diff options
Diffstat (limited to 'gn3/base')
| -rw-r--r-- | gn3/base/__init__.py | 0 | ||||
| -rw-r--r-- | gn3/base/data_set.py | 886 | ||||
| -rw-r--r-- | gn3/base/mrna_assay_tissue_data.py | 94 | ||||
| -rw-r--r-- | gn3/base/species.py | 64 | ||||
| -rw-r--r-- | gn3/base/trait.py | 366 | ||||
| -rw-r--r-- | gn3/base/webqtlCaseData.py | 84 |
6 files changed, 1494 insertions, 0 deletions
diff --git a/gn3/base/__init__.py b/gn3/base/__init__.py new file mode 100644 index 0000000..e69de29 --- /dev/null +++ b/gn3/base/__init__.py diff --git a/gn3/base/data_set.py b/gn3/base/data_set.py new file mode 100644 index 0000000..e61e4eb --- /dev/null +++ b/gn3/base/data_set.py @@ -0,0 +1,886 @@ + +import json +import math +import collections +import requests +from redis import Redis +from flask import g +from gn3.utility.db_tools import escape +from gn3.utility.db_tools import mescape +from gn3.utility.db_tools import create_in_clause +from gn3.utility.tools import locate_ignore_error +from gn3.db.calls import fetch1 +from gn3.db.calls import fetchone +from gn3.db.webqtlDatabaseFunction import retrieve_species +from gn3.utility import chunks + +from gn3.utility import get_group_samplelists +from gn3.base.species import TheSpecies +r = Redis() + +# should probably move this to its own configuration files + +USE_REDIS = True + +# todo move to config file +GN2_BASE_URL = "https://genenetwork.org/" + +DS_NAME_MAP = {} + +# pylint: disable-all +#todo file not linted +# pylint: disable=C0103 + + + +def create_dataset(dataset_name, dataset_type=None, get_samplelist=True, group_name=None): + + if dataset_name == "Temp": + dataset_type = "Temp" + + if dataset_type is None: + dataset_type = Dataset_Getter(dataset_name) + dataset_ob = DS_NAME_MAP[dataset_type] + dataset_class = globals()[dataset_ob] + + if dataset_type == "Temp": + results = dataset_class(dataset_name, get_samplelist, group_name) + + else: + results = dataset_class(dataset_name, get_samplelist) + + return results + + +class DatasetType: + def __init__(self, redis_instance): + self.redis_instance = redis_instance + self.datasets = {} + + data = self.redis_instance.get("dataset_structure") + if data: + self.datasets = json.loads(data) + + else: + + try: + + data = json.loads(requests.get( + GN2_BASE_URL + "/api/v_pre1/gen_dropdown", timeout=5).content) + + # todo:Refactor code below n^4 loop + + for species in data["datasets"]: + for group in data["datasets"][species]: + for dataset_type in data['datasets'][species][group]: + for dataset in data['datasets'][species][group][dataset_type]: + + short_dataset_name = dataset[1] + if dataset_type == "Phenotypes": + new_type = "Publish" + + elif dataset_type == "Genotypes": + new_type = "Geno" + else: + new_type = "ProbeSet" + + self.datasets[short_dataset_name] = new_type + + except Exception as e: + raise e + + self.redis_instance.set( + "dataset_structure", json.dumps(self.datasets)) + + def set_dataset_key(self, t, name): + """If name is not in the object's dataset dictionary, set it, and update + dataset_structure in Redis + + args: + t: Type of dataset structure which can be: 'mrna_expr', 'pheno', + 'other_pheno', 'geno' + name: The name of the key to inserted in the datasets dictionary + + """ + + sql_query_mapping = { + 'mrna_expr': ("""SELECT ProbeSetFreeze.Id FROM """ + + """ProbeSetFreeze WHERE ProbeSetFreeze.Name = "{}" """), + 'pheno': ("""SELECT InfoFiles.GN_AccesionId """ + + """FROM InfoFiles, PublishFreeze, InbredSet """ + + """WHERE InbredSet.Name = '{}' AND """ + + """PublishFreeze.InbredSetId = InbredSet.Id AND """ + + """InfoFiles.InfoPageName = PublishFreeze.Name"""), + 'other_pheno': ("""SELECT PublishFreeze.Name """ + + """FROM PublishFreeze, InbredSet """ + + """WHERE InbredSet.Name = '{}' AND """ + + """PublishFreeze.InbredSetId = InbredSet.Id"""), + 'geno': ("""SELECT GenoFreeze.Id FROM GenoFreeze WHERE """ + + """GenoFreeze.Name = "{}" """) + } + + dataset_name_mapping = { + "mrna_expr": "ProbeSet", + "pheno": "Publish", + "other_pheno": "Publish", + "geno": "Geno", + } + + group_name = name + if t in ['pheno', 'other_pheno']: + group_name = name.replace("Publish", "") + + results = g.db.execute( + sql_query_mapping[t].format(group_name)).fetchone() + if results: + self.datasets[name] = dataset_name_mapping[t] + self.redis_instance.set( + "dataset_structure", json.dumps(self.datasets)) + + return True + + return None + + def __call__(self, name): + if name not in self.datasets: + for t in ["mrna_expr", "pheno", "other_pheno", "geno"]: + + if(self.set_dataset_key(t, name)): + # This has side-effects, with the end result being a truth-y value + break + + return self.datasets.get(name, None) + + +# Do the intensive work at startup one time only +# could replace the code below +Dataset_Getter = DatasetType(r) + + +class DatasetGroup: + """ + Each group has multiple datasets; each species has multiple groups. + + For example, Mouse has multiple groups (BXD, BXA, etc), and each group + has multiple datasets associated with it. + + """ + + def __init__(self, dataset, name=None): + """This sets self.group and self.group_id""" + if name == None: + self.name, self.id, self.genetic_type = fetchone( + dataset.query_for_group) + + else: + self.name, self.id, self.genetic_type = fetchone( + "SELECT InbredSet.Name, InbredSet.Id, InbredSet.GeneticType FROM InbredSet where Name='%s'" % name) + + if self.name == 'BXD300': + self.name = "BXD" + + self.f1list = None + + self.parlist = None + + self.get_f1_parent_strains() + + # remove below not used in correlation + + self.mapping_id, self.mapping_names = self.get_mapping_methods() + + self.species = retrieve_species(self.name) + + def get_f1_parent_strains(self): + try: + # should import ParInfo + raise e + # NL, 07/27/2010. ParInfo has been moved from webqtlForm.py to webqtlUtil.py; + f1, f12, maternal, paternal = webqtlUtil.ParInfo[self.name] + except Exception as e: + f1 = f12 = maternal = paternal = None + + if f1 and f12: + self.f1list = [f1, f12] + + if maternal and paternal: + self.parlist = [maternal, paternal] + + def get_mapping_methods(self): + mapping_id = g.db.execute( + "select MappingMethodId from InbredSet where Name= '%s'" % self.name).fetchone()[0] + + if mapping_id == "1": + mapping_names = ["GEMMA", "QTLReaper", "R/qtl"] + elif mapping_id == "2": + mapping_names = ["GEMMA"] + + elif mapping_id == "3": + mapping_names = ["R/qtl"] + + elif mapping_id == "4": + mapping_names = ["GEMMA", "PLINK"] + + else: + mapping_names = [] + + return mapping_id, mapping_names + + def get_samplelist(self): + result = None + key = "samplelist:v3:" + self.name + if USE_REDIS: + result = r.get(key) + + if result is not None: + + self.samplelist = json.loads(result) + + else: + # logger.debug("Cache not hit") + # should enable logger + genotype_fn = locate_ignore_error(self.name+".geno", 'genotype') + if genotype_fn: + self.samplelist = get_group_samplelists.get_samplelist( + "geno", genotype_fn) + + else: + self.samplelist = None + + if USE_REDIS: + r.set(key, json.dumps(self.samplelist)) + r.expire(key, 60*5) + + +class DataSet: + """ + DataSet class defines a dataset in webqtl, can be either Microarray, + Published phenotype, genotype, or user input dataset(temp) + + """ + + def __init__(self, name, get_samplelist=True, group_name=None): + + assert name, "Need a name" + self.name = name + self.id = None + self.shortname = None + self.fullname = None + self.type = None + self.data_scale = None # ZS: For example log2 + + self.setup() + + if self.type == "Temp": # Need to supply group name as input if temp trait + # sets self.group and self.group_id and gets genotype + self.group = DatasetGroup(self, name=group_name) + else: + self.check_confidentiality() + self.retrieve_other_names() + # sets self.group and self.group_id and gets genotype + self.group = DatasetGroup(self) + self.accession_id = self.get_accession_id() + if get_samplelist == True: + self.group.get_samplelist() + self.species = TheSpecies(self) + + def get_desc(self): + """Gets overridden later, at least for Temp...used by trait's get_given_name""" + return None + + # Delete this eventually + @property + def riset(): + Weve_Renamed_This_As_Group + + def get_accession_id(self): + if self.type == "Publish": + results = g.db.execute("""select InfoFiles.GN_AccesionId from InfoFiles, PublishFreeze, InbredSet where + InbredSet.Name = %s and + PublishFreeze.InbredSetId = InbredSet.Id and + InfoFiles.InfoPageName = PublishFreeze.Name and + PublishFreeze.public > 0 and + PublishFreeze.confidentiality < 1 order by + PublishFreeze.CreateTime desc""", (self.group.name)).fetchone() + elif self.type == "Geno": + results = g.db.execute("""select InfoFiles.GN_AccesionId from InfoFiles, GenoFreeze, InbredSet where + InbredSet.Name = %s and + GenoFreeze.InbredSetId = InbredSet.Id and + InfoFiles.InfoPageName = GenoFreeze.ShortName and + GenoFreeze.public > 0 and + GenoFreeze.confidentiality < 1 order by + GenoFreeze.CreateTime desc""", (self.group.name)).fetchone() + else: + results = None + + if results != None: + return str(results[0]) + else: + return "None" + + def retrieve_other_names(self): + """This method fetches the the dataset names in search_result. + + If the data set name parameter is not found in the 'Name' field of + the data set table, check if it is actually the FullName or + ShortName instead. + + This is not meant to retrieve the data set info if no name at + all is passed. + + """ + + try: + if self.type == "ProbeSet": + query_args = tuple(escape(x) for x in ( + self.name, + self.name, + self.name)) + + self.id, self.name, self.fullname, self.shortname, self.data_scale, self.tissue = fetch1(""" + SELECT ProbeSetFreeze.Id, ProbeSetFreeze.Name, ProbeSetFreeze.FullName, ProbeSetFreeze.ShortName, ProbeSetFreeze.DataScale, Tissue.Name + FROM ProbeSetFreeze, ProbeFreeze, Tissue + WHERE ProbeSetFreeze.ProbeFreezeId = ProbeFreeze.Id + AND ProbeFreeze.TissueId = Tissue.Id + AND (ProbeSetFreeze.Name = '%s' OR ProbeSetFreeze.FullName = '%s' OR ProbeSetFreeze.ShortName = '%s') + """ % (query_args), "/dataset/"+self.name+".json", + lambda r: (r["id"], r["name"], r["full_name"], + r["short_name"], r["data_scale"], r["tissue"]) + ) + else: + query_args = tuple(escape(x) for x in ( + (self.type + "Freeze"), + self.name, + self.name, + self.name)) + + self.tissue = "N/A" + self.id, self.name, self.fullname, self.shortname = fetchone(""" + SELECT Id, Name, FullName, ShortName + FROM %s + WHERE (Name = '%s' OR FullName = '%s' OR ShortName = '%s') + """ % (query_args)) + + except TypeError as e: + logger.debug( + "Dataset {} is not yet available in GeneNetwork.".format(self.name)) + pass + + def get_trait_data(self, sample_list=None): + if sample_list: + self.samplelist = sample_list + else: + self.samplelist = self.group.samplelist + + if self.group.parlist != None and self.group.f1list != None: + if (self.group.parlist + self.group.f1list) in self.samplelist: + self.samplelist += self.group.parlist + self.group.f1list + + query = """ + SELECT Strain.Name, Strain.Id FROM Strain, Species + WHERE Strain.Name IN {} + and Strain.SpeciesId=Species.Id + and Species.name = '{}' + """.format(create_in_clause(self.samplelist), *mescape(self.group.species)) + # logger.sql(query) + results = dict(g.db.execute(query).fetchall()) + sample_ids = [results[item] for item in self.samplelist] + + # MySQL limits the number of tables that can be used in a join to 61, + # so we break the sample ids into smaller chunks + # Postgres doesn't have that limit, so we can get rid of this after we transition + chunk_size = 50 + number_chunks = int(math.ceil(len(sample_ids) / chunk_size)) + trait_sample_data = [] + for sample_ids_step in chunks.divide_into_chunks(sample_ids, number_chunks): + if self.type == "Publish": + dataset_type = "Phenotype" + else: + dataset_type = self.type + temp = ['T%s.value' % item for item in sample_ids_step] + if self.type == "Publish": + query = "SELECT {}XRef.Id,".format(escape(self.type)) + else: + query = "SELECT {}.Name,".format(escape(dataset_type)) + data_start_pos = 1 + query += ', '.join(temp) + query += ' FROM ({}, {}XRef, {}Freeze) '.format(*mescape(dataset_type, + self.type, + self.type)) + + for item in sample_ids_step: + query += """ + left join {}Data as T{} on T{}.Id = {}XRef.DataId + and T{}.StrainId={}\n + """.format(*mescape(self.type, item, item, self.type, item, item)) + + if self.type == "Publish": + query += """ + WHERE {}XRef.InbredSetId = {}Freeze.InbredSetId + and {}Freeze.Name = '{}' + and {}.Id = {}XRef.{}Id + order by {}.Id + """.format(*mescape(self.type, self.type, self.type, self.name, + dataset_type, self.type, dataset_type, dataset_type)) + else: + query += """ + WHERE {}XRef.{}FreezeId = {}Freeze.Id + and {}Freeze.Name = '{}' + and {}.Id = {}XRef.{}Id + order by {}.Id + """.format(*mescape(self.type, self.type, self.type, self.type, + self.name, dataset_type, self.type, self.type, dataset_type)) + + results = g.db.execute(query).fetchall() + trait_sample_data.append(results) + + trait_count = len(trait_sample_data[0]) + self.trait_data = collections.defaultdict(list) + + # put all of the separate data together into a dictionary where the keys are + # trait names and values are lists of sample values + for trait_counter in range(trait_count): + trait_name = trait_sample_data[0][trait_counter][0] + for chunk_counter in range(int(number_chunks)): + self.trait_data[trait_name] += ( + trait_sample_data[chunk_counter][trait_counter][data_start_pos:]) + + +class MrnaAssayDataSet(DataSet): + ''' + An mRNA Assay is a quantitative assessment (assay) associated with an mRNA trait + + This used to be called ProbeSet, but that term only refers specifically to the Affymetrix + platform and is far too specific. + + ''' + DS_NAME_MAP['ProbeSet'] = 'MrnaAssayDataSet' + + def setup(self): + # Fields in the database table + self.search_fields = ['Name', + 'Description', + 'Probe_Target_Description', + 'Symbol', + 'Alias', + 'GenbankId', + 'UniGeneId', + 'RefSeq_TranscriptId'] + + # Find out what display_fields is + self.display_fields = ['name', 'symbol', + 'description', 'probe_target_description', + 'chr', 'mb', + 'alias', 'geneid', + 'genbankid', 'unigeneid', + 'omim', 'refseq_transcriptid', + 'blatseq', 'targetseq', + 'chipid', 'comments', + 'strand_probe', 'strand_gene', + 'proteinid', 'uniprotid', + 'probe_set_target_region', + 'probe_set_specificity', + 'probe_set_blat_score', + 'probe_set_blat_mb_start', + 'probe_set_blat_mb_end', + 'probe_set_strand', + 'probe_set_note_by_rw', + 'flag'] + + # Fields displayed in the search results table header + self.header_fields = ['Index', + 'Record', + 'Symbol', + 'Description', + 'Location', + 'Mean', + 'Max LRS', + 'Max LRS Location', + 'Additive Effect'] + + # Todo: Obsolete or rename this field + self.type = 'ProbeSet' + + self.query_for_group = ''' + SELECT + InbredSet.Name, InbredSet.Id, InbredSet.GeneticType + FROM + InbredSet, ProbeSetFreeze, ProbeFreeze + WHERE + ProbeFreeze.InbredSetId = InbredSet.Id AND + ProbeFreeze.Id = ProbeSetFreeze.ProbeFreezeId AND + ProbeSetFreeze.Name = "%s" + ''' % escape(self.name) + + def check_confidentiality(self): + return geno_mrna_confidentiality(self) + + def get_trait_info(self, trait_list=None, species=''): + + # Note: setting trait_list to [] is probably not a great idea. + if not trait_list: + trait_list = [] + + for this_trait in trait_list: + + if not this_trait.haveinfo: + this_trait.retrieveInfo(QTL=1) + + if not this_trait.symbol: + this_trait.symbol = "N/A" + + # XZ, 12/08/2008: description + # XZ, 06/05/2009: Rob asked to add probe target description + description_string = str( + str(this_trait.description).strip(codecs.BOM_UTF8), 'utf-8') + target_string = str( + str(this_trait.probe_target_description).strip(codecs.BOM_UTF8), 'utf-8') + + if len(description_string) > 1 and description_string != 'None': + description_display = description_string + else: + description_display = this_trait.symbol + + if (len(description_display) > 1 and description_display != 'N/A' and + len(target_string) > 1 and target_string != 'None'): + description_display = description_display + '; ' + target_string.strip() + + # Save it for the jinja2 template + this_trait.description_display = description_display + + if this_trait.chr and this_trait.mb: + this_trait.location_repr = 'Chr%s: %.6f' % ( + this_trait.chr, float(this_trait.mb)) + + # Get mean expression value + query = ( + """select ProbeSetXRef.mean from ProbeSetXRef, ProbeSet + where ProbeSetXRef.ProbeSetFreezeId = %s and + ProbeSet.Id = ProbeSetXRef.ProbeSetId and + ProbeSet.Name = '%s' + """ % (escape(str(this_trait.dataset.id)), + escape(this_trait.name))) + + #logger.debug("query is:", pf(query)) + logger.sql(query) + result = g.db.execute(query).fetchone() + + mean = result[0] if