diff options
| author | Pjotr Prins | 2016-05-29 17:20:20 +0000 |
|---|---|---|
| committer | Pjotr Prins | 2016-05-29 17:20:20 +0000 |
| commit | 4b083f2cdfa493f7b2ccc3c30cc5bb6cad694d3a (patch) | |
| tree | c2c2503d88868f13b06dd66328720486e72f647a /wqflask | |
| parent | 33d817c81b4b22bc051dbde2b26c5d4de028369e (diff) | |
| parent | 0d22d3bc72cfc35cb23efce3d1687477880a5b3e (diff) | |
| download | genenetwork2-4b083f2cdfa493f7b2ccc3c30cc5bb6cad694d3a.tar.gz | |
Merge branch 'master' of github.com:genenetwork/genenetwork2
Diffstat (limited to 'wqflask')
19 files changed, 939 insertions, 513 deletions
diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index 053b45fc..4953e728 100644 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -711,7 +711,7 @@ class PhenotypeDataSet(DataSet): def retrieve_sample_data(self, trait): query = """ SELECT - Strain.Name, PublishData.value, PublishSE.error, NStrain.count + Strain.Name, PublishData.value, PublishSE.error, NStrain.count, Strain.Name2 FROM (PublishData, Strain, PublishXRef, PublishFreeze) left join PublishSE on @@ -803,7 +803,7 @@ class GenotypeDataSet(DataSet): def retrieve_sample_data(self, trait): query = """ SELECT - Strain.Name, GenoData.value, GenoSE.error, GenoData.Id + Strain.Name, GenoData.value, GenoSE.error, GenoData.Id, Sample.Name2 FROM (GenoData, GenoFreeze, Strain, Geno, GenoXRef) left join GenoSE on @@ -1031,7 +1031,7 @@ class MrnaAssayDataSet(DataSet): def retrieve_sample_data(self, trait): query = """ SELECT - Strain.Name, ProbeSetData.value, ProbeSetSE.error, ProbeSetData.Id + Strain.Name, ProbeSetData.value, ProbeSetSE.error, ProbeSetData.Id, Strain.Name2 FROM (ProbeSetData, ProbeSetFreeze, Strain, ProbeSet, ProbeSetXRef) left join ProbeSetSE on diff --git a/wqflask/base/trait.py b/wqflask/base/trait.py index 6c5ca8b2..d1c0be83 100644 --- a/wqflask/base/trait.py +++ b/wqflask/base/trait.py @@ -180,13 +180,15 @@ class GeneralTrait(object): samples = [] vals = [] the_vars = [] + sample_aliases = [] for sample_name, sample_data in self.data.items(): if sample_data.value != None: if not include_variance or sample_data.variance != None: samples.append(sample_name) vals.append(sample_data.value) the_vars.append(sample_data.variance) - return samples, vals, the_vars + sample_aliases.append(sample_data.name2) + return samples, vals, the_vars, sample_aliases # @@ -230,7 +232,7 @@ class GeneralTrait(object): if results: for item in results: - name, value, variance, num_cases = item + name, value, variance, num_cases, name2 = item if not samplelist or (samplelist and name in samplelist): self.data[name] = webqtlCaseData(*item) #name, value, variance, num_cases) @@ -313,9 +315,9 @@ class GeneralTrait(object): self.confidential = 0 if self.pre_publication_description and not self.pubmed_id: self.confidential = 1 - - description = self.post_publication_description + description = self.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 @@ -329,7 +331,7 @@ class GeneralTrait(object): # # description = self.pre_publication_description - if len(description) > 0: + if description: self.description_display = description.strip() else: self.description_display = "" @@ -479,7 +481,7 @@ class GeneralTrait(object): else: self.locus = self.lrs = self.additive = "" - if self.locus_chr != "" and self.locus_mb != "": + if (self.dataset.type == 'Publish' or self.dataset.type == "ProbeSet") and self.locus_chr != "" and self.locus_mb != "": #XZ: LRS_location_value is used for sorting try: LRS_location_value = int(self.locus_chr)*1000 + float(self.locus_mb) diff --git a/wqflask/base/webqtlCaseData.py b/wqflask/base/webqtlCaseData.py index 42763aed..99a34866 100755 --- a/wqflask/base/webqtlCaseData.py +++ b/wqflask/base/webqtlCaseData.py @@ -29,8 +29,9 @@ print("Mr. Mojo Risin 2") class webqtlCaseData(object): """one case data in one trait""" - def __init__(self, name, value=None, variance=None, num_cases=None): + 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 diff --git a/wqflask/base/webqtlConfig.py b/wqflask/base/webqtlConfig.py index 0358bcbf..d0016b33 100755..100644 --- a/wqflask/base/webqtlConfig.py +++ b/wqflask/base/webqtlConfig.py @@ -69,7 +69,7 @@ GENERATED_TEXT_DIR = mk_dir(TMPDIR+'/generated_text/') # Flat file directories GENODIR = flat_files('genotype')+'/' -JSON_GENODIR = assert_dir(GENODIR+'json/') +JSON_GENODIR = flat_files('json')+'/' PORTADDR = "http://50.16.251.170" diff --git a/wqflask/utility/webqtlUtil.py b/wqflask/utility/webqtlUtil.py index f842dde0..1108614b 100755 --- a/wqflask/utility/webqtlUtil.py +++ b/wqflask/utility/webqtlUtil.py @@ -509,7 +509,7 @@ def calCorrelationRank(xVals,yVals,N): j = 0 for i in range(len(xVals)): - if xVals[i]!= None and yVals[i]!= None: + if (xVals[i]!= None and yVals[i]!= None) and (xVals[i] != "None" and yVals[i] != "None"): XX.append((j,xVals[i])) YY.append((j,yVals[i])) j = j+1 diff --git a/wqflask/wqflask/correlation/show_corr_results.py b/wqflask/wqflask/correlation/show_corr_results.py index dd661092..6d8dd76a 100755..100644 --- a/wqflask/wqflask/correlation/show_corr_results.py +++ b/wqflask/wqflask/correlation/show_corr_results.py @@ -50,6 +50,7 @@ from dbFunction import webqtlDatabaseFunction import utility.webqtlUtil #this is for parallel computing only. from wqflask.correlation import correlation_functions from utility.benchmark import Bench +import utility.webqtlUtil from MySQLdb import escape_string as escape @@ -159,6 +160,9 @@ class CorrelationResults(object): self.correlation_data = {} + db_filename = self.getFileName(target_db_name = self.target_dataset.name) + cache_available = db_filename in os.listdir(webqtlConfig.TEXTDIR) + if self.corr_type == "tissue": self.trait_symbol_dict = self.dataset.retrieve_genes("Symbol") @@ -174,9 +178,25 @@ class CorrelationResults(object): self.get_sample_r_and_p_values(trait, self.target_dataset.trait_data[trait]) elif self.corr_type == "sample": - # print("self.target_dataset.trait_data: %d" % len(self.target_dataset.trait_data)) - for trait, values in self.target_dataset.trait_data.iteritems(): - self.get_sample_r_and_p_values(trait, values) + if self.dataset.type == "ProbeSet" and cache_available: + dataset_file = open(webqtlConfig.TEXTDIR+db_filename,'r') + + #XZ, 01/08/2009: read the first line + line = dataset_file.readline() + dataset_strains = webqtlUtil.readLineCSV(line)[1:] + + self.this_trait_vals = [] + for item in dataset_strains: + if item in self.sample_data: + self.this_trait_vals.append(self.sample_data[item]) + else: + self.this_trait_vals.append("None") + num_overlap = len(self.this_trait_vals) + + self.do_parallel_correlation(db_filename, num_overlap) + else: + for trait, values in self.target_dataset.trait_data.iteritems(): + self.get_sample_r_and_p_values(trait, values) self.correlation_data = collections.OrderedDict(sorted(self.correlation_data.items(), key=lambda t: -abs(t[1][0]))) @@ -190,7 +210,7 @@ class CorrelationResults(object): range_chr_as_int = order_id for _trait_counter, trait in enumerate(self.correlation_data.keys()[:self.return_number]): - trait_object = GeneralTrait(dataset=self.target_dataset, name=trait, get_qtl_info=True) + trait_object = GeneralTrait(dataset=self.target_dataset, name=trait, get_qtl_info=True, get_sample_info=False) if self.dataset.type == "ProbeSet" or self.dataset.type == "Geno": #ZS: Convert trait chromosome to an int for the location range option @@ -308,7 +328,7 @@ class CorrelationResults(object): #traitList = self.correlate() - #_log.info("Done doing correlation calculation") + #print("Done doing correlation calculation") ############################################################################################################################################ @@ -521,27 +541,126 @@ class CorrelationResults(object): """ - # print("len(self.sample_data):", len(self.sample_data)) - - this_trait_vals = [] + self.this_trait_vals = [] target_vals = [] for index, sample in enumerate(self.target_dataset.samplelist): if sample in self.sample_data: sample_value = self.sample_data[sample] target_sample_value = target_samples[index] - this_trait_vals.append(sample_value) + self.this_trait_vals.append(sample_value) target_vals.append(target_sample_value) - this_trait_vals, target_vals, num_overlap = corr_result_helpers.normalize_values( - 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': - sample_r, sample_p = scipy.stats.pearsonr(this_trait_vals, target_vals) + sample_r, sample_p = scipy.stats.pearsonr(self.this_trait_vals, target_vals) else: - sample_r, sample_p = scipy.stats.spearmanr(this_trait_vals, target_vals) + sample_r, sample_p = scipy.stats.spearmanr(self.this_trait_vals, target_vals) - self.correlation_data[trait] = [sample_r, sample_p, num_overlap] + if num_overlap > 5: + self.correlation_data[trait] = [sample_r, sample_p, num_overlap] + + + """ + correlations = [] + + #XZ: Use the fast method only for probeset dataset, and this dataset must have been created. + #XZ: Otherwise, use original method + #print("Entering correlation") + + #db_filename = self.getFileName(target_db_name=self.target_db_name) + # + #cache_available = db_filename in os.listdir(webqtlConfig.TEXTDIR) + + # If the cache file exists, do a cached correlation for probeset data + if self.dataset.type == "ProbeSet": +# if self.method in [METHOD_SAMPLE_PEARSON, METHOD_SAMPLE_RANK] and cache_available: +# traits = do_parallel_correlation() +# +# else: + + traits = self.get_traits(self.vals) + + for trait in traits: + trait.calculate_correlation(vals, self.method) + + self.record_count = len(traits) #ZS: This isn't a good way to get this value, so I need to change it later + + #XZ, 3/31/2010: Theoretically, we should create one function 'comTissueCorr' + #to compare each trait by their tissue corr p values. + #But because the tissue corr p values are generated by permutation test, + #the top ones always have p value 0. So comparing p values actually does nothing. + #In addition, for the tissue data in our database, the N is always the same. + #So it's safe to compare with tissue corr statistic value. + #That's the same as literature corr. + #if self.method in [METHOD_LIT, METHOD_TISSUE_PEARSON, METHOD_TISSUE_RANK] and self.gene_id: + # traits.sort(webqtlUtil.cmpLitCorr) + #else: + #if self.method in TISSUE_METHODS: + # sort(traits, key=lambda A: math.fabs(A.tissue_corr)) + #elif self.method == METHOD_LIT: + # traits.sort(traits, key=lambda A: math.fabs(A.lit_corr)) + #else: + traits = sortTraitCorrelations(traits, self.method) + + # Strip to the top N correlations + traits = traits[:min(self.returnNumber, len(traits))] + + addLiteratureCorr = False + addTissueCorr = False + + trait_list = [] + for trait in traits: + db_trait = webqtlTrait(db=self.db, name=trait.name, cursor=self.cursor) + db_trait.retrieveInfo( QTL='Yes' ) + + db_trait.Name = trait.name + db_trait.corr = trait.correlation + db_trait.nOverlap = trait.overlap + db_trait.corrPValue = trait.p_value + + # NL, 07/19/2010 + # js function changed, add a new parameter rankOrder for js function 'showTissueCorrPlot' + db_trait.RANK_ORDER = self.RANK_ORDERS[self.method] + + #XZ, 26/09/2008: Method is 4 or 5. Have fetched tissue corr, but no literature correlation yet. + if self.method in TISSUE_METHODS: + db_trait.tissueCorr = trait.tissue_corr + db_trait.tissuePValue = trait.p_tissue + addTissueCorr = True + + + #XZ, 26/09/2008: Method is 3, Have fetched literature corr, but no tissue corr yet. + elif self.method == METHOD_LIT: + db_trait.LCorr = trait.lit_corr + db_trait.mouse_geneid = self.translateToMouseGeneID(self.species, db_trait.geneid) + addLiteratureCorr = True + + #XZ, 26/09/2008: Method is 1 or 2. Have NOT fetched literature corr and tissue corr yet. + # Phenotype data will not have geneid, and neither will some probes + # we need to handle this because we will get an attribute error + else: + if self.input_trait_mouse_gene_id and self.db.type=="ProbeSet": + addLiteratureCorr = True + if self.trait_symbol and self.db.type=="ProbeSet": + addTissueCorr = True + + trait_list.append(db_trait) + + if addLiteratureCorr: + trait_list = self.getLiteratureCorrelationByList(self.input_trait_mouse_gene_id, + self.species, trait_list) + if addTissueCorr: + trait_list = self.getTissueCorrelationByList( + primaryTraitSymbol = self.trait_symbol, + traitList = trait_list, + TissueProbeSetFreezeId = TISSUE_MOUSE_DB, + method=self.method) + + return trait_list + """ + def do_tissue_corr_for_all_traits_2(self): """Comments Possibly Out of Date!!!!! @@ -670,38 +789,6 @@ class CorrelationResults(object): # except: return False - def get_all_dataset_data(self): - - """ - SELECT ProbeSet.Name, T128.value, T129.value, T130.value, T131.value, T132.value, T134.value, T135.value, T138.value, T139.value, T140.value, T141.value, T142.value, T144 - .value, T145.value, T147.value, T148.value, T149.value, T487.value, T919.value, T920.value, T922.value - FROM (ProbeSet, ProbeSetXRef, ProbeSetFreeze) - left join ProbeSetData as T128 on T128.Id = ProbeSetXRef.DataId and T128.StrainId=128 - left join ProbeSetData as T129 on T129.Id = ProbeSetXRef.DataId and T129.StrainId=129 - left join ProbeSetData as T130 on T130.Id = ProbeSetXRef.DataId and T130.StrainId=130 - left join ProbeSetData as T131 on T131.Id = ProbeSetXRef.DataId and T131.StrainId=131 - left join ProbeSetData as T132 on T132.Id = ProbeSetXRef.DataId and T132.StrainId=132 - left join ProbeSetData as T134 on T134.Id = ProbeSetXRef.DataId and T134.StrainId=134 - left join ProbeSetData as T135 on T135.Id = ProbeSetXRef.DataId and T135.StrainId=135 - left join ProbeSetData as T138 on T138.Id = ProbeSetXRef.DataId and T138.StrainId=138 - left join ProbeSetData as T139 on T139.Id = ProbeSetXRef.DataId and T139.StrainId=139 - left join ProbeSetData as T140 on T140.Id = ProbeSetXRef.DataId and T140.StrainId=140 - left join ProbeSetData as T141 on T141.Id = ProbeSetXRef.DataId and T141.StrainId=141 - left join ProbeSetData as T142 on T142.Id = ProbeSetXRef.DataId and T142.StrainId=142 - left join ProbeSetData as T144 on T144.Id = ProbeSetXRef.DataId and T144.StrainId=144 - left join ProbeSetData as T145 on T145.Id = ProbeSetXRef.DataId and T145.StrainId=145 - left join ProbeSetData as T147 on T147.Id = ProbeSetXRef.DataId and T147.StrainId=147 - left join ProbeSetData as T148 on T148.Id = ProbeSetXRef.DataId and T148.StrainId=148 - left join ProbeSetData as T149 on T149.Id = ProbeSetXRef.DataId and T149.StrainId=149 - left join ProbeSetData as T487 on T487.Id = ProbeSetXRef.DataId and T487.StrainId=487 - left join ProbeSetData as T919 on T919.Id = ProbeSetXRef.DataId and T919.StrainId=919 - left join ProbeSetData as T920 on T920.Id = ProbeSetXRef.DataId and T920.StrainId=920 - left join ProbeSetData as T922 on T922.Id = ProbeSetXRef.DataId and T922.StrainId=922 - WHERE ProbeSetXRef.ProbeSetFreezeId = ProbeSetFreeze.Id and - ProbeSetFreeze.Name = 'HC_M2_0606_P' and - ProbeSet.Id = ProbeSetXRef.ProbeSetId order by ProbeSet.Id - """ - def process_samples(self, start_vars, sample_names, excluded_samples=None): if not excluded_samples: excluded_samples = () @@ -988,59 +1075,7 @@ class CorrelationResults(object): totalTraits = len(traits) #XZ, 09/18/2008: total trait number return traits - - - def do_parallel_correlation(self): - _log.info("Invoking parallel computing") - input_line_list = datasetFile.readlines() - _log.info("Read lines from the file") - all_line_number = len(input_line_list) - - step = 1000 - job_number = math.ceil( float(all_line_number)/step ) - - job_input_lists = [] - - _log.info("Configuring jobs") - - for job_index in range( int(job_number) ): - starti = job_index*step - endi = min((job_index+1)*step, all_line_number) - - one_job_input_list = [] - - for i in range( starti, endi ): - one_job_input_list.append( input_line_list[i] ) - - job_input_lists.append( one_job_input_list ) - - _log.info("Creating pp servers") - - ppservers = () - # Creates jobserver with automatically detected number of workers - job_server = pp.Server(ppservers=ppservers) - - _log.info("Done creating servers") - - jobs = [] - results = [] - - _log.info("Starting parallel computation, submitting jobs") - for one_job_input_list in job_input_lists: #pay attention to modules from outside - jobs.append( job_server.submit(func=compute_corr, args=(nnCorr, _newvals, one_job_input_list, self.method), depfuncs=(), modules=("utility.webqtlUtil",)) ) - _log.info("Done submitting jobs") - - for one_job in jobs: - one_result = one_job() - results.append( one_result ) - - _log.info("Acquiring results") - - for one_result in results: - for one_traitinfo in one_result: - allcorrelations.append( one_traitinfo ) - - _log.info("Appending the results") + def calculate_corr_for_all_tissues(self, tissue_dataset_id=None): symbol_corr_dict = {} @@ -1065,10 +1100,7 @@ class CorrelationResults(object): # SymbolValueDict) return (symbolCorrDict, symbolPvalueDict) - datasetFile.close() - totalTraits = len(allcorrelations) - _log.info("Done correlating using the fast method") - + def correlate(self): self.correlation_data = collections.defaultdict(list) @@ -1083,107 +1115,254 @@ 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 + Used by correlationPage""" + + dataset_id = str(self.target_dataset.id) + dataset_fullname = self.target_dataset.fullname.replace(' ','_') + dataset_fullname = dataset_fullname.replace('/','_') + FileName = 'ProbeSetFreezeId_' + dataset_id + '_FullName_' + dataset_fullname + '.txt' - """ - correlations = [] - - #XZ: Use the fast method only for probeset dataset, and this dataset must have been created. - #XZ: Otherwise, use original method - #_log.info("Entering correlation") - - #db_filename = self.getFileName(target_db_name=self.target_db_name) - # - #cache_available = db_filename in os.listdir(webqtlConfig.TEXTDIR) - - # If the cache file exists, do a cached correlation for probeset data - if self.dataset.type == "ProbeSet": -# if self.method in [METHOD_SAMPLE_PEARSON, METHOD_SAMPLE_RANK] and cache_available: -# traits = do_parallel_correlation() -# -# else: - - traits = self.get_traits(self.vals) + 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 + def compute_corr(input_nnCorr, input_trait, input_list, corr_method): + + import math + import reaper + + def calCorrelation(dbdata,userdata,N): + X = [] + Y = [] + for i in range(N): + if (dbdata[i] != None and userdata[i] != None) and (dbdata[i] != "None" and userdata[i] != "None"): + X.append(float(dbdata[i])) + Y.append(float(userdata[i])) + NN = len(X) + if NN <6: + return (0.0,NN) + sx = reduce(lambda x,y:x+y,X,0.0) + sy = reduce(lambda x,y:x+y,Y,0.0) + meanx = sx/NN + meany = sy/NN + xyd = 0.0 + sxd = 0.0 + syd = 0.0 + for i in range(NN): + xyd += (X[i] - meanx)*(Y[i]-meany) + sxd += (X[i] - meanx)*(X[i] - meanx) + syd += (Y[i] - meany)*(Y[i] - meany) + try: + corr = xyd/(math.sqrt(sxd)*math.sqrt(syd)) + except: + corr = 0 + return (corr,NN) + + def calCorrelationRank(xVals,yVals,N): + """ + Calculated Spearman Ranked Correlation. The algorithm works + by setting all tied ranks to the average of those ranks (for + example, if ranks 5-10 all have the same value, each will be set + to rank 7.5). + """ + + XX = [] + YY = [] + j = 0 + + for i in range(len(xVals)): + if (xVals[i]!= None and yVals[i]!= None) and (xVals[i] != "None" and yVals[i] != "None"): + XX.append((j,float(xVals[i]))) + YY.append((j,float(yVals[i]))) + j = j+1 + + NN = len(XX) + if NN <6: + return (0.0,NN) + XX.sort(cmpOrder2) + YY.sort(cmpOrder2) + X = [0]*NN + Y = [0]*NN + + j = 1 + rank = 0.0 + t = 0.0 + sx = 0.0 + + while j < NN: + + if XX[j][1] != XX[j-1][1]: + X[XX[j-1][0]] = j + j = j+1 - for trait in traits: - trait.calculate_correlation(vals, self.method) + else: + jt = j+1 + ji = j + for jt in range(j+1, NN): + if (XX[jt][1] != XX[j-1][1]): + break + rank = 0.5*(j+jt) + for ji in range(j-1, jt): + X[XX[ji][0]] = rank + t = jt-j + sx = sx + (t*t*t-t) + if (jt == NN-1): + if (XX[jt][1] == XX[j-1][1]): + X[XX[NN-1][0]] = rank + j = jt+1 + + if j == NN: + if X[XX[NN-1][0]] == 0: + X[XX[NN-1][0]] = NN + + j = 1 + rank = 0.0 + t = 0.0 + sy = 0.0 + + while j < NN: + + if YY[j][1] != YY[j-1][1]: + Y[YY[j-1][0]] = j + j = j+1 + else: + jt = j+1 + ji = j + for jt in range(j+1, NN): + if (YY[jt][1] != YY[j-1][1]): + break + rank = 0.5*(j+jt) + for ji in range(j-1, jt): + Y[YY[ji][0]] = rank + t = jt - j + sy = sy + (t*t*t-t) + if (jt == NN-1): + if (YY[jt][1] == YY[j-1][1]): + Y[YY[NN-1][0]] = rank + j = jt+1 + + if j == NN: + if Y[YY[NN-1][0]] == 0: + Y[YY[NN-1][0]] = NN + + D = 0.0 + + for i in range(NN): + D += (X[i]-Y[i])*(X[i]-Y[i]) + + fac = (1.0 -sx/(NN*NN*NN-NN))*(1.0-sy/(NN*NN*NN-NN)) + + return ((1-(6.0/(NN*NN*NN-NN))*(D+(sx+sy)/12.0))/math.sqrt(fac),NN) + + # allcorrelations = [] + + correlation_data = {} + for i, line in enumerate(input_list): + if i == 0: + continue + tokens = line.split('","') + tokens[-1] = tokens[-1][:-2] #remove the last " + tokens[0] = tokens[0][1:] #remove the first " + + traitdataName = tokens[0] + database_trait = tokens[1:] + + #print("database_trait:", database_trait) + + #ZS: 2015 could add biweight correlation, see http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3465711/ + # if corr_method == 'pearson': + # sample_r, sample_p = scipy.stats.pearsonr(input_trait, database_trait) + # else: + # sample_r, sample_p = scipy.stats.spearmanr(input_trait, database_trait) + + if corr_method == "pearson": #XZ: Pearson's r + sample_r, nOverlap = calCorrelation(input_trait, database_trait, input_nnCorr) + else: #XZ: Spearman's rho + sample_r, nOverlap = calCorrelationRank(input_trait, database_trait, input_nnCorr) + + #XZ: calculate corrPValue + if nOverlap < 3: + sample_p = 1.0 + else: + if abs(sample_r) >= 1.0: + sample_p = 0.0 + else: + z_value = 0.5*math.log((1.0+sample_r)/(1.0-sample_r)) + 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] + + # traitinfo = [traitdataName, sample_r, nOverlap] + # allcorrelations.append(traitinfo) - self.record_count = len(traits) #ZS: This isn't a good way to get this value, so I need to change it later + return correlation_data + # return allcorrelations + + + datasetFile = open(webqtlConfig.TEXTDIR+db_filename,'r') + + print("Invoking parallel computing") + input_line_list = datasetFile.readlines() + print("Read lines from the file") + all_line_number = len(input_line_list) - #XZ, 3/31/2010: Theoretically, we should create one function 'comTissueCorr' - #to compare each trait by their tissue corr p values. - #But because the tissue corr p values are generated by permutation test, - #the top ones always have p value 0. So comparing p values actually does nothing. - #In addition, for the tissue data in our database, the N is always the same. - #So it's safe to compare with tissue corr statistic value. - #That's the same as literature corr. - #if self.method in [METHOD_LIT, METHOD_TISSUE_PEARSON, METHOD_TISSUE_RANK] and self.gene_id: - # traits.sort(webqtlUtil.cmpLitCorr) - #else: - #if self.method in TISSUE_METHODS: - # sort(traits, key=lambda A: math.fabs(A.tissue_corr)) - #elif self.method == METHOD_LIT: - # traits.sort(traits, key=lambda A: math.fabs(A.lit_corr)) - #else: - traits = sortTraitCorrelations(traits, self.method) + step = 1000 + job_number = math.ceil( float(all_line_number)/step ) - # Strip to the top N correlations - traits = traits[:min(self.returnNumber, len(traits))] + print("JOB NUMBER", job_number) + + job_input_lists = [] - addLiteratureCorr = False - addTissueCorr = False + print("Configuring jobs") - trait_list = [] - for trait in traits: - db_trait = webqtlTrait(db=self.db, name=trait.name, cursor=self.cursor) - db_trait.retrieveInfo( QTL='Yes' ) + for job_index in range( int(job_number) ): + starti = job_index*step + endi = min((job_index+1)*step, all_line_number) - db_trait.Name = trait.name - db_trait.corr = trait.correlation - db_trait.nOverlap = trait.overlap - db_trait.corrPValue = trait.p_value + one_job_input_list = [] - # NL, 07/19/2010 - # js function changed, add a new parameter rankOrder for js function 'showTissueCorrPlot' - db_trait.RANK_ORDER = self.RANK_ORDERS[self.method] + for i in range( starti, endi ): + one_job_input_list.append( input_line_list[i] ) - #XZ, 26/09/2008: Method is 4 or 5. Have fetched tissue corr, but no literature correlation yet. - if self.method in TISSUE_METHODS: - db_trait.tissueCorr = trait.tissue_corr - db_trait.tissuePValue = trait.p_tissue - addTissueCorr = True + job_input_lists.append( one_job_input_list ) + print("Creating pp servers") - #XZ, 26/09/2008: Method is 3, Have fetched literature corr, but no tissue corr yet. - elif self.method == METHOD_LIT: - db_trait.LCorr = trait.lit_corr - db_trait.mouse_geneid = self.translateToMouseGeneID(self.species, db_trait.geneid) - addLiteratureCorr = True + ppservers = () + # Creates jobserver with automatically detected number of workers + job_server = pp.Server(ppservers=ppservers) - #XZ, 26/09/2008: Method is 1 or 2. Have NOT fetched literature corr and tissue corr yet. - # Phenotype data will not have geneid, and neither will some probes - # we need to handle this because we will get an attribute error - else: - if self.input_trait_mouse_gene_id and self.db.type=="ProbeSet": - addLiteratureCorr = True - if self.trait_symbol and self.db.type=="ProbeSet": - addTissueCorr = True + print("Done creating servers") - trait_list.append(db_trait) + jobs = [] + results = [] - if addLiteratureCorr: - trait_list = self.getLiteratureCorrelationByList(self.input_trait_mouse_gene_id, - self.species, trait_list) - if addTissueCorr: - trait_list = self.getTissueCorrelationByList( - primaryTraitSymbol = self.trait_symbol, - traitList = trait_list, - TissueProbeSetFreezeId = TISSUE_MOUSE_DB, - method=self.method) + print("Starting parallel computation, submitting jobs") + for one_job_input_list in job_input_lists: #pay attention to modules from outside + jobs.append( job_server.submit(func=compute_corr, args=(num_overlap, self.this_trait_vals, one_job_input_list, self.corr_method), depfuncs=(), modules=("webqtlUtil",)) ) + print("Done submitting jobs") - return trait_list - """ + for one_job in jobs: + one_result = one_job() + self.correlation_data.update(one_result) + # one_result = one_job() + # results.append( one_result ) + #print("CORRELATION DATA:", self.correlation_data) + + # print("Acquiring results") + # for one_result in results: + # for one_traitinfo in one_result: + # allcorrelations.append( one_traitinfo ) diff --git a/wqflask/wqflask/gsearch.py b/wqflask/wqflask/gsearch.py index 4cd3874c..4f9dc316 100644 --- a/wqflask/wqflask/gsearch.py +++ b/wqflask/wqflask/gsearch.py @@ -1,94 +1,94 @@ from __future__ import absolute_import, print_function, division from flask import Flask, g -from base.data_set import create_dataset -from base.trait import GeneralTrait -from dbFunction import webqtlDatabaseFunction +#from base.data_set import create_dataset +#from base.trait import GeneralTrait +#from dbFunction import webqtlDatabaseFunction -from utility.benchmark import Bench +#from utility.benchmark import Bench class GSearch(object): def __init__(self, kw): self.type = kw['type'] self.terms = kw['terms'] - if self.type == "gene": - sql = """ - SELECT - Species.`Name` AS species_name, - InbredSet.`Name` AS inbredset_name, - Tissue.`Name` AS tissue_name, - ProbeSetFreeze.Name AS probesetfreeze_name, - ProbeSet.Name AS probeset_name, - ProbeSet.Symbol AS probeset_symbol, - ProbeSet.`description` AS probeset_description, - ProbeSet.Chr AS chr, - ProbeSet.Mb AS mb, - ProbeSetXRef.Mean AS mean, - ProbeSetXRef.LRS AS lrs, - ProbeSetXRef.`Locus` AS locus, - ProbeSetXRef.`pValue` AS pvalue, - ProbeSetXRef.`additive` AS additive - FROM Species, InbredSet, ProbeSetXRef, ProbeSet, ProbeFreeze, ProbeSetFreeze, Tissue - WHERE InbredSet.`SpeciesId`=Species.`Id` - AND ProbeFreeze.InbredSetId=InbredSet.`Id` - AND ProbeFreeze.`TissueId`=Tissue.`Id` - AND ProbeSetFreeze.ProbeFreezeId=ProbeFreeze.Id - AND ( MATCH (ProbeSet.Name,ProbeSet.description,ProbeSet.symbol,alias,GenbankId, UniGeneId, Probe_Target_Description) AGAINST ('%s' IN BOOLEAN MODE) ) - AND ProbeSet.Id = ProbeSetXRef.ProbeSetId - AND ProbeSetXRef.ProbeSetFreezeId=ProbeSetFreeze.Id - AND ProbeSetFreeze.public > 0 - ORDER BY species_name, inbredset_name, tissue_name, probesetfreeze_name, probeset_name - LIMIT 6000 - """ % (self.terms) - with Bench("Running query"): - re = g.db.execute(sql).fetchall() - self.trait_list = [] - with Bench("Creating trait objects"): - for line in re: - dataset = create_dataset(line[3], "ProbeSet", get_samplelist=False) - trait_id = line[4] - #with Bench("Building trait object"): - this_trait = GeneralTrait(dataset=dataset, name=trait_id, get_qtl_info=True, get_sample_info=False) - self.trait_list.append(this_trait) + # if self.type == "gene": + # sql = """ + # SELECT + # Species.`Name` AS species_name, + # InbredSet.`Name` AS inbredset_name, + # Tissue.`Name` AS tissue_name, + # ProbeSetFreeze.Name AS probesetfreeze_name, + # ProbeSet.Name AS probeset_name, + # ProbeSet.Symbol AS probeset_symbol, + # ProbeSet.`description` AS probeset_description, + # ProbeSet.Chr AS chr, + # ProbeSet.Mb AS mb, + # ProbeSetXRef.Mean AS mean, + # ProbeSetXRef.LRS AS lrs, + # ProbeSetXRef.`Locus` AS locus, + # ProbeSetXRef.`pValue` AS pvalue, + # ProbeSetXRef.`additive` AS additive + # FROM Species, InbredSet, ProbeSetXRef, ProbeSet, ProbeFreeze, ProbeSetFreeze, Tissue + # WHERE InbredSet.`SpeciesId`=Species.`Id` + # AND ProbeFreeze.InbredSetId=InbredSet.`Id` + # AND ProbeFreeze.`TissueId`=Tissue.`Id` + # AND ProbeSetFreeze.ProbeFreezeId=ProbeFreeze.Id + # AND ( MATCH (ProbeSet.Name,ProbeSet.description,ProbeSet.symbol,alias,GenbankId, UniGeneId, Probe_Target_Description) AGAINST ('%s' IN BOOLEAN MODE) ) + # AND ProbeSet.Id = ProbeSetXRef.ProbeSetId + # AND ProbeSetXRef.ProbeSetFreezeId=ProbeSetFreeze.Id + # AND ProbeSetFreeze.public > 0 + # ORDER BY species_name, inbredset_name, tissue_name, probesetfreeze_name, probeset_name + # LIMIT 6000 + # """ % (self.terms) + # with Bench("Running query"): + # re = g.db.execute(sql).fetchall() + # self.trait_list = [] + # with Bench("Creating trait objects"): + # for line in re: + # dataset = create_dataset(line[3], "ProbeSet", get_samplelist=False) + # trait_id = line[4] + # with Bench("Building trait object"): + # this_trait = GeneralTrait(dataset=dataset, name=trait_id, get_qtl_info=True, get_sample_info=False) + # self.trait_list.append(this_trait) - elif self.type == "phenotype": - sql = """ - SELECT - Species.`Name`, - InbredSet.`Name`, - PublishFreeze.`Name`, - PublishXRef.`Id`, - Phenotype.`Post_publication_description`, - Publication.`Authors`, - Publication.`Year`, - PublishXRef.`LRS`, - PublishXRef.`Locus`, - PublishXRef.`additive` - FROM Species,InbredSet,PublishFreeze,PublishXRef,Phenotype,Publication - WHERE PublishXRef.`InbredSetId`=InbredSet.`Id` - AND PublishFreeze.`InbredSetId`=InbredSet.`Id` - AND InbredSet.`SpeciesId`=Species.`Id` - AND PublishXRef.`PhenotypeId`=Phenotype.`Id` - AND PublishXRef.`PublicationId`=Publication.`Id` - AND (Phenotype.Post_publication_description REGEXP "[[:<:]]%s[[:>:]]" - OR Phenotype.Pre_publication_description REGEXP "[[:<:]]%s[[:>:]]" - OR Phenotype.Pre_publication_abbreviation REGEXP "[[:<:]]%s[[:>:]]" - OR Phenotype.Post_publication_abbreviation REGEXP "[[:<:]]%s[[:>:]]" - OR Phenotype.Lab_code REGEXP "[[:<:]]%s[[:>:]]" - OR Publication.PubMed_ID REGEXP "[[:<:]]%s[[:>:]]" - OR Publication.Abstract REGEXP "[[:<:]]%s[[:>:]]" - OR Publication.Title REGEXP "[[:<:]]%s[[:>:]]" - OR Publication.Authors REGEXP "[[:<:]]%s[[:>:]]" - OR PublishXRef.Id REGEXP "[[:<:]]%s[[:>:]]") - ORDER BY Species.`Name`, InbredSet.`Name`, PublishXRef.`Id` - LIMIT 6000 - """ % (self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms) - re = g.db.execute(sql).fetchall() - self.trait_list = [] - with Bench("Creating trait objects"): - for line in re: - dataset = create_dataset(line[2], "Publish") - trait_id = line[3] - this_trait = GeneralTrait(dataset=dataset, name=trait_id, get_qtl_info=True, get_sample_info=False) - self.trait_list.append(this_trait) + # elif self.type == "phenotype": + # sql = """ + # SELECT + # Species.`Name`, + # InbredSet.`Name`, + # PublishFreeze.`Name`, + # PublishXRef.`Id`, + # Phenotype.`Post_publication_description`, + # Publication.`Authors`, + # Publication.`Year`, + # PublishXRef.`LRS`, + # PublishXRef.`Locus`, + # PublishXRef.`additive` + # FROM Species,InbredSet,PublishFreeze,PublishXRef,Phenotype,Publication + # WHERE PublishXRef.`InbredSetId`=InbredSet.`Id` + # AND PublishFreeze.`InbredSetId`=InbredSet.`Id` + # AND InbredSet.`SpeciesId`=Species.`Id` + # AND PublishXRef.`PhenotypeId`=Phenotype.`Id` + # AND PublishXRef.`PublicationId`=Publication.