diff options
| -rwxr-xr-x | bin/genenetwork2 | 8 | ||||
| -rw-r--r-- | wqflask/base/webqtlConfig.py | 1 | ||||
| -rw-r--r-- | wqflask/utility/tools.py | 13 | ||||
| -rw-r--r-- | wqflask/wqflask/correlation/show_corr_results.py | 100 | ||||
| -rw-r--r-- | wqflask/wqflask/marker_regression/gemma_mapping.py | 16 |
5 files changed, 93 insertions, 45 deletions
diff --git a/bin/genenetwork2 b/bin/genenetwork2 index 145ce395..5438c1c0 100755 --- a/bin/genenetwork2 +++ b/bin/genenetwork2 @@ -45,6 +45,7 @@ SCRIPT=$(realpath "$0") echo SCRIPT=$SCRIPT +echo GN2_PROFILE=$GN2_PROFILE GN2_BASE_DIR=$(dirname $(dirname "$SCRIPT")) GN2_ID=$(cat /etc/hostname):$(basename $GN2_BASE_DIR) @@ -125,7 +126,9 @@ else export PLINK_COMMAND="$GN2_PROFILE/bin/plink2" export PYLMM_COMMAND="$GN2_PROFILE/bin/pylmm_redis" export GEMMA_COMMAND="$GN2_PROFILE/bin/gemma" - export GEMMA_WRAPPER_COMMAND="$GN2_PROFILE/bin/gemma-wrapper" + if [ -z $GEMMA_WRAPPER_COMMAND ]; then + export GEMMA_WRAPPER_COMMAND="$GN2_PROFILE/bin/gemma-wrapper" + fi while IFS=":" read -ra PPATH; do for PPart in "${PPATH[@]}"; do if [ ! -d $PPart ] ; then echo "PYTHONPATH not valid "$PYTHONPATH ; exit 1 ; fi @@ -198,7 +201,8 @@ if [ "$1" = '-gunicorn-prod' ] ; then cd $GN2_BASE_DIR/wqflask echo PYTHONPATH=$PYTHONPATH if [ -z $SERVER_PORT ]; then echo "ERROR: Provide a SERVER_PORT" ; exit 1 ; fi - cmd="--bind 0.0.0.0:$SERVER_PORT --workers=32 --max-requests 1000 --timeout 1200 wsgi" + PID=$TMPDIR/gunicorn.$USER.pid + cmd="--bind 0.0.0.0:$SERVER_PORT --pid $PID -k eventlet --workers 20 --keep-alive 1200 --max-requests 1000 --timeout 1200 wsgi" echo RUNNING gunicorn $cmd gunicorn $cmd exit $? diff --git a/wqflask/base/webqtlConfig.py b/wqflask/base/webqtlConfig.py index 1ef2bc26..1e66e957 100644 --- a/wqflask/base/webqtlConfig.py +++ b/wqflask/base/webqtlConfig.py @@ -82,6 +82,7 @@ assert_writable_dir(GENERATED_TEXT_DIR) # Flat file directories GENODIR = flat_files('genotype')+'/' assert_dir(GENODIR) +assert_dir(GENODIR+'bimbam') # for gemma # JSON genotypes are OBSOLETE JSON_GENODIR = flat_files('genotype/json')+'/' diff --git a/wqflask/utility/tools.py b/wqflask/utility/tools.py index ec673cf5..005f9b0f 100644 --- a/wqflask/utility/tools.py +++ b/wqflask/utility/tools.py @@ -16,7 +16,7 @@ OVERRIDES = {} def app_set(command_id, value): """Set application wide value""" app.config.setdefault(command_id,value) - value + return value def get_setting(command_id,guess=None): """Resolve a setting from the environment or the global settings in @@ -51,7 +51,7 @@ def get_setting(command_id,guess=None): return None # ---- Check whether environment exists - logger.debug("Looking for "+command_id+"\n") + # print("Looking for "+command_id+"\n") command = value(os.environ.get(command_id)) if command is None or command == "": command = OVERRIDES.get(command_id) @@ -63,7 +63,7 @@ def get_setting(command_id,guess=None): if command is None or command == "": # print command raise Exception(command_id+' setting unknown or faulty (update default_settings.py?).') - logger.debug("Set "+command_id+"="+str(command)) + # print("Set "+command_id+"="+str(command)) return command def get_setting_bool(id): @@ -280,6 +280,7 @@ SMTP_PASSWORD = get_setting_safe('SMTP_PASSWORD') PYLMM_COMMAND = app_set("PYLMM_COMMAND",pylmm_command()) GEMMA_COMMAND = app_set("GEMMA_COMMAND",gemma_command()) +assert(GEMMA_COMMAND is not None) PLINK_COMMAND = app_set("PLINK_COMMAND",plink_command()) GEMMA_WRAPPER_COMMAND = gemma_wrapper_command() TEMPDIR = tempdir() # defaults to UNIX TMPDIR @@ -293,7 +294,7 @@ from six import string_types if os.environ.get('WQFLASK_OVERRIDES'): jsonfn = get_setting('WQFLASK_OVERRIDES') - logger.error("WQFLASK_OVERRIDES: %s" % jsonfn) + logger.info("WQFLASK_OVERRIDES: %s" % jsonfn) with open(jsonfn) as data_file: overrides = json.load(data_file) for k in overrides: @@ -305,8 +306,4 @@ if os.environ.get('WQFLASK_OVERRIDES'): logger.debug(OVERRIDES) # assert_file(PHEWAS_FILES+"/auwerx/PheWAS_pval_EMMA_norm.RData") -# assert_dir(get_setting("JS_BIODALLIANCE")) -# assert_file(get_setting("JS_BIODALLIANCE")+"/build/dalliance-all.js") -# assert_file(get_setting("JS_BIODALLIANCE")+"/build/worker-all.js") -# assert_dir(get_setting("JS_TWITTER_POST_FETCHER")) assert_file(JS_TWITTER_POST_FETCHER_PATH+"/js/twitterFetcher_min.js") diff --git a/wqflask/wqflask/correlation/show_corr_results.py b/wqflask/wqflask/correlation/show_corr_results.py index 24432ad0..3d1c0d17 100644 --- a/wqflask/wqflask/correlation/show_corr_results.py +++ b/wqflask/wqflask/correlation/show_corr_results.py @@ -75,6 +75,46 @@ def print_mem(stage=""): mem = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss #print("{}: {}".format(stage, mem/1024)) +def is_float(value): + try: + float(value) + return True + except: + return False + +def is_int(value): + try: + int(value) + return True + except: + return False + +def is_str(value): + if value is None: + return False + try: + str(value) + return True + except: + return False + +def get_float(vars,name,default=None): + if name in vars: + if is_float(vars[name]): + return float(vars[name]) + return None + +def get_int(vars,name,default=None): + if name in vars: + if is_int(vars[name]): + return float(vars[name]) + return default + +def get_string(vars,name,default=None): + if name in vars: + if not vars[name] is None: + return str(vars[name]) + return default class AuthException(Exception): pass @@ -96,7 +136,19 @@ class CorrelationResults(object): # get trait list from db (database name) # calculate correlation with Base vector and targets - print("TESTING...") + # Check parameters + assert('corr_type' in start_vars) + assert(is_str(start_vars['corr_type'])) + assert('dataset' in start_vars) + # assert('group' in start_vars) permitted to be empty? + assert('corr_sample_method' in start_vars) + assert('corr_samples_group' in start_vars) + assert('corr_dataset' in start_vars) + assert('min_expr' in start_vars) + assert('corr_return_results' in start_vars) + if 'loc_chr' in start_vars: + assert('min_loc_mb' in start_vars) + assert('max_loc_mb' in start_vars) with Bench("Doing correlations"): if start_vars['dataset'] == "Temp": @@ -115,27 +167,17 @@ class CorrelationResults(object): self.sample_data = {} self.corr_type = start_vars['corr_type'] self.corr_method = start_vars['corr_sample_method'] - if 'min_expr' in start_vars: - if start_vars['min_expr'] != "": - self.min_expr = float(start_vars['min_expr']) - else: - self.min_expr = None - self.p_range_lower = float(start_vars['p_range_lower']) - self.p_range_upper = float(start_vars['p_range_upper']) + self.min_expr = get_float(start_vars,'min_expr') + self.p_range_lower = get_float(start_vars,'p_range_lower',-1.0) + self.p_range_upper = get_float(start_vars,'p_range_upper',1.0) if ('loc_chr' in start_vars and 'min_loc_mb' in start_vars and 'max_loc_mb' in start_vars): - self.location_chr = start_vars['loc_chr'] - if start_vars['min_loc_mb'].isdigit(): - self.min_location_mb = start_vars['min_loc_mb'] - else: - self.min_location_mb = None - if start_vars['max_loc_mb'].isdigit(): - self.max_location_mb = start_vars['max_loc_mb'] - else: - self.max_location_mb = None + self.location_chr = get_string(start_vars,'loc_chr') + self.min_location_mb = get_int(start_vars,'min_loc_mb') + self.max_location_mb = get_int(start_vars,'max_loc_mb') self.get_formatted_corr_type() self.return_number = int(start_vars['corr_return_results']) @@ -183,7 +225,7 @@ class CorrelationResults(object): else: for trait, values in self.target_dataset.trait_data.iteritems(): self.get_sample_r_and_p_values(trait, values) - + elif self.corr_type == "lit": self.trait_geneid_dict = self.dataset.retrieve_genes("GeneId") lit_corr_data = self.do_lit_correlation_for_all_traits() @@ -564,7 +606,7 @@ class CorrelationResults(object): self.this_trait_vals.append(sample_value) target_vals.append(target_sample_value) - self.this_trait_vals, target_vals, num_overlap = corr_result_helpers.normalize_values(self.this_trait_vals, target_vals) + self.this_trait_vals, target_vals, num_overlap = corr_result_helpers.normalize_values(self.this_trait_vals, target_vals) #ZS: 2015 could add biweight correlation, see http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3465711/ if self.corr_method == 'pearson': @@ -574,8 +616,8 @@ class