result else 0 + + if mean: + this_trait.mean = "%2.3f" % mean + + # LRS and its location + this_trait.LRS_score_repr = 'N/A' + this_trait.LRS_location_repr = 'N/A' + + # Max LRS and its Locus location + if this_trait.lrs and this_trait.locus: + query = """ + select Geno.Chr, Geno.Mb from Geno, Species + where Species.Name = '{}' and + Geno.Name = '{}' and + Geno.SpeciesId = Species.Id + """.format(species, this_trait.locus) + logger.sql(query) + result = g.db.execute(query).fetchone() + + if result: + lrs_chr, lrs_mb = result + this_trait.LRS_score_repr = '%3.1f' % this_trait.lrs + this_trait.LRS_location_repr = 'Chr%s: %.6f' % ( + lrs_chr, float(lrs_mb)) + + return trait_list + + def retrieve_sample_data(self, trait): + query = """ + SELECT + Strain.Name, ProbeSetData.value, ProbeSetSE.error, NStrain.count, Strain.Name2 + FROM + (ProbeSetData, ProbeSetFreeze, Strain, ProbeSet, ProbeSetXRef) + left join ProbeSetSE on + (ProbeSetSE.DataId = ProbeSetData.Id AND ProbeSetSE.StrainId = ProbeSetData.StrainId) + left join NStrain on + (NStrain.DataId = ProbeSetData.Id AND + NStrain.StrainId = ProbeSetData.StrainId) + WHERE + ProbeSet.Name = '%s' AND ProbeSetXRef.ProbeSetId = ProbeSet.Id AND + ProbeSetXRef.ProbeSetFreezeId = ProbeSetFreeze.Id AND + ProbeSetFreeze.Name = '%s' AND + ProbeSetXRef.DataId = ProbeSetData.Id AND + ProbeSetData.StrainId = Strain.Id + Order BY + Strain.Name + """ % (escape(trait), escape(self.name)) + # logger.sql(query) + results = g.db.execute(query).fetchall() + #logger.debug("RETRIEVED RESULTS HERE:", results) + return results + + def retrieve_genes(self, column_name): + query = """ + select ProbeSet.Name, ProbeSet.%s + from ProbeSet,ProbeSetXRef + where ProbeSetXRef.ProbeSetFreezeId = %s and + ProbeSetXRef.ProbeSetId=ProbeSet.Id; + """ % (column_name, escape(str(self.id))) + # logger.sql(query) + results = g.db.execute(query).fetchall() + + return dict(results) + + +class TempDataSet(DataSet): + '''Temporary user-generated data set''' + + DS_NAME_MAP['Temp'] = 'TempDataSet' + + def setup(self): + self.search_fields = ['name', + 'description'] + + self.display_fields = ['name', + 'description'] + + self.header_fields = ['Name', + 'Description'] + + self.type = 'Temp' + + # Need to double check later how these are used + self.id = 1 + self.fullname = 'Temporary Storage' + self.shortname = 'Temp' + + +class PhenotypeDataSet(DataSet): + DS_NAME_MAP['Publish'] = 'PhenotypeDataSet' + + def setup(self): + + #logger.debug("IS A PHENOTYPEDATASET") + + # Fields in the database table + self.search_fields = ['Phenotype.Post_publication_description', + 'Phenotype.Pre_publication_description', + 'Phenotype.Pre_publication_abbreviation', + 'Phenotype.Post_publication_abbreviation', + 'PublishXRef.mean', + 'Phenotype.Lab_code', + 'Publication.PubMed_ID', + 'Publication.Abstract', + 'Publication.Title', + 'Publication.Authors', + 'PublishXRef.Id'] + + # Figure out what display_fields is + self.display_fields = ['name', 'group_code', + 'pubmed_id', + 'pre_publication_description', + 'post_publication_description', + 'original_description', + 'pre_publication_abbreviation', + 'post_publication_abbreviation', + 'mean', + 'lab_code', + 'submitter', 'owner', + 'authorized_users', + 'authors', 'title', + 'abstract', 'journal', + 'volume', 'pages', + 'month', 'year', + 'sequence', 'units', 'comments'] + + # Fields displayed in the search results table header + self.header_fields = ['Index', + 'Record', + 'Description', + 'Authors', + 'Year', + 'Max LRS', + 'Max LRS Location', + 'Additive Effect'] + + self.type = 'Publish' + + self.query_for_group = ''' + SELECT + InbredSet.Name, InbredSet.Id, InbredSet.GeneticType + FROM + InbredSet, PublishFreeze + WHERE + PublishFreeze.InbredSetId = InbredSet.Id AND + PublishFreeze.Name = "%s" + ''' % escape(self.name) + + def check_confidentiality(self): + # (Urgently?) Need to write this + pass + + def get_trait_info(self, trait_list, species=''): + for this_trait in trait_list: + + if not this_trait.haveinfo: + this_trait.retrieve_info(get_qtl_info=True) + + description = this_trait.post_publication_description + + # If the dataset is confidential and the user has access to confidential + # phenotype traits, then display the pre-publication description instead + # of the post-publication description + if this_trait.confidential: + this_trait.description_display = "" + continue # todo for now, because no authorization features + + if not webqtlUtil.has_access_to_confidentail_phenotype_trait( + privilege=self.privilege, + userName=self.userName, + authorized_users=this_trait.authorized_users): + + description = this_trait.pre_publication_description + + if len(description) > 0: + this_trait.description_display = description.strip() + else: + this_trait.description_display = "" + + if not this_trait.year.isdigit(): + this_trait.pubmed_text = "N/A" + else: + this_trait.pubmed_text = this_trait.year + + if this_trait.pubmed_id: + this_trait.pubmed_link = webqtlConfig.PUBMEDLINK_URL % this_trait.pubmed_id + + # LRS and its location + this_trait.LRS_score_repr = "N/A" + this_trait.LRS_location_repr = "N/A" + + if this_trait.lrs: + query = """ + select Geno.Chr, Geno.Mb from Geno, Species + where Species.Name = '%s' and + Geno.Name = '%s' and + Geno.SpeciesId = Species.Id + """ % (species, this_trait.locus) + + result = g.db.execute(query).fetchone() + + if result: + if result[0] and result[1]: + LRS_Chr = result[0] + LRS_Mb = result[1] + + this_trait.LRS_score_repr = LRS_score_repr = '%3.1f' % this_trait.lrs + this_trait.LRS_location_repr = LRS_location_repr = 'Chr%s: %.6f' % ( + LRS_Chr, float(LRS_Mb)) + + def retrieve_sample_data(self, trait): + query = """ + SELECT + Strain.Name, PublishData.value, PublishSE.error, NStrain.count, Strain.Name2 + FROM + (PublishData, Strain, PublishXRef, PublishFreeze) + left join PublishSE on + (PublishSE.DataId = PublishData.Id AND PublishSE.StrainId = PublishData.StrainId) + left join NStrain on + (NStrain.DataId = PublishData.Id AND + NStrain.StrainId = PublishData.StrainId) + WHERE + PublishXRef.InbredSetId = PublishFreeze.InbredSetId AND + PublishData.Id = PublishXRef.DataId AND PublishXRef.Id = %s AND + PublishFreeze.Id = %s AND PublishData.StrainId = Strain.Id + Order BY + Strain.Name + """ + + results = g.db.execute(query, (trait, self.id)).fetchall() + return results + + +class GenotypeDataSet(DataSet): + DS_NAME_MAP['Geno'] = 'GenotypeDataSet' + + def setup(self): + # Fields in the database table + self.search_fields = ['Name', + 'Chr'] + + # Find out what display_fields is + self.display_fields = ['name', + 'chr', + 'mb', + 'source2', + 'sequence'] + + # Fields displayed in the search results table header + self.header_fields = ['Index', + 'ID', + 'Location'] + + # Todo: Obsolete or rename this field + self.type = 'Geno' + + self.query_for_group = ''' + SELECT + InbredSet.Name, InbredSet.Id, InbredSet.GeneticType + FROM + InbredSet, GenoFreeze + WHERE + GenoFreeze.InbredSetId = InbredSet.Id AND + GenoFreeze.Name = "%s" + ''' % escape(self.name) + + def check_confidentiality(self): + return geno_mrna_confidentiality(self) + + def get_trait_info(self, trait_list, species=None): + for this_trait in trait_list: + if not this_trait.haveinfo: + this_trait.retrieveInfo() + + if this_trait.chr and this_trait.mb: + this_trait.location_repr = 'Chr%s: %.6f' % ( + this_trait.chr, float(this_trait.mb)) + + def retrieve_sample_data(self, trait): + query = """ + SELECT + Strain.Name, GenoData.value, GenoSE.error, "N/A", Strain.Name2 + FROM + (GenoData, GenoFreeze, Strain, Geno, GenoXRef) + left join GenoSE on + (GenoSE.DataId = GenoData.Id AND GenoSE.StrainId = GenoData.StrainId) + WHERE + Geno.SpeciesId = %s AND Geno.Name = %s AND GenoXRef.GenoId = Geno.Id AND + GenoXRef.GenoFreezeId = GenoFreeze.Id AND + GenoFreeze.Name = %s AND + GenoXRef.DataId = GenoData.Id AND + GenoData.StrainId = Strain.Id + Order BY + Strain.Name + """ + results = g.db.execute(query, + (webqtlDatabaseFunction.retrieve_species_id(self.group.name), + trait, self.name)).fetchall() + return results + + +def geno_mrna_confidentiality(ob): + dataset_table = ob.type + "Freeze" + #logger.debug("dataset_table [%s]: %s" % (type(dataset_table), dataset_table)) + + query = '''SELECT Id, Name, FullName, confidentiality, + AuthorisedUsers FROM %s WHERE Name = "%s"''' % (dataset_table, ob.name) + # + result = g.db.execute(query) + + (_dataset_id, + _name, + _full_name, + confidential, + _authorized_users) = result.fetchall()[0] + + if confidential: + return True diff --git a/gn3/base/mrna_assay_tissue_data.py b/gn3/base/mrna_assay_tissue_data.py new file mode 100644 index 0000000..0f51ade --- /dev/null +++ b/gn3/base/mrna_assay_tissue_data.py @@ -0,0 +1,94 @@ + +# pylint: disable-all +import collections + +from flask import g + +from gn3.utility.db_tools import create_in_clause +from gn3.utility.db_tools import escape +from gn3.utility.bunch import Bunch + + +# from utility.logger import getLogger +# logger = getLogger(__name__ ) + +class MrnaAssayTissueData(object): + + def __init__(self, gene_symbols=None): + self.gene_symbols = gene_symbols + if self.gene_symbols == None: + self.gene_symbols = [] + + self.data = collections.defaultdict(Bunch) + + query = '''select t.Symbol, t.GeneId, t.DataId, t.Chr, t.Mb, t.description, t.Probe_Target_Description + from ( + select Symbol, max(Mean) as maxmean + from TissueProbeSetXRef + where TissueProbeSetFreezeId=1 and ''' + + # Note that inner join is necessary in this query to get distinct record in one symbol group + # with highest mean value + # Due to the limit size of TissueProbeSetFreezeId table in DB, + # performance of inner join is acceptable.MrnaAssayTissueData(gene_symbols=symbol_list) + if len(gene_symbols) == 0: + query += '''Symbol!='' and Symbol Is Not Null group by Symbol) + as x inner join TissueProbeSetXRef as t on t.Symbol = x.Symbol + and t.Mean = x.maxmean; + ''' + else: + in_clause = create_in_clause(gene_symbols) + + # ZS: This was in the query, not sure why: http://docs.python.org/2/library/string.html?highlight=lower#string.lower + query += ''' Symbol in {} group by Symbol) + as x inner join TissueProbeSetXRef as t on t.Symbol = x.Symbol + and t.Mean = x.maxmean; + '''.format(in_clause) + + results = g.db.execute(query).fetchall() + + lower_symbols = [] + for gene_symbol in gene_symbols: + if gene_symbol != None: + lower_symbols.append(gene_symbol.lower()) + + for result in results: + symbol = result[0] + if symbol.lower() in lower_symbols: + symbol = symbol.lower() + + self.data[symbol].gene_id = result.GeneId + self.data[symbol].data_id = result.DataId + self.data[symbol].chr = result.Chr + self.data[symbol].mb = result.Mb + self.data[symbol].description = result.description + self.data[symbol].probe_target_description = result.Probe_Target_Description + + ########################################################################### + # Input: cursor, symbolList (list), dataIdDict(Dict) + # output: symbolValuepairDict (dictionary):one dictionary of Symbol and Value Pair, + # key is symbol, value is one list of expression values of one probeSet; + # function: get one dictionary whose key is gene symbol and value is tissue expression data (list type). + # Attention! All keys are lower case! + ########################################################################### + + def get_symbol_values_pairs(self): + id_list = [self.data[symbol].data_id for symbol in self.data] + + symbol_values_dict = {} + + if len(id_list) > 0: + query = """SELECT TissueProbeSetXRef.Symbol, TissueProbeSetData.value + FROM TissueProbeSetXRef, TissueProbeSetData + WHERE TissueProbeSetData.Id IN {} and + TissueProbeSetXRef.DataId = TissueProbeSetData.Id""".format(create_in_clause(id_list)) + + results = g.db.execute(query).fetchall() + for result in results: + if result.Symbol.lower() not in symbol_values_dict: + symbol_values_dict[result.Symbol.lower()] = [result.value] + else: + symbol_values_dict[result.Symbol.lower()].append( + result.value) + + return symbol_values_dict diff --git a/gn3/base/species.py b/gn3/base/species.py new file mode 100644 index 0000000..9fb08fb --- /dev/null +++ b/gn3/base/species.py @@ -0,0 +1,64 @@ + +# pylint: disable-all +import collections +from flask import g +from dataclasses import dataclass + +class TheSpecies: + def __init__(self, dataset=None, species_name=None): + if species_name is not None: + self.name = species_name + + self.chromosomes = Chromosomes(species=self.name) + + else: + self.dataset = dataset + self.chromosomes = Chromosomes(dataset=self.dataset) + + +class Chromosomes: + def __init__(self, dataset=None, species=None): + self.chromosomes = collections.OrderedDict() + + if species is not None: + query = """ + Select + Chr_Length.Name, Chr_Length.OrderId, Length from Chr_Length, Species + where + Chr_Length.SpeciesId = Species.SpeciesId AND + Species.Name = '%s' + Order by OrderId + """ % species.capitalize() + + else: + self.dataset = dataset + + query = """ + Select + Chr_Length.Name, Chr_Length.OrderId, Length from Chr_Length, InbredSet + where + Chr_Length.SpeciesId = InbredSet.SpeciesId AND + InbredSet.Name = '%s' + Order by OrderId + """ % self.dataset.group.name + + # logger.sql(query) + + results = g.db.execute(query).fetchall() + + for item in results: + self.chromosomes[item.OrderId] = IndChromosome( + item.Name, item.Length) + + +# @dataclass +class IndChromosome: + def __init__(self,name,length): + self.name= name + self.length = length + + @property + def mb_length(self): + """Chromosome length in megabases""" + return self.length/ 1000000 + diff --git a/gn3/base/trait.py b/gn3/base/trait.py new file mode 100644 index 0000000..f4be61c --- /dev/null +++ b/gn3/base/trait.py @@ -0,0 +1,366 @@ + +# pylint: disable-all +from flask import g +from redis import Redis +from gn3.utility.db_tools import escape +from gn3.base.webqtlCaseData import webqtlCaseData + + +def check_resource_availability(dataset, name=None): + + # todo add code for this + # should probably work on this has to do with authentication + return {'data': ['no-access', 'view'], 'metadata': ['no-access', 'view'], 'admin': ['not-admin']} + + +def create_trait(**kw): + # work on check resource availability deals with authentication + assert bool(kw.get("dataset")) != bool( + kw.get('dataset_name')), "Needs dataset ob. or name" + + assert bool(kw.get("name")), "Need trait name" + + if kw.get('dataset_name'): + if kw.get('dataset_name') != "Temp": + dataset = create_dataset(kw.get('dataset_name')) + else: + dataset = kw.get('dataset') + + if dataset.type == 'Publish': + permissions = check_resource_availability( + dataset, kw.get('name')) + else: + permissions = check_resource_availability(dataset) + + if "view" in permissions['data']: + the_trait = GeneralTrait(**kw) + if the_trait.dataset.type != "Temp": + the_trait = retrieve_trait_info( + the_trait, + the_trait.dataset, + get_qtl_info=kw.get('get_qtl_info')) + + + return the_trait + + return None + + +class GeneralTrait: + def __init__(self, get_qtl_info=False, get_sample_info=True, **kw): + assert bool(kw.get('dataset')) != bool( + kw.get('dataset_name')), "Needs dataset ob. or name" + # Trait ID, ProbeSet ID, Published ID, etc. + self.name = kw.get('name') + if kw.get('dataset_name'): + if kw.get('dataset_name') == "Temp": + temp_group = self.name.split("_")[2] + self.dataset = create_dataset( + dataset_name="Temp", + dataset_type="Temp", + group_name=temp_group) + + else: + self.dataset = create_dataset(kw.get('dataset_name')) + + else: + self.dataset = kw.get("dataset") + + self.cellid = kw.get('cellid') + self.identification = kw.get('identification', 'un-named trait') + self.haveinfo = kw.get('haveinfo', False) + self.sequence = kw.get('sequence') + self.data = kw.get('data', {}) + self.view = True + + # Sets defaults + self.locus = None + self.lrs = None + self.pvalue = None + self.mean = None + self.additive = None + self.num_overlap = None + self.strand_probe = None + self.symbol = None + self.display_name = self.name + self.LRS_score_repr = "N/A" + self.LRS_location_repr = "N/A" + + if kw.get('fullname'): + name2 = value.split("::") + if len(name2) == 2: + self.dataset, self.name = name2 + + elif len(name2) == 3: + self.dataset, self.name, self.cellid = name2 + + # Todo: These two lines are necessary most of the time, but + # perhaps not all of the time So we could add a simple if + # statement to short-circuit this if necessary + if get_sample_info is not False: + self = retrieve_sample_data(self, self.dataset) + + +def retrieve_sample_data(trait, dataset, samplelist=None): + if samplelist is None: + samplelist = [] + + if dataset.type == "Temp": + results = Redis.get(trait.name).split() + + else: + results = dataset.retrieve_sample_data(trait.name) + + # Todo: is this necessary? If not