`Id` + # AND (Phenotype.Post_publication_description REGEXP "[[:<:]]%s[[:>:]]" + # OR Phenotype.Pre_publication_description REGEXP "[[:<:]]%s[[:>:]]" + # OR Phenotype.Pre_publication_abbreviation REGEXP "[[:<:]]%s[[:>:]]" + # OR Phenotype.Post_publication_abbreviation REGEXP "[[:<:]]%s[[:>:]]" + # OR Phenotype.Lab_code REGEXP "[[:<:]]%s[[:>:]]" + # OR Publication.PubMed_ID REGEXP "[[:<:]]%s[[:>:]]" + # OR Publication.Abstract REGEXP "[[:<:]]%s[[:>:]]" + # OR Publication.Title REGEXP "[[:<:]]%s[[:>:]]" + # OR Publication.Authors REGEXP "[[:<:]]%s[[:>:]]" + # OR PublishXRef.Id REGEXP "[[:<:]]%s[[:>:]]") + # ORDER BY Species.`Name`, InbredSet.`Name`, PublishXRef.`Id` + # LIMIT 6000 + # """ % (self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms) + # re = g.db.execute(sql).fetchall() + # self.trait_list = [] + # with Bench("Creating trait objects"): + # for line in re: + # dataset = create_dataset(line[2], "Publish") + # trait_id = line[3] + # this_trait = GeneralTrait(dataset=dataset, name=trait_id, get_qtl_info=True, get_sample_info=False) + # self.trait_list.append(this_trait) diff --git a/wqflask/wqflask/heatmap/heatmap.py b/wqflask/wqflask/heatmap/heatmap.py index 2445b37f..19c330eb 100644 --- a/wqflask/wqflask/heatmap/heatmap.py +++ b/wqflask/wqflask/heatmap/heatmap.py @@ -136,7 +136,7 @@ class Heatmap(object): this_trait = trait_db[0] #this_db = trait_db[1] genotype = self.dataset.group.read_genotype_file() - samples, values, variances = this_trait.export_informative() + samples, values, variances, sample_aliases = this_trait.export_informative() trimmed_samples = [] trimmed_values = [] diff --git a/wqflask/wqflask/marker_regression/marker_regression.py b/wqflask/wqflask/marker_regression/marker_regression.py index 08f422f0..1e0a618e 100644 --- a/wqflask/wqflask/marker_regression/marker_regression.py +++ b/wqflask/wqflask/marker_regression/marker_regression.py @@ -56,11 +56,21 @@ class MarkerRegression(object): self.samples = [] # Want only ones with values self.vals = [] - + + #for sample in self.this_trait.data.keys(): for sample in self.dataset.group.samplelist: - value = start_vars['value:' + sample] - self.samples.append(str(sample)) - self.vals.append(value) + in_trait_data = False + for item in self.this_trait.data: + if self.this_trait.data[item].name2 == 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['value:' + sample] + self.samples.append(sample) + self.vals.append(value) self.mapping_method = start_vars['method'] if start_vars['manhattan_plot'] == "True": @@ -203,6 +213,8 @@ class MarkerRegression(object): if 'lod_score' in marker.keys(): self.qtl_results.append(marker) + self.trimmed_markers = trim_markers_for_table(results) + for qtl in enumerate(self.qtl_results): self.json_data['chr1'].append(str(qtl['chr1'])) self.json_data['chr2'].append(str(qtl['chr2'])) @@ -641,15 +653,17 @@ class MarkerRegression(object): if self.manhattan_plot != True: genotype = genotype.addinterval() - samples, values, variances = self.this_trait.export_informative() - + samples, values, variances, sample_aliases = self.this_trait.export_informative() + trimmed_samples = [] trimmed_values = [] for i in range(0, len(samples)): - if samples[i] in self.dataset.group.samplelist: - trimmed_samples.append(samples[i]) + if self.this_trait.data[samples[i]].name2 in self.dataset.group.samplelist: + trimmed_samples.append(sample_aliases[i]) trimmed_values.append(values[i]) + #print("THE SAMPLES:", trimmed_samples) + if self.num_perm < 100: self.suggestive = 0 self.significant = 0 diff --git a/wqflask/wqflask/show_trait/show_trait.py b/wqflask/wqflask/show_trait/show_trait.py index 074c78bf..f7a33d4f 100755..100644 --- a/wqflask/wqflask/show_trait/show_trait.py +++ b/wqflask/wqflask/show_trait/show_trait.py @@ -1184,17 +1184,16 @@ class ShowTrait(object): all_samples_ordered = self.dataset.group.all_samples_ordered() primary_sample_names = list(all_samples_ordered) - - print("self.dataset.group", pf(self.dataset.group.__dict__)) - print("-*- primary_samplelist is:", pf(primary_sample_names)) - + other_sample_names = [] for sample in this_trait.data.keys(): - if sample not in all_samples_ordered: + if (this_trait.data[sample].name2 in primary_sample_names) and (this_trait.data[sample].name not in primary_sample_names): + primary_sample_names.append(this_trait.data[sample].name) + primary_sample_names.remove(this_trait.data[sample].name2) + elif sample not in all_samples_ordered: all_samples_ordered.append(sample) other_sample_names.append(sample) - print("species:", self.dataset.group.species) if self.dataset.group.species == "human": primary_sample_names += other_sample_names diff --git a/wqflask/wqflask/static/new/javascript/dataset_menu_structure.json b/wqflask/wqflask/static/new/javascript/dataset_menu_structure.json index 2b16383f..12a30e84 100755 --- a/wqflask/wqflask/static/new/javascript/dataset_menu_structure.json +++ b/wqflask/wqflask/static/new/javascript/dataset_menu_structure.json @@ -1596,6 +1596,11 @@ "B6D2F2": { "Brain mRNA": [ [ + "76", + "BRF2_M_0805_M", + "OHSU/VA B6D2F2 Brain mRNA M430 (Aug05) MAS5" + ], + [ "78", "BRF2_M_0805_P", "OHSU/VA B6D2F2 Brain mRNA M430 (Aug05) PDNN" @@ -1606,11 +1611,6 @@ "OHSU/VA B6D2F2 Brain mRNA M430 (Aug05) RMA" ], [ - "76", - "BRF2_M_0805_M", - "OHSU/VA B6D2F2 Brain mRNA M430 (Aug05) MAS5" - ], - [ "33", "BRF2_M_0304_P", "OHSU/VA B6D2F2 Brain mRNA M430A (Mar04) PDNN" @@ -1846,6 +1846,18 @@ ] }, "BXD": { + "Adipose Proteome": [ + [ + "797", + "EPFL_AdiPro0416", + "EPFL/ETHZ BXD Brown Adipose, Total Tissue Proteome, Chow Diet (Apr16) Light SWATH **" + ], + [ + "798", + "EPFL_AdiMitPro0416", + "EPFL/ETHZ BXD Brown Adipose, Isolated Mitochondria Proteome, Chow Diet (Apr16) Light SWATH **" + ] + ], "Adipose mRNA": [ [ "469", @@ -1916,21 +1928,21 @@ [ "414", "UCLA_BXD-on_Femur_0113_RSN", - "UCLA GSE27483 BXD Only Bone Femur ILM Mouse WG-6 v2.0 (Jan13) RSN" + "UCLA GSE27483 BXD Only Bone Femur ILM Mouse WG-6 v1.1 (Jan13) RSN" ] ], "Brain mRNA": [ [ - "164", - "UTHSC_BXD_WB_RNASeq1112", - "UTHSC Mouse BXD Whole Brain RNA Sequence (Nov12) RPKM Untrimmed" - ], - [ "590", "UTHSC_BXD_WB_RNASeqtrim1_1112", "UTHSC Mouse BXD Whole Brain RNA Sequence (Nov12) RPKM Trimmed 1.0" ], [ + "164", + "UTHSC_BXD_WB_RNASeq1112", + "UTHSC Mouse BXD Whole Brain RNA Sequence (Nov12) RPKM Untrimmed" + ], + [ "394", "UTHSC_BXD_WB_RNASeqEx1112", "UTHSC Mouse BXD Whole Brain RNA Sequence Exon Level (Nov12) RPKM" @@ -1946,6 +1958,11 @@ "UTHSC Brain mRNA U74Av2 (Nov05) PDNN" ], [ + "82", + "BR_U_0805_R", + "UTHSC Brain mRNA U74Av2 (Aug05) RMA" + ], + [ "81", "BR_U_0805_P", "UTHSC Brain mRNA U74Av2 (Aug05) PDNN" @@ -1956,11 +1973,6 @@ "UTHSC Brain mRNA U74Av2 (Aug05) MAS5" ], [ - "82", - "BR_U_0805_R", - "UTHSC Brain mRNA U74Av2 (Aug05) RMA" - ], - [ "42", "CB_M_0204_P", "INIA Brain mRNA M430 (Feb04) PDNN" @@ -2027,6 +2039,11 @@ "Eye M430v2 WT Gpnmb (Sep08) RMA" ], [ + "278", + "Eye_M2_0908_R_MT", + "Eye M430v2 Mutant Tyrp1 (Sep08) RMA" + ], + [ "382", "Eye_M2_0908_WTWT", "Eye M430v2 WT WT (Sep08) RMA" @@ -2037,11 +2054,6 @@ "Eye M430v2 WT Tyrp1 (Sep08) RMA" ], [ - "278", - "Eye_M2_0908_R_MT", - "Eye M430v2 Mutant Tyrp1 (Sep08) RMA" - ], - [ "400", "DBA2J-ONH-1212", "Howell et al. 2011, DBA/2J Glaucoma Optic Nerve Head M430 2.0 (Dec12) RMA" @@ -2557,29 +2569,29 @@ ], "Retina mRNA": [ [ + "267", + "Illum_Retina_BXD_RankInv0410", + "Full HEI Retina Illumina V6.2 (Apr10) RankInv" + ], + [ + "302", + "G2NEI_ILM_Retina_BXD_RI0410", + "HEI Retina Normal Illumina V6.2 (Apr10) RankInv" + ], + [ "709", "DoDCMMRPRetMoGene2_0515", - "DoD CDMRP Retina Affy MoGene 2.0 ST (May15) RMA Gene Level" + "DoD Retina Normal Affy MoGene 2.0 ST (May15) RMA Gene Level" ], [ "710", "DoDCMMRPRetMoGene2Ex_0515", - "DoD CDMRP Retina Affy MoGene 2.0 ST (May15) RMA Exon Level" + "DoD Retina Normal Affy MoGene 2.0 ST (May15) RMA Exon Level" ], [ "385", "ONCRetILM6_0412", "ONC HEI Retina (April 2012) RankInv" - ], - [ - "302", - "G2NEI_ILM_Retina_BXD_RI0410", - "Normal HEI Retina (April 2010) RankInv" - ], - [ - "267", - "Illum_Retina_BXD_RankInv0410", - "Full HEI Retina (April 2010) RankInv" ] ], "Spleen mRNA": [ @@ -2704,6 +2716,23 @@ "RTC_1106_R", "HZI Treg M430v2 (Feb11) RMA" ] + ], + "Ventral Tegmental Area mRNA": [ + [ + "228", + "VCUSal_0609_R", + "VCU BXD VTA Sal M430 2.0 (Jun09) RMA" + ], + [ + "230", + "VCUEtvsSal_0609_R", + "VCU BXD VTA Et vs Sal M430 2.0 (Jun09) RMA" + ], + [ + "229", + "VCUEtOH_0609_R", + "VCU BXD VTA EtOH M430 2.0 (Jun09) RMA" + ] ] }, "BXH": { @@ -2810,6 +2839,16 @@ "170", "UCLA_CTB6B6CTF2_ADIPOSE_2005", "UCLA CTB6/B6CTF2 Adipose (2005) mlratio" + ], + [ + "189", + "UCLA_CTB6B6CTF2_ADIPOSE_FEMALE", + "UCLA CTB6B6CTF2 Adipose Female mlratio" + ], + [ + "188", + "UCLA_CTB6B6CTF2_ADIPOSE_MALE", + "UCLA CTB6B6CTF2 Adipose Male mlratio" ] ], "Brain mRNA": [ @@ -2841,6 +2880,16 @@ "172", "UCLA_CTB6B6CTF2_LIVER_2005", "UCLA CTB6/B6CTF2 Liver (2005) mlratio" + ], + [ + "193", + "UCLA_CTB6B6CTF2_LIVER_FEMALE", + "UCLA CTB6B6CTF2 Liver Female mlratio" + ], + [ + "192", + "UCLA_CTB6B6CTF2_LIVER_MALE", + "UCLA CTB6B6CTF2 Liver Male mlratio" ] ], "Muscle mRNA": [ @@ -2848,6 +2897,16 @@ "173", "UCLA_CTB6B6CTF2_MUSCLE_2005", "UCLA CTB6/B6CTF2 Muscle (2005) mlratio" + ], + [ + "195", + "UCLA_CTB6B6CTF2_MUSCLE_FEMALE", + "UCLA CTB6B6CTF2 Muscle Female mlratio" + ], + [ + "194", + "UCLA_CTB6B6CTF2_MUSCLE_MALE", + "UCLA CTB6B6CTF2 Muscle Male mlratio" ] ], "Phenotypes": [ @@ -2942,6 +3001,11 @@ ], "Hippocampus mRNA": [ [ + "212", + "Illum_LXS_Hipp_RSE_1008", + "Hippocampus Illumina RSE (Oct08) RankInv beta" + ], + [ "214", "Illum_LXS_Hipp_NOE_1008", "Hippocampus Illumina NOE (Oct08) RankInv beta" @@ -2962,11 +3026,6 @@ "Hippocampus Illumina NON (Oct08) RankInv beta" ], [ - "212", - "Illum_LXS_Hipp_RSE_1008", - "Hippocampus Illumina RSE (Oct08) RankInv beta" - ], - [ "143", "Illum_LXS_Hipp_loess0807", "Hippocampus Illumina (Aug07) LOESS" @@ -3130,16 +3189,32 @@ "Scripps-2013": {} }, "rat": { - "HSNIH": { + "HSNIH-Palmer": { "Phenotypes": [ [ - "619", - "HSNIHPublish", - "HSNIH Published Phenotypes" + "None", + "HSNIH-PalmerPublish", + "HSNIH-Palmer Published Phenotypes" + ] + ] + }, + "HSNIH-RGSMC": { + "Phenotypes": [ + [ + "None", + "HSNIH-RGSMCPublish", + "HSNIH-RGSMC Published Phenotypes" ] ] }, "HXBBXH": { + "Adipose mRNA": [ + [ + "799", + "FGUCAS_BAdip0516", + "FGUCAS BXH/HXB Brown Adipose Affy Rat Gene 2.0 ST (May16) log2 **" + ] + ], "Adrenal Gland mRNA": [ [ "220", @@ -3460,8 +3535,12 @@ ], "rat": [ [ - "HSNIH", - "NIH Heterogeneous Stock" + "HSNIH-Palmer", + "NIH Heterogeneous Stock (Palmer)" + ], + [ + "HSNIH-RGSMC", + "NIH Heterogeneous Stock (RGSMC 2013)" ], [ "HXBBXH", @@ -4351,6 +4430,10 @@ "Adipose mRNA" ], [ + "Adipose Proteome", + "Adipose Proteome" + ], + [ "Adrenal Gland mRNA", "Adrenal Gland mRNA" ], @@ -4465,6 +4548,10 @@ [ "T Cell (regulatory) mRNA", "T Cell (regulatory) mRNA" + ], + [ + "Ventral Tegmental Area mRNA", + "Ventral Tegmental Area mRNA" ] ], "BXH": [ @@ -4656,7 +4743,13 @@ "Scripps-2013": [] }, "rat": { - "HSNIH": [ + "HSNIH-Palmer": [ + [ + "Phenotypes", + "Phenotypes" + ] + ], + "HSNIH-RGSMC": [ [ "Phenotypes", "Phenotypes" @@ -4672,6 +4765,10 @@ "Genotypes" ], [ + "Adipose mRNA", + "Adipose mRNA" + ], + [ "Adrenal Gland mRNA", "Adrenal Gland mRNA" ], diff --git a/wqflask/wqflask/static/new/javascript/show_trait.js b/wqflask/wqflask/static/new/javascript/show_trait.js index 2fa77ae0..34d1a139 100644 --- a/wqflask/wqflask/static/new/javascript/show_trait.js +++ b/wqflask/wqflask/static/new/javascript/show_trait.js @@ -50,7 +50,7 @@ }, { vn: "interquartile", pretty: "Interquartile Range", - url: "/glossary.html#Interquartile", + url: "http://www.genenetwork.org/glossary.html#Interquartile", digits: 2 } ]; @@ -313,6 +313,12 @@ return $("#trait_data_form").submit(); }; + submit_corr = function(){ + var url; + url = "/corr_compute"; + return submit_special(url); + }; + $(".corr_compute").on("click", (function(_this) { return function() { var url; diff --git a/wqflask/wqflask/templates/gsearch_gene.html b/wqflask/wqflask/templates/gsearch_gene.html index 7cc9a1bd..92b0b411 100755 --- a/wqflask/wqflask/templates/gsearch_gene.html +++ b/wqflask/wqflask/templates/gsearch_gene.html @@ -2,10 +2,6 @@ {% block title %}Search Results{% endblock %} {% block css %} <link rel="stylesheet" type="text/css" href="/static/new/packages/DataTables/css/jquery.dataTables.css" /> - <link rel="stylesheet" type="text/css" href="/static/packages/DT_bootstrap/DT_bootstrap.css" /> - <link rel="stylesheet" type="text/css" href="/static/packages/TableTools/media/css/TableTools.css" /> - <link rel="stylesheet" type="text/css" href="/static/new/packages/DataTables/extensions/dataTables.fixedHeader.css" > - <link rel="stylesheet" type="text/css" href="//cdn.datatables.net/fixedcolumns/3.0.4/css/dataTables.fixedColumns.css"> <link rel="stylesheet" type="text/css" href="/static/new/packages/DataTables/extensions/buttons.bootstrap.css" /> {% endblock %} {% block content %} @@ -13,6 +9,7 @@ <div class="container"> + <p>You searched for {{ terms }}.</p> <p>To study a record, click on its ID below.<br />Check records below and click Add button to add to selection.</p> <div> @@ -26,46 +23,28 @@ <br /> <br /> - <table width="2000px" class="table table-hover table-striped" id="trait_table"> + <div style="width: 2000px;"> + <table width="2000px" id="trait_table" class="table table-hover table-striped" > <thead> - <tr> - <th style="width: 30px;"></th> - <th>Index</th> - <th>Species</th> - <th>Group</th> - <th>Tissue</th> - <th>Dataset</th> - <th>Record</th> - <th>Symbol</th> - <th>Description</th> - <th>Location</th> - <th>Mean</th> - <th style="text-align: right;">Max <br>LRS<a href="http://genenetwork.org//glossary.html#L" target="_blank"><sup style="color:#f00"> ?</sup></a></th> - <th>Max LRS Location</th> - <th style="text-align: right;">Additive<br>Effect<a href="http://genenetwork.org//glossary.html#A" target="_blank"><sup style="color:#f00"> ?</sup></a></th> - </tr> - </thead> - <tbody> - {% for this_trait in trait_list %} - <TR id="trait:{{ this_trait.name }}:{{ this_trait.dataset.name }}"> - <TD><INPUT TYPE="checkbox" NAME="searchResult" class="checkbox trait_checkbox" style="transform: scale(1.5);" VALUE="{{ data_hmac('{}:{}'.format(this_trait.name, this_trait.dataset.name)) }}"></TD> - <TD>{{ loop.index }}</TD> - <TD>{{ this_trait.dataset.group.species }}</TD> - <TD>{{ this_trait.dataset.group.name }}</TD> - <TD>{{ this_trait.dataset.tissue }}</TD> - <TD>{{ this_trait.dataset.fullname }}</TD> - <TD><a href="{{ url_for('show_trait_page', trait_id = this_trait.name, dataset = this_trait.dataset.name)}}">{{ this_trait.name }}</a></TD> - <TD>{{ this_trait.symbol }}</TD> - <TD>{{ this_trait.description_display }}</TD> - <TD>{{ this_trait.location_repr }}</TD> - <TD align="right">{{ '%0.3f' % this_trait.mean|float }}</TD> - <TD align="right">{{ '%0.3f' % this_trait.LRS_score_repr|float }}</TD> - <TD>{{ this_trait.LRS_location_repr }}</TD> - <TD align="right">{{ '%0.3f' % this_trait.additive|float }}</TD> - </TR> - {% endfor %} - </tbody> - </table> + <tr> + <th></th> + <th>Index</th> + <th>Species</th> + <th>Group</th> + <th>Tissue</th> + <th>Dataset</th> + <th>Record</th> + <th>Symbol</th> + <th>Description</th> + <th>Location</th> + <th>Mean</th> + <th>Max<br>LRS<a href="http://genenetwork.org/glossary.html#L" target="_blank"><sup style="color:#f00"> ?</sup></a></th> + <th>Max LRS Location</th> + <th>Additive<br>Effect<a href="http://genenetwork.org/glossary.html#A" target="_blank"><sup style="color:#f00"> ?