CorrelationResults(object): if num_overlap > 5: self.correlation_data[trait] = [sample_r, sample_p, num_overlap] - - + + """ correlations = [] @@ -673,8 +715,8 @@ class CorrelationResults(object): method=self.method) return trait_list - """ - + """ + def do_tissue_corr_for_all_traits_2(self): """Comments Possibly Out of Date!!!!! @@ -1089,7 +1131,7 @@ class CorrelationResults(object): totalTraits = len(traits) #XZ, 09/18/2008: total trait number return traits - + def calculate_corr_for_all_tissues(self, tissue_dataset_id=None): symbol_corr_dict = {} @@ -1129,7 +1171,7 @@ class CorrelationResults(object): values_2.append(target_value) correlation = calCorrelation(values_1, values_2) self.correlation_data[trait] = correlation - + def getFileName(self, target_db_name): ### dcrowell August 2008 """Returns the name of the reference database file with which correlations are calculated. Takes argument cursor which is a cursor object of any instance of a subclass of templatePage @@ -1144,7 +1186,7 @@ class CorrelationResults(object): return FileName def do_parallel_correlation(self, db_filename, num_overlap): - + #XZ, 01/14/2009: This method is for parallel computing only. #XZ: It is supposed to be called when "Genetic Correlation, Pearson's r" (method 1) #XZ: or "Genetic Correlation, Spearman's rho" (method 2) is selected @@ -1313,7 +1355,7 @@ class CorrelationResults(object): z_value = z_value*math.sqrt(nOverlap-3) sample_p = 2.0*(1.0 - reaper.normp(abs(z_value))) - correlation_data[traitdataName] = [sample_r, sample_p, nOverlap] + correlation_data[traitdataName] = [sample_r, sample_p, nOverlap] # traitinfo = [traitdataName, sample_r, nOverlap] # allcorrelations.append(traitinfo) @@ -1321,7 +1363,7 @@ class CorrelationResults(object): return correlation_data # return allcorrelations - + datasetFile = open(webqtlConfig.GENERATED_TEXT_DIR+db_filename,'r') print("Invoking parallel computing") @@ -1378,5 +1420,3 @@ class CorrelationResults(object): # for one_result in results: # for one_traitinfo in one_result: # allcorrelations.append( one_traitinfo ) - - diff --git a/wqflask/wqflask/marker_regression/gemma_mapping.py b/wqflask/wqflask/marker_regression/gemma_mapping.py index a24e43d4..68920130 100644 --- a/wqflask/wqflask/marker_regression/gemma_mapping.py +++ b/wqflask/wqflask/marker_regression/gemma_mapping.py @@ -3,7 +3,7 @@ import os, math, string, random, json from base import webqtlConfig from base.trait import GeneralTrait from base.data_set import create_dataset -from utility.tools import flat_files, GEMMA_COMMAND, GEMMA_WRAPPER_COMMAND, TEMPDIR +from utility.tools import flat_files, GEMMA_COMMAND, GEMMA_WRAPPER_COMMAND, TEMPDIR, assert_bin, assert_file import utility.logger logger = utility.logger.getLogger(__name__ ) @@ -11,6 +11,7 @@ logger = utility.logger.getLogger(__name__ ) def run_gemma(this_dataset, samples, vals, covariates, method, use_loco): """Generates p-values for each marker using GEMMA""" + assert_bin(GEMMA_COMMAND); if this_dataset.group.genofile != None: genofile_name = this_dataset.group.genofile[:-5] else: @@ -27,7 +28,7 @@ def run_gemma(this_dataset, samples, vals, covariates, method, use_loco): if i < (len(this_chromosomes) - 1): chr_list_string += this_chromosomes[i+1].name + "," else: - chr_list_string += this_chromosomes[i+1].name + chr_list_string += this_chromosomes[i+1].name if covariates != "": gen_covariates_file(this_dataset, covariates) @@ -209,8 +210,13 @@ def parse_gemma_output(genofile_name): def parse_loco_output(this_dataset, gwa_output_filename): output_filelist = [] - with open("{}/gn2/".format(TEMPDIR) + gwa_output_filename + ".json") as data_file: - data = json.load(data_file) + jsonfn = "{}/gn2/".format(TEMPDIR) + gwa_output_filename + ".json" + assert_file(jsonfn) + try: + with open(jsonfn) as data_file: + data = json.load(data_file) + except: + logger.error("Can not parse "+jsonfn) files = data['files'] for file in files: @@ -247,4 +253,4 @@ def parse_loco_output(this_dataset, gwa_output_filename): included_markers.append(line.split("\t")[1]) p_values.append(float(line.split("\t")[10])) - return marker_obs
\ No newline at end of file + return marker_obs |