remove + trait.data.clear() + + if results: + if dataset.type == "Temp": + all_samples_ordered = dataset.group.all_samples_ordered() + for i, item in enumerate(results): + try: + trait.data[all_samples_ordered[i]] = webqtlCaseData( + all_samples_ordered[i], float(item)) + + except Exception as e: + pass + + + else: + for item in results: + name, value, variance, num_cases, name2 = item + if not samplelist or (samplelist and name in samplelist): + trait.data[name] = webqtlCaseData(*item) + + return trait + +def retrieve_trait_info(trait, dataset, get_qtl_info=False): + assert dataset, "Dataset doesn't exist" + + the_url = None + # some code should be added added here + + try: + response = requests.get(the_url).content + trait_info = json.loads(response) + except: # ZS: I'm assuming the trait is viewable if the try fails for some reason; it should never reach this point unless the user has privileges, since that's dealt with in create_trait + if dataset.type == 'Publish': + query = """ + SELECT + PublishXRef.Id, InbredSet.InbredSetCode, Publication.PubMed_ID, + CAST(Phenotype.Pre_publication_description AS BINARY), + CAST(Phenotype.Post_publication_description AS BINARY), + CAST(Phenotype.Original_description AS BINARY), + CAST(Phenotype.Pre_publication_abbreviation AS BINARY), + CAST(Phenotype.Post_publication_abbreviation AS BINARY), PublishXRef.mean, + Phenotype.Lab_code, Phenotype.Submitter, Phenotype.Owner, Phenotype.Authorized_Users, + CAST(Publication.Authors AS BINARY), CAST(Publication.Title AS BINARY), CAST(Publication.Abstract AS BINARY), + CAST(Publication.Journal AS BINARY), Publication.Volume, Publication.Pages, + Publication.Month, Publication.Year, PublishXRef.Sequence, + Phenotype.Units, PublishXRef.comments + FROM + PublishXRef, Publication, Phenotype, PublishFreeze, InbredSet + WHERE + PublishXRef.Id = %s AND + Phenotype.Id = PublishXRef.PhenotypeId AND + Publication.Id = PublishXRef.PublicationId AND + PublishXRef.InbredSetId = PublishFreeze.InbredSetId AND + PublishXRef.InbredSetId = InbredSet.Id AND + PublishFreeze.Id = %s + """ % (trait.name, dataset.id) + + trait_info = g.db.execute(query).fetchone() + + # XZ, 05/08/2009: Xiaodong add this block to use ProbeSet.Id to find the probeset instead of just using ProbeSet.Name + # XZ, 05/08/2009: to avoid the problem of same probeset name from different platforms. + elif dataset.type == 'ProbeSet': + display_fields_string = ', ProbeSet.'.join(dataset.display_fields) + display_fields_string = 'ProbeSet.' + display_fields_string + query = """ + SELECT %s + FROM ProbeSet, ProbeSetFreeze, ProbeSetXRef + WHERE + ProbeSetXRef.ProbeSetFreezeId = ProbeSetFreeze.Id AND + ProbeSetXRef.ProbeSetId = ProbeSet.Id AND + ProbeSetFreeze.Name = '%s' AND + ProbeSet.Name = '%s' + """ % (escape(display_fields_string), + escape(dataset.name), + escape(str(trait.name))) + + trait_info = g.db.execute(query).fetchone() + # XZ, 05/08/2009: We also should use Geno.Id to find marker instead of just using Geno.Name + # to avoid the problem of same marker name from different species. + elif dataset.type == 'Geno': + display_fields_string = ',Geno.'.join(dataset.display_fields) + display_fields_string = 'Geno.' + display_fields_string + query = """ + SELECT %s + FROM Geno, GenoFreeze, GenoXRef + WHERE + GenoXRef.GenoFreezeId = GenoFreeze.Id AND + GenoXRef.GenoId = Geno.Id AND + GenoFreeze.Name = '%s' AND + Geno.Name = '%s' + """ % (escape(display_fields_string), + escape(dataset.name), + escape(trait.name)) + + trait_info = g.db.execute(query).fetchone() + else: # Temp type + query = """SELECT %s FROM %s WHERE Name = %s""" + + trait_info = g.db.execute(query, + ','.join(dataset.display_fields), + dataset.type, trait.name).fetchone() + + if trait_info: + trait.haveinfo = True + for i, field in enumerate(dataset.display_fields): + holder = trait_info[i] + if isinstance(holder, bytes): + holder = holder.decode("utf-8", errors="ignore") + setattr(trait, field, holder) + + if dataset.type == 'Publish': + if trait.group_code: + trait.display_name = trait.group_code + "_" + str(trait.name) + + trait.confidential = 0 + if trait.pre_publication_description and not trait.pubmed_id: + trait.confidential = 1 + + description = trait.post_publication_description + + # If the dataset is confidential and the user has access to confidential + # phenotype traits, then display the pre-publication description instead + # of the post-publication description + trait.description_display = "" + if not trait.pubmed_id: + trait.abbreviation = trait.pre_publication_abbreviation + trait.description_display = trait.pre_publication_description + else: + trait.abbreviation = trait.post_publication_abbreviation + if description: + trait.description_display = description.strip() + + if not trait.year.isdigit(): + trait.pubmed_text = "N/A" + else: + trait.pubmed_text = trait.year + + # moved to config + + PUBMEDLINK_URL = "http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=%s&dopt=Abstract" + + if trait.pubmed_id: + trait.pubmed_link = PUBMEDLINK_URL % trait.pubmed_id + + if dataset.type == 'ProbeSet' and dataset.group: + description_string = trait.description + target_string = trait.probe_target_description + + if str(description_string or "") != "" and description_string != 'None': + description_display = description_string + else: + description_display = trait.symbol + + if (str(description_display or "") != "" and + description_display != 'N/A' and + str(target_string or "") != "" and target_string != 'None'): + description_display = description_display + '; ' + target_string.strip() + + # Save it