</sup></a></th> + </tr> + </thead> + </table> + </div> </div> </div> @@ -156,23 +135,18 @@ console.time("Creating table"); $('#trait_table').DataTable( { - "columns": [ - { "type": "natural" }, - { "type": "natural" }, - { "type": "natural" }, - { "type": "cust-txt" }, - { "type": "natural" }, - { "type": "natural" }, - { "type": "natural" }, - { "type": "natural" }, - { "type": "natural", "width": "15%" }, - { "type": "natural" }, - { "type": "natural" }, - { "type": "natural" }, - { "type": "natural" }, - { "type": "cust-txt" } - ], - "order": [[ 1, "asc" ]], + "processing": true, + "serverSide": true, + "paging": false, + "ajax": { + "url": "/gsearch_updating?terms={{ terms }}&type={{ type }}", + "type": "POST", + "dataType": "json", + "contentType": "application/json; charset=utf-8", + "data": function ( args ) { + return { "args": JSON.stringify( args ) }; + } + }, "buttons": [ { extend: 'csvHtml5', @@ -184,15 +158,26 @@ } } ], - "sDom": "RZBtir", + "columns": [ + { "data": "checkbox", "orderable" : false }, + { "data": "index", "orderable" : true }, + { "data": "species", "orderable" : true }, + { "data": "group", "orderable" : true }, + { "data": "tissue", "orderable" : true }, + { "data": "dataset", "orderable" : true }, + { "data": "record", "orderable" : true }, + { "data": "symbol", "orderable" : true }, + { "data": "description", "orderable" : true }, + { "data": "location", "orderable" : true }, + { "data": "mean", "orderable" : true }, + { "data": "max_lrs", "orderable" : true }, + { "data": "max_lrs_location", "orderable" : true }, + { "data": "additive_effect", "orderable" : true } + ], + "sDom": "Bfrti", "autoWidth": false, - "bLengthChange": true, - "bDeferRender": true, - "scrollCollapse": false, - "colResize": { - "tableWidthFixed": false, - }, - "paging": false + "scrollY": "800px", + "bDeferRender": true } ); console.timeEnd("Creating table"); diff --git a/wqflask/wqflask/templates/show_trait.html b/wqflask/wqflask/templates/show_trait.html index 64638fc7..5e2dc6fa 100755 --- a/wqflask/wqflask/templates/show_trait.html +++ b/wqflask/wqflask/templates/show_trait.html @@ -29,7 +29,7 @@ <h3>{{ this_trait.description_fmt }}</h3> </div> - <form method="post" target="_blank" action="/corr_compute" name="trait_page" id="trait_data_form" + <form method="post" action="/corr_compute" target="_blank" name="trait_page" id="trait_data_form" class="form-horizontal"> <div id="hidden_inputs"> <input type="hidden" name="trait_hmac" value="{{ data_hmac('{}:{}'.format(this_trait.name, dataset.name)) }}"> diff --git a/wqflask/wqflask/templates/show_trait_calculate_correlations.html b/wqflask/wqflask/templates/show_trait_calculate_correlations.html index 80fafa5e..0e15ce9c 100755 --- a/wqflask/wqflask/templates/show_trait_calculate_correlations.html +++ b/wqflask/wqflask/templates/show_trait_calculate_correlations.html @@ -90,12 +90,15 @@ </div> <div class="form-group"> <label class="col-xs-1 control-label">Range</label> - <div class="col-xs-3 controls"> + <div class="col-xs-4 controls"> <input name="p_range_lower" value="" type="hidden"> <input name="p_range_upper" value="" type="hidden"> - <div id="p_range_slider" ></div> - <span style="font: 400 12px Arial; color: #888; display: block; margin: 15px 0;" id="p_range_lower"></span> - <span style="font: 400 12px Arial; color: #888; display: block; margin: 15px 0;" id="p_range_upper"></span> + <span style="display: inline;"> + <div id="p_range_slider" style="width: 200px;"></div> + <span style="font: 400 12px Arial; color: #888; display: inline; margin: 25px 0; width: 20px;" id="p_range_lower"></span> + <span> </span> + <span style="font: 400 12px Arial; color: #888; display: inline; margin: 15px 0; width: 20px;" id="p_range_upper"></span> + </span> </div> </div> diff --git a/wqflask/wqflask/templates/show_trait_details.html b/wqflask/wqflask/templates/show_trait_details.html index 95a3b967..d5fb0cf2 100755 --- a/wqflask/wqflask/templates/show_trait_details.html +++ b/wqflask/wqflask/templates/show_trait_details.html @@ -35,8 +35,7 @@ <tr> <td>Target Score</td> <td> - <a href="/blatInfo.html" target="_blank" - title="Values higher than 2 for the specificity are good"> + <a href="http://genenetwork.org/blatInfo.html" title="Values higher than 2 for the specificity are good"> BLAT Specificity </a>: {{ "%0.3f" | format(this_trait.probe_set_specificity|float) }} @@ -51,25 +50,25 @@ <td>Resource Links</td> <td> {% if this_trait.geneid != None %} - <a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=gene&cmd=Retrieve&dopt=Graphics&list_uids={{ this_trait.geneid }}" target="_blank" title="Info from NCBI Entrez Gene"> + <a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=gene&cmd=Retrieve&dopt=Graphics&list_uids={{ this_trait.geneid }}" title="Info from NCBI Entrez Gene"> Gene </a> {% endif %} {% if this_trait.omim != None %} - <a href="http://www.ncbi.nlm.nih.gov/omim/{{ this_trait.omim }}" target="_blank" title="Summary from On Mendelion Inheritance in Man"> + <a href="http://www.ncbi.nlm.nih.gov/omim/{{ this_trait.omim }}" title="Summary from On Mendelion Inheritance in Man"> OMIM </a> {% endif %} {% if this_trait.genbankid != None %} - <a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Nucleotide&cmd=search&doptcmdl=DocSum&term={{ this_trait.genbankid }}" target="_blank" title="Find the original GenBank sequence used to design the probes"> + <a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Nucleotide&cmd=search&doptcmdl=DocSum&term={{ this_trait.genbankid }}" title="Find the original GenBank sequence used to design the probes"> GenBank </a> {% endif %} {% if this_trait.symbol != None %} - <a href="http://www.genotation.org/Getd2g.pl?gene_list={{ this_trait.symbol }}" target="_blank" title="Related descriptive, genomic, clinical, functional and drug-therapy information"> + <a href="http://www.genotation.org/Getd2g.pl?gene_list={{ this_trait.symbol }}" title="Related descriptive, genomic, clinical, functional and drug-therapy information"> Genotation </a> @@ -87,40 +86,40 @@ </a> {% if this_trait.dataset.type == 'ProbeSet' %} {% if this_trait.symbol != None %} - <a href="#redirect" onclick="window.open('http://www.genenetwork.org/webqtl/main.py?cmd=sch&gene={{ this_trait.symbol }}&alias=1&species={{ species_name }}')"> + <a href="http://www.genenetwork.org/webqtl/main.py?cmd=sch&gene={{ this_trait.symbol }}&alias=1&species={{ species_name }}"> <button type="button" class="btn btn-default" title="Find similar expression data"> <i class="icon-search"></i> Find </button> </a> {% endif %} {% if UCSC_BLAT_URL != "" %} - <a href="#redirect" onclick="window.open('{{ UCSC_BLAT_URL }}')"> + <a href="{{ UCSC_BLAT_URL }}"> <button type="button" class="btn btn-default" title="Check probe locations at UCSC"> <i class="icon-ok"></i> Verify </button> </a> {% endif %} {% if this_trait.symbol != None %} - <a href="#redirect" onclick="window.open('http://genenetwork.org/webqtl/main.py?FormID=geneWiki&symbol={{ this_trait.symbol }}')"> + <a href="http://genenetwork.org/webqtl/main.py?FormID=geneWiki&symbol={{ this_trait.symbol }}"> <button type="button" class="btn btn-default" title="Write or review comments about this gene"> <i class="icon-edit"></i> GeneWiki </button> </a> - <a href="#redirect" onclick="window.open('http://genenetwork.org/webqtl/main.py?FormID=SnpBrowserResultPage&submitStatus=1&diffAlleles=True&customStrain=True&geneName={{ this_trait.symbol }}')"> + <a href="http://genenetwork.org/webqtl/main.py?FormID=SnpBrowserResultPage&submitStatus=1&diffAlleles=True&customStrain=True&geneName={{ this_trait.symbol }}"> <button type="button" class="btn btn-default" title="View SNPs and Indels"> <i class="icon-road"></i> SNPs </button> </a> {% endif %} {% if UTHSC_BLAT_URL != "" %} - <a href="#redirect" onclick="window.open('{{ UTHSC_BLAT_URL }}')"> + <a href="{{ UTHSC_BLAT_URL }}"> <button type="button" class="btn btn-default" title="View probes, SNPs, and RNA-seq at UTHSC"> <i class="icon-eye-close"></i> RNA-seq </button> </a> {% endif %} {% if show_probes == "True" %} - <a href="#redirect" onclick="window.open('http://genenetwork.org/webqtl/main.py?FormID=showProbeInfo&database={{ this_trait.dataset.name }}&ProbeSetID={{ this_trait.name }}&CellID={{ this_trait.cellid }}&RISet={{ dataset.group.name }}&incparentsf1=ON')"> + <a href="http://genenetwork.org/webqtl/main.py?FormID=showProbeInfo&database={{ this_trait.dataset.name }}&ProbeSetID={{ this_trait.name }}&CellID={{ this_trait.cellid }}&RISet={{ dataset.group.name }}&incparentsf1=ON"> <button type="button" class="btn btn-default" title="Check sequence of probes"> <i class="icon-list"></i> Probes </button> diff --git a/wqflask/wqflask/templates/show_trait_mapping_tools.html b/wqflask/wqflask/templates/show_trait_mapping_tools.html index 067dfc67..3d9c2521 100755 --- a/wqflask/wqflask/templates/show_trait_mapping_tools.html +++ b/wqflask/wqflask/templates/show_trait_mapping_tools.html @@ -6,13 +6,13 @@ <ul class="nav nav-pills"> {% if use_pylmm_rqtl and not use_plink_gemma and dataset.group.species != "human" %} <li class="active"> - <a href="#pylmm" data-toggle="tab">pyLMM</a> + <a href="#interval_mapping" data-toggle="tab">Interval Mapping</a> </li> <li> - <a href="#rqtl_geno" data-toggle="tab">R/qtl</a> + <a href="#pylmm" data-toggle="tab">pyLMM</a> </li> <li> - <a href="#interval_mapping" data-toggle="tab">Interval Mapping</a> + <a href="#rqtl_geno" data-toggle="tab">R/qtl</a> </li> {% endif %} {% if use_plink_gemma %} @@ -30,7 +30,89 @@ <div class="tab-content"> {% if use_pylmm_rqtl and not use_plink_gemma and dataset.group.species != "human" %} - <div class="tab-pane active" id="pylmm"> + <div class="tab-pane active" id="interval_mapping"> + <div style="margin-top: 20px" class="form-horizontal"> + <div class="mapping_method_fields form-group"> + <label for="mapping_permutations" class="col-xs-3 control-label">Permutations</label> + <div style="margin-left: 20px;" class="col-xs-4 controls"> + <input name="num_perm_reaper" value="2000" type="text" class="form-control"> + </div> + </div> + <div class="mapping_method_fields form-group"> + <label for="mapping_bootstraps" class="col-xs-3 control-label">Bootstraps</label> + <div style="margin-left: 20px;" class="col-xs-4 controls"> + <input name="num_bootstrap" value="2000" type="text" class="form-control"> + </div> + </div> + <div class="mapping_method_fields form-group"> + <label for="control_for" class="col-xs-3 control-label">Control for</label> + <div style="margin-left: 20px;" class="col-xs-4 controls"> + {% if dataset.type == 'ProbeSet' and this_trait.locus_chr != "" %} + <input name="control_reaper" value="{{ nearest_marker }}" type="text" style="width: 160px;" class="form-control" /> + {% else %} + <input name="control_reaper" value="" type="text" class="form-control" /> + {% endif %} + <label class="radio-inline"> + <input type="radio" name="do_control_reaper" value="true"> + Yes + </label> + <label class="radio-inline"> + <input type="radio" name="do_control_reaper" value="false" checked=""> + No + </label> + </div> + </div> + +<!-- + <div class="mapping_method_fields form-group"> + <label for="mapping_bootstraps" class="col-xs-3 control-label" title="Bootstrapping Resamples">Bootstrap Test (n=2000)</label> + <div class="col-xs-4 controls"> + <label> + <input type="checkbox" name="bootCheck" id="bootCheck"> Bootstrap Test (n=2000) + </label> + </div> + </div> + + <div class="mapping_method_fields form-group"> + <label style="text-align:left;" class="col-xs-12 control-label">Display Additive Effect</label> + <div class="col-xs-12 controls" id="display_additive_effect"> + <label class="radio-inline"> + <input type="radio" name="display_additive" id="display_additive" value="yes" checked=""> + Yes + </label> + <label class="radio-inline"> + <input type="radio" name="display_additive" id="display_additive" value="no"> + No + </label> + </div> + </div> +--> + + + <div class="mapping_method_fields form-group"> + <label style="text-align:left;" class="col-xs-12 control-label">Marker Regr.</label> + <div class="col-xs-12 controls"> + <label class="radio-inline"> + <input type="radio" name="manhattan_plot_reaper" value="True"> + Yes + </label> + <label class="radio-inline"> + <input type="radio" name="manhattan_plot_reaper" value="False" checked=""> + No + </label> + </div> + </div> + <div class="form-group"> + <div style="padding-left:15px;" class="controls"> + <button id="interval_mapping_compute" class="btn submit_special btn-primary" data-url="/marker_regression" title="Compute Interval Mapping"> + <i class="icon-ok-circle icon-white"></i> Compute + </button> + </div> + </div> + <!--<div id="alert_placeholder"></div>--> + </div> + </div> + <div class="tab-pane" id="pylmm"> <div style="margin-top: 20px" class="form-horizontal"> <div class="mapping_method_fields form-group"> <label for="mapping_permutations" class="col-xs-3 control-label">Permutations</label> @@ -179,88 +261,6 @@ </div> </div> </div> - <div class="tab-pane" id="interval_mapping"> - <div style="margin-top: 20px" class="form-horizontal"> - <div class="mapping_method_fields form-group"> - <label for="mapping_permutations" class="col-xs-3 control-label">Permutations</label> - <div style="margin-left: 20px;" class="col-xs-4 controls"> - <input name="num_perm_reaper" value="2000" type="text" class="form-control"> - </div> - </div> - <div class="mapping_method_fields form-group"> - <label for="mapping_bootstraps" class="col-xs-3 control-label">Bootstraps</label> - <div style="margin-left: 20px;" class="col-xs-4 controls"> - <input name="num_bootstrap" value="2000" type="text" class="form-control"> - </div> - </div> - <div class="mapping_method_fields form-group"> - <label for="control_for" class="col-xs-3 control-label">Control for</label> - <div style="margin-left: 20px;" class="col-xs-4 controls"> - {% if dataset.type == 'ProbeSet' and this_trait.locus_chr != "" %} - <input name="control_reaper" value="{{ nearest_marker }}" type="text" style="width: 160px;" class="form-control" /> - {% else %} - <input name="control_reaper" value="" type="text" class="form-control" /> - {% endif %} - <label class="radio-inline"> - <input type="radio" name="do_control_reaper" value="true"> - Yes - </label> - <label class="radio-inline"> - <input type="radio" name="do_control_reaper" value="false" checked=""> - No - </label> - </div> - </div> - -<!-- - <div class="mapping_method_fields form-group"> - <label for="mapping_bootstraps" class="col-xs-3 control-label" title="Bootstrapping Resamples">Bootstrap Test (n=2000)</label> - <div class="col-xs-4 controls"> - <label> - <input type="checkbox" name="bootCheck" id="bootCheck"> Bootstrap Test (n=2000) - </label> - </div> - </div> - - <div class="mapping_method_fields form-group"> - <label style="text-align:left;" class="col-xs-12 control-label">Display Additive Effect</label> - <div class="col-xs-12 controls" id="display_additive_effect"> - <label class="radio-inline"> - <input type="radio" name="display_additive" id="display_additive" value="yes" checked=""> - Yes - </label> - <label class="radio-inline"> - <input type="radio" name="display_additive" id="display_additive" value="no"> - No - </label> - </div> - </div> ---> - - - <div class="mapping_method_fields form-group"> - <label style="text-align:left;" class="col-xs-12 control-label">Marker Regr.</label> - <div class="col-xs-12 controls"> - <label class="radio-inline"> - <input type="radio" name="manhattan_plot_reaper" value="True"> - Yes - </label> - <label class="radio-inline"> - <input type="radio" name="manhattan_plot_reaper" value="False" checked=""> - No - </label> - </div> - </div> - <div class="form-group"> - <div style="padding-left:15px;" class="controls"> - <button id="interval_mapping_compute" class="btn submit_special btn-primary" data-url="/marker_regression" title="Compute Interval Mapping"> - <i class="icon-ok-circle icon-white"></i> Compute - </button> - </div> - </div> - <!--<div id="alert_placeholder"></div>--> - </div> - </div> {% endif %} {% if use_plink_gemma %} <div class="tab-pane" id="plink"> diff --git a/wqflask/wqflask/update_search_results.py b/wqflask/wqflask/update_search_results.py new file mode 100644 index 00000000..ffd7fd51 --- /dev/null +++ b/wqflask/wqflask/update_search_results.py @@ -0,0 +1,129 @@ +from __future__ import absolute_import, print_function, division + +import json + +from flask import Flask, g +from base.data_set import create_dataset +from base.trait import GeneralTrait +from dbFunction import webqtlDatabaseFunction + +from utility.benchmark import Bench + +class GSearch(object): + + def __init__(self, kw): + self.type = kw['type'] + self.terms = kw['terms'] + #self.row_range = kw['row_range'] + if self.type == "gene": + sql = """ + SELECT + Species.`Name` AS species_name, + InbredSet.`Name` AS inbredset_name, + Tissue.`Name` AS tissue_name, + ProbeSetFreeze.Name AS probesetfreeze_name, + ProbeSet.Name AS probeset_name, + ProbeSet.Symbol AS probeset_symbol, + ProbeSet.`description` AS probeset_description, + ProbeSet.Chr AS chr, + ProbeSet.Mb AS mb, + ProbeSetXRef.Mean AS mean, + ProbeSetXRef.LRS AS lrs, + ProbeSetXRef.`Locus` AS locus, + ProbeSetXRef.`pValue` AS pvalue, + ProbeSetXRef.`additive` AS additive + FROM Species, InbredSet, ProbeSetXRef, ProbeSet, ProbeFreeze, ProbeSetFreeze, Tissue + WHERE InbredSet.`SpeciesId`=Species.`Id` + AND ProbeFreeze.InbredSetId=InbredSet.`Id` + AND ProbeFreeze.`TissueId`=Tissue.`Id` + AND ProbeSetFreeze.ProbeFreezeId=ProbeFreeze.Id + AND ( MATCH (ProbeSet.Name,ProbeSet.description,ProbeSet.symbol,alias,GenbankId, UniGeneId, Probe_Target_Description) AGAINST ('%s' IN BOOLEAN MODE) ) + AND ProbeSet.Id = ProbeSetXRef.ProbeSetId + AND ProbeSetXRef.ProbