] == marker['rs']:
filtered_annotations.append(annot_marker)
break
self.annotations_for_browser = filtered_annotations
browser_files = write_input_for_browser(
self.dataset, self.results_for_browser,
self.annotations_for_browser)
else:
browser_files = write_input_for_browser(
self.dataset, self.results_for_browser,
self.annotations_for_browser)
with Bench("Trimming Markers for Table"):
self.trimmed_markers = trim_markers_for_table(results)
chr_lengths = get_chr_lengths(self.mapping_scale,
self.mapping_method,
self.dataset, self.qtl_results)
#ZS: For zooming into genome browser, need to pass chromosome name instead of number
if self.dataset.group.species == "mouse":
if self.selected_chr == 20:
this_chr = "X"
else:
this_chr = str(self.selected_chr)
elif self.dataset.group.species == "rat":
if self.selected_chr == 21:
this_chr = "X"
else:
this_chr = str(self.selected_chr)
else:
if self.selected_chr == 22:
this_chr = "X"
elif self.selected_chr == 23:
this_chr = "Y"
else:
this_chr = str(self.selected_chr)
if self.mapping_method != "gemma":
if self.score_type == "LRS":
significant_for_browser = self.significant / 4.61
else:
significant_for_browser = self.significant
self.js_data = dict(
#result_score_type = self.score_type,
#this_trait = self.this_trait.name,
#data_set = self.dataset.name,
#maf = self.maf,
#manhattan_plot = self.manhattan_plot,
#mapping_scale = self.mapping_scale,
#chromosomes = chromosome_mb_lengths,
#qtl_results = self.qtl_results,
categorical_vars=self.categorical_vars,
chr_lengths=chr_lengths,
num_perm=self.num_perm,
perm_results=self.perm_output,
significant=significant_for_browser,
browser_files=browser_files,
selected_chr=this_chr,
total_markers=total_markers)
else:
self.js_data = dict(chr_lengths=chr_lengths,
browser_files=browser_files,
selected_chr=this_chr,
total_markers=total_markers)
def __init__(self, start_vars, temp_uuid):
helper_functions.get_species_dataset_trait(self, start_vars)
self.temp_uuid = temp_uuid #needed to pass temp_uuid to gn1 mapping code (marker_regression_gn1.py)
self.json_data = {}
self.json_data['lodnames'] = ['lod.hk']
self.samples = [] # Want only ones with values
self.vals = []
all_samples_ordered = self.dataset.group.all_samples_ordered()
primary_sample_names = list(all_samples_ordered)
for sample in self.dataset.group.samplelist:
# sample is actually the name of an individual
in_trait_data = False
for item in self.this_trait.data:
if self.this_trait.data[item].name == sample:
value = start_vars['value:' +
self.this_trait.data[item].name]
self.samples.append(self.this_trait.data[item].name)
self.vals.append(value)
in_trait_data = True
break
if not in_trait_data:
value = start_vars.get('value:' + sample)
if value:
self.samples.append(sample)
self.vals.append(value)
#ZS: Check if genotypes exist in the DB in order to create links for markers
if "geno_db_exists" in start_vars:
self.geno_db_exists = start_vars['geno_db_exists']
else:
try:
self.geno_db_exists = "True"
except:
self.geno_db_exists = "False"
self.mapping_method = start_vars['method']
if "results_path" in start_vars:
self.mapping_results_path = start_vars['results_path']
else:
mapping_results_filename = self.dataset.group.name + "_" + ''.join(
random.choice(string.ascii_uppercase + string.digits)
for _ in range(6))
self.mapping_results_path = "{}{}.csv".format(
webqtlConfig.GENERATED_IMAGE_DIR, mapping_results_filename)
if start_vars['manhattan_plot'] == "True":
self.manhattan_plot = True
else:
self.manhattan_plot = False
self.maf = start_vars['maf'] # Minor allele frequency
if "use_loco" in start_vars:
self.use_loco = start_vars['use_loco']
else:
self.use_loco = None
self.suggestive = ""
self.significant = ""
self.pair_scan = False # Initializing this since it is checked in views to determine which template to use
self.score_type = "LRS" #ZS: LRS or LOD
self.mapping_scale = "physic"
self.num_perm = 0
self.perm_output = []
self.bootstrap_results = []
self.covariates = start_vars[
'covariates'] if "covariates" in start_vars else None
#ZS: This is passed to GN1 code for single chr mapping
self.selected_chr = -1
if "selected_chr" in start_vars:
if int(
start_vars['selected_chr']
) != -1: #ZS: Needs to be -1 if showing full map; there's probably a better way to fix this
self.selected_chr = int(start_vars['selected_chr']) + 1
else:
self.selected_chr = int(start_vars['selected_chr'])
if "startMb" in start_vars:
self.startMb = start_vars['startMb']
if "endMb" in start_vars:
self.endMb = start_vars['endMb']
if "graphWidth" in start_vars:
self.graphWidth = start_vars['graphWidth']
if "lrsMax" in start_vars:
self.lrsMax = start_vars['lrsMax']
if "haplotypeAnalystCheck" in start_vars:
self.haplotypeAnalystCheck = start_vars['haplotypeAnalystCheck']
if "startMb" in start_vars: #ZS: This is to ensure showGenes, Legend, etc are checked the first time you open the mapping page, since startMb will only not be set during the first load
if "permCheck" in start_vars:
self.permCheck = "ON"
else:
self.permCheck = False
self.num_perm = int(start_vars['num_perm'])
self.LRSCheck = start_vars['LRSCheck']
if "showSNP" in start_vars:
self.showSNP = start_vars['showSNP']
else:
self.showSNP = False
if "showGenes" in start_vars:
self.showGenes = start_vars['showGenes']
else:
self.showGenes = False
if "viewLegend" in start_vars:
self.viewLegend = start_vars['viewLegend']
else:
self.viewLegend = False
else:
try:
if int(start_vars['num_perm']) > 0:
self.num_perm = int(start_vars['num_perm'])
except:
self.num_perm = 0
if self.num_perm > 0:
self.permCheck = "ON"
else:
self.permCheck = False
self.showSNP = "ON"
self.showGenes = "ON"
self.viewLegend = "ON"
if 'genofile' in start_vars:
if start_vars['genofile'] != "":
self.genofile_string = start_vars['genofile']
self.dataset.group.genofile = self.genofile_string.split(
":")[0]
self.dataset.group.get_markers()
if self.mapping_method == "gemma":
self.score_type = "-log(p)"
self.manhattan_plot = True
with Bench("Running GEMMA"):
marker_obs = gemma_mapping.run_gemma(self.dataset,
self.samples, self.vals,
self.covariates,
self.use_loco)
results = marker_obs
elif self.mapping_method == "rqtl_plink":
results = self.run_rqtl_plink()
elif self.mapping_method == "rqtl_geno":
self.score_type = "LOD"
self.mapping_scale = "morgan"
self.control_marker = start_vars['control_marker']
self.do_control = start_vars['do_control']
self.method = start_vars['mapmethod_rqtl_geno']
self.model = start_vars['mapmodel_rqtl_geno']
#if start_vars['pair_scan'] == "true":
# self.pair_scan = True
if self.permCheck and self.num_perm > 0:
self.perm_output, self.suggestive, self.significant, results = rqtl_mapping.run_rqtl_geno(
self.vals, self.dataset, self.method, self.model,
self.permCheck, self.num_perm, self.do_control,
self.control_marker, self.manhattan_plot, self.pair_scan)
else:
results = rqtl_mapping.run_rqtl_geno(
self.vals, self.dataset, self.method, self.model,
self.permCheck, self.num_perm, self.do_control,
self.control_marker, self.manhattan_plot, self.pair_scan)
elif self.mapping_method == "reaper":
if "startMb" in start_vars: #ZS: Check if first time page loaded, so it can default to ON
if "additiveCheck" in start_vars:
self.additiveCheck = start_vars['additiveCheck']
else:
self.additiveCheck = False
if "bootCheck" in start_vars:
self.bootCheck = "ON"
else:
self.bootCheck = False
self.num_bootstrap = int(start_vars['num_bootstrap'])
else:
self.additiveCheck = "ON"
try:
if int(start_vars['num_bootstrap']) > 0:
self.bootCheck = "ON"
self.num_bootstrap = int(start_vars['num_bootstrap'])
else:
self.bootCheck = False
self.num_bootstrap = 0
except:
self.bootCheck = False
self.num_bootstrap = 0
self.control_marker = start_vars['control_marker']
self.do_control = start_vars['do_control']
logger.info("Running qtlreaper")
results, self.json_data, self.perm_output, self.suggestive, self.significant, self.bootstrap_results = qtlreaper_mapping.gen_reaper_results(
self.this_trait, self.dataset, self.samples, self.vals,
self.json_data, self.num_perm, self.bootCheck,
self.num_bootstrap, self.do_control, self.control_marker,
self.manhattan_plot)
elif self.mapping_method == "plink":
self.score_type = "-log(p)"
self.manhattan_plot = True
results = plink_mapping.run_plink(self.this_trait, self.dataset,
self.species, self.vals,
self.maf)
#results = self.run_plink()
elif self.mapping_method == "pylmm":
logger.debug("RUNNING PYLMM")
if self.num_perm > 0:
self.run_permutations(str(temp_uuid))
results = self.gen_data(str(temp_uuid))
else:
logger.debug("RUNNING NOTHING")
if self.pair_scan == True:
self.qtl_results = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker['chr1'] > 0 or marker['chr1'] == "X" or marker[
'chr1'] == "X/Y":
if marker['chr1'] > highest_chr or marker[
'chr1'] == "X" or marker['chr1'] == "X/Y":
highest_chr = marker['chr1']
if 'lod_score' in marker.keys():
self.qtl_results.append(marker)
self.trimmed_markers = results
for qtl in enumerate(self.qtl_results):
self.json_data['chr1'].append(str(qtl['chr1']))
self.json_data['chr2'].append(str(qtl['chr2']))
self.json_data['Mb'].append(qtl['Mb'])
self.json_data['markernames'].append(qtl['name'])
self.js_data = dict(json_data=self.json_data,
this_trait=self.this_trait.name,
data_set=self.dataset.name,
maf=self.maf,
manhattan_plot=self.manhattan_plot,
mapping_scale=self.mapping_scale,
qtl_results=self.qtl_results)
else:
self.qtl_results = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker['chr'] > 0 or marker['chr'] == "X" or marker[
'chr'] == "X/Y":
if marker['chr'] > highest_chr or marker[
'chr'] == "X" or marker['chr'] == "X/Y":
highest_chr = marker['chr']
if ('lod_score' in marker.keys()) or ('lrs_value'
in marker.keys()):
self.qtl_results.append(marker)
with Bench("Exporting Results"):
export_mapping_results(self.dataset, self.this_trait,
self.qtl_results,
self.mapping_results_path,
self.mapping_scale, self.score_type)
with Bench("Trimming Markers for Figure"):
if len(self.qtl_results) > 30000:
self.qtl_results = trim_markers_for_figure(
self.qtl_results)
with Bench("Trimming Markers for Table"):
self.trimmed_markers = trim_markers_for_table(results)
if self.mapping_method != "gemma":
self.json_data['chr'] = []
self.json_data['pos'] = []
self.json_data['lod.hk'] = []
self.json_data['markernames'] = []
self.json_data['suggestive'] = self.suggestive
self.json_data['significant'] = self.significant
#Need to convert the QTL objects that qtl reaper returns into a json serializable dictionary
for index, qtl in enumerate(self.qtl_results):
#if index<40:
# logger.debug("lod score is:", qtl['lod_score'])
if qtl['chr'] == highest_chr and highest_chr != "X" and highest_chr != "X/Y":
#logger.debug("changing to X")
self.json_data['chr'].append("X")
else:
self.json_data['chr'].append(str(qtl['chr']))
self.json_data['pos'].append(qtl['Mb'])
if 'lrs_value' in qtl.keys():
self.json_data['lod.hk'].append(str(qtl['lrs_value']))
else:
self.json_data['lod.hk'].append(str(qtl['lod_score']))
self.json_data['markernames'].append(qtl['name'])
#Get chromosome lengths for drawing the interval map plot
chromosome_mb_lengths = {}
self.json_data['chrnames'] = []
for key in self.species.chromosomes.chromosomes.keys():
self.json_data['chrnames'].append([
self.species.chromosomes.chromosomes[key].name,
self.species.chromosomes.chromosomes[key].mb_length
])
chromosome_mb_lengths[
key] = self.species.chromosomes.chromosomes[
key].mb_length
self.js_data = dict(
result_score_type=self.score_type,
json_data=self.json_data,
this_trait=self.this_trait.name,
data_set=self.dataset.name,
maf=self.maf,
manhattan_plot=self.manhattan_plot,
mapping_scale=self.mapping_scale,
chromosomes=chromosome_mb_lengths,
qtl_results=self.qtl_results,
num_perm=self.num_perm,
perm_results=self.perm_output,
)
def __init__(self, start_vars, temp_uuid):
helper_functions.get_species_dataset_trait(self, start_vars)
#tempdata = temp_data.TempData(temp_uuid)
self.temp_uuid = temp_uuid #needed to pass temp_uuid to gn1 mapping code (marker_regression_gn1.py)
self.json_data = {}
self.json_data['lodnames'] = ['lod.hk']
self.samples = [] # Want only ones with values
self.vals = []
for sample in self.dataset.group.samplelist:
value = start_vars['value:' + sample]
self.samples.append(str(sample))
self.vals.append(value)
self.mapping_method = start_vars['method']
if start_vars['manhattan_plot'] == "true":
self.manhattan_plot = True
else:
self.manhattan_plot = False
self.maf = start_vars['maf'] # Minor allele frequency
self.suggestive = ""
self.significant = ""
self.pair_scan = False # Initializing this since it is checked in views to determine which template to use
self.score_type = "LRS" #ZS: LRS or LOD
self.mapping_scale = "physic"
self.num_perm = 0
#ZS: This is passed to GN1 code for single chr mapping
self.selected_chr = -1
if "selected_chr" in start_vars:
self.selected_chr = int(start_vars['selected_chr'])
self.dataset.group.get_markers()
if self.mapping_method == "gemma":
self.score_type = "LOD"
included_markers, p_values = gemma_mapping.run_gemma(self.dataset, self.samples, self.vals)
self.dataset.group.get_specified_markers(markers = included_markers)
self.dataset.group.markers.add_pvalues(p_values)
results = self.dataset.group.markers.markers
elif self.mapping_method == "rqtl_plink":
results = self.run_rqtl_plink()
elif self.mapping_method == "rqtl_geno":
self.score_type = "LOD"
self.mapping_scale = "morgan"
if start_vars['num_perm'] == "":
self.num_perm = 0
else:
self.num_perm = start_vars['num_perm']
self.control = start_vars['control_marker']
self.do_control = start_vars['do_control']
self.method = start_vars['mapmethod_rqtl_geno']
self.model = start_vars['mapmodel_rqtl_geno']
if start_vars['pair_scan'] == "true":
self.pair_scan = True
results = self.run_rqtl_geno()
elif self.mapping_method == "reaper":
if start_vars['num_perm'] == "":
self.num_perm = 0
else:
self.num_perm = int(start_vars['num_perm'])
self.additive = False
self.control = start_vars['control_marker']
self.do_control = start_vars['do_control']
results = self.gen_reaper_results()
elif self.mapping_method == "plink":
results = self.run_plink()
elif self.mapping_method == "pylmm":
print("RUNNING PYLMM")
self.num_perm = start_vars['num_perm']
if self.num_perm != "":
if int(self.num_perm) > 0:
self.run_permutations(str(temp_uuid))
results = self.gen_data(str(temp_uuid))
else:
print("RUNNING NOTHING")
if self.pair_scan == True:
self.qtl_results = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker['chr1'] > 0 or marker['chr1'] == "X" or marker['chr1'] == "X/Y":
if marker['chr1'] > highest_chr or marker['chr1'] == "X" or marker['chr1'] == "X/Y":
highest_chr = marker['chr1']
if 'lod_score' in marker.keys():
self.qtl_results.append(marker)
for qtl in enumerate(self.qtl_results):
self.json_data['chr1'].append(str(qtl['chr1']))
self.json_data['chr2'].append(str(qtl['chr2']))
self.json_data['Mb'].append(qtl['Mb'])
self.json_data['markernames'].append(qtl['name'])
self.js_data = dict(
json_data = self.json_data,
this_trait = self.this_trait.name,
data_set = self.dataset.name,
maf = self.maf,
manhattan_plot = self.manhattan_plot,
mapping_scale = self.mapping_scale,
qtl_results = self.qtl_results,
)
else:
self.cutoff = 2
self.qtl_results = []
highest_chr = 1 #This is needed in order to convert the highest chr to X/Y
for marker in results:
if marker['chr'] > 0 or marker['chr'] == "X" or marker['chr'] == "X/Y":
if marker['chr'] > highest_chr or marker['chr'] == "X" or marker['chr'] == "X/Y":
highest_chr = marker['chr']
if ('lod_score' in marker.keys()) or ('lrs_value' in marker.keys()):
self.qtl_results.append(marker)
self.json_data['chr'] = []
self.json_data['pos'] = []
self.json_data['lod.hk'] = []
self.json_data['markernames'] = []
self.json_data['suggestive'] = self.suggestive
self.json_data['significant'] = self.significant
#Need to convert the QTL objects that qtl reaper returns into a json serializable dictionary
for index, qtl in enumerate(self.qtl_results):
#if index<40:
# print("lod score is:", qtl['lod_score'])
if qtl['chr'] == highest_chr and highest_chr != "X" and highest_chr != "X/Y":
print("changing to X")
self.json_data['chr'].append("X")
else:
self.json_data['chr'].append(str(qtl['chr']))
self.json_data['pos'].append(qtl['Mb'])
if 'lrs_value' in qtl.keys():
self.json_data['lod.hk'].append(str(qtl['lrs_value']))
else:
self.json_data['lod.hk'].append(str(qtl['lod_score']))
self.json_data['markernames'].append(qtl['name'])
#Get chromosome lengths for drawing the interval map plot
chromosome_mb_lengths = {}
self.json_data['chrnames'] = []
for key in self.species.chromosomes.chromosomes.keys():
self.json_data['chrnames'].append([self.species.chromosomes.chromosomes[key].name, self.species.chromosomes.chromosomes[key].mb_length])
chromosome_mb_lengths[key] = self.species.chromosomes.chromosomes[key].mb_length
# print("json_data:", self.json_data)
self.js_data = dict(
result_score_type = self.score_type,
json_data = self.json_data,
this_trait = self.this_trait.name,
data_set = self.dataset.name,
maf = self.maf,
manhattan_plot = self.manhattan_plot,
mapping_scale = self.mapping_scale,
chromosomes = chromosome_mb_lengths,
qtl_results = self.qtl_results,
)