for the jinja2 template + trait.description_display = description_display + + trait.location_repr = 'N/A' + if trait.chr and trait.mb: + trait.location_repr = 'Chr%s: %.6f' % ( + trait.chr, float(trait.mb)) + + elif dataset.type == "Geno": + trait.location_repr = 'N/A' + if trait.chr and trait.mb: + trait.location_repr = 'Chr%s: %.6f' % ( + trait.chr, float(trait.mb)) + + if get_qtl_info: + # LRS and its location + trait.LRS_score_repr = "N/A" + trait.LRS_location_repr = "N/A" + trait.locus = trait.locus_chr = trait.locus_mb = trait.lrs = trait.pvalue = trait.additive = "" + if dataset.type == 'ProbeSet' and not trait.cellid: + trait.mean = "" + query = """ + SELECT + ProbeSetXRef.Locus, ProbeSetXRef.LRS, ProbeSetXRef.pValue, ProbeSetXRef.mean, ProbeSetXRef.additive + FROM + ProbeSetXRef, ProbeSet + WHERE + ProbeSetXRef.ProbeSetId = ProbeSet.Id AND + ProbeSet.Name = "{}" AND + ProbeSetXRef.ProbeSetFreezeId ={} + """.format(trait.name, dataset.id) + + trait_qtl = g.db.execute(query).fetchone() + if trait_qtl: + trait.locus, trait.lrs, trait.pvalue, trait.mean, trait.additive = trait_qtl + if trait.locus: + query = """ + select Geno.Chr, Geno.Mb from Geno, Species + where Species.Name = '{}' and + Geno.Name = '{}' and + Geno.SpeciesId = Species.Id + """.format(dataset.group.species, trait.locus) + + result = g.db.execute(query).fetchone() + if result: + trait.locus_chr = result[0] + trait.locus_mb = result[1] + else: + trait.locus = trait.locus_chr = trait.locus_mb = trait.additive = "" + else: + trait.locus = trait.locus_chr = trait.locus_mb = trait.additive = "" + + if dataset.type == 'Publish': + query = """ + SELECT + PublishXRef.Locus, PublishXRef.LRS, PublishXRef.additive + FROM + PublishXRef, PublishFreeze + WHERE + PublishXRef.Id = %s AND + PublishXRef.InbredSetId = PublishFreeze.InbredSetId AND + PublishFreeze.Id =%s + """ % (trait.name, dataset.id) + + trait_qtl = g.db.execute(query).fetchone() + if trait_qtl: + trait.locus, trait.lrs, trait.additive = trait_qtl + if trait.locus: + query = """ + select Geno.Chr, Geno.Mb from Geno, Species + where Species.Name = '{}' and + Geno.Name = '{}' and + Geno.SpeciesId = Species.Id + """.format(dataset.group.species, trait.locus) + + result = g.db.execute(query).fetchone() + if result: + trait.locus_chr = result[0] + trait.locus_mb = result[1] + else: + trait.locus = trait.locus_chr = trait.locus_mb = trait.additive = "" + else: + trait.locus = trait.locus_chr = trait.locus_mb = trait.additive = "" + else: + trait.locus = trait.lrs = trait.additive = "" + if (dataset.type == 'Publish' or dataset.type == "ProbeSet") and str(trait.locus_chr or "") != "" and str(trait.locus_mb or "") != "": + trait.LRS_location_repr = LRS_location_repr = 'Chr%s: %.6f' % ( + trait.locus_chr, float(trait.locus_mb)) + if str(trait.lrs or "") != "": + trait.LRS_score_repr = LRS_score_repr = '%3.1f' % trait.lrs + else: + raise KeyError(repr(trait.name) + + ' information is not found in the database.') + return trait diff --git a/gn3/base/webqtlCaseData.py b/gn3/base/webqtlCaseData.py new file mode 100644 index 0000000..8395af8 --- /dev/null +++ b/gn3/base/webqtlCaseData.py @@ -0,0 +1,84 @@ +# Copyright (C) University of Tennessee Health Science Center, Memphis, TN. +# +# This program is free software: you can redistribute it and/or modify it +# under the terms of the GNU Affero General Public License +# as published by the Free Software Foundation, either version 3 of the +# License, or (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +# See the GNU Affero General Public License for more details. +# +# This program is available from Source Forge: at GeneNetwork Project +# (sourceforge.net/projects/genenetwork/). +# +# Contact Drs. Robert W. Williams and Xiaodong Zhou (2010) +# at rwilliams@uthsc.edu and xzhou15@uthsc.edu +# +# This module is used by GeneNetwork project (www.genenetwork.org) +# +# Created by GeneNetwork Core Team 2010/08/10 + + +# uncomment below + +# from utility.logger import getLogger +# logger = getLogger(__name__) + +# import utility.tools + +# utility.tools.show_settings() +# pylint: disable-all + +class webqtlCaseData: + """one case data in one trait""" + + def __init__(self, name, value=None, variance=None, num_cases=None, name2=None): + self.name = name + self.name2 = name2 # Other name (for traits like BXD65a) + self.value = value # Trait Value + self.variance = variance # Trait Variance + self.num_cases = num_cases # Number of individuals/cases + self.extra_attributes = None + self.this_id = None # Set a sane default (can't be just "id" cause that's a reserved word) + self.outlier = None # Not set to True/False until later + + def __repr__(self): + case_data_string = "<webqtlCaseData> " + if self.value is not None: + case_data_string += "value=%2.3f" % self.value + if self.variance is not None: + case_data_string += " variance=%2.3f" % self.variance + if self.num_cases: + case_data_string += " ndata=%s" % self.num_cases + if self.name: + case_data_string += " name=%s" % self.name + if self.name2: + case_data_string += " name2=%s" % self.name2 + return case_data_string + + @property + def class_outlier(self): + """Template helper""" + if self.outlier: + return "outlier" + return "" + + @property + def display_value(self): + if self.value is not None: + return "%2.3f" % self.value + return "x" + + @property + def display_variance(self): + if self.variance is not None: + return "%2.3f" % self.variance + return "x" + + @property + def display_num_cases(self): + if self.num_cases is not None: + return "%s" % self.num_cases + return "x"
\ No newline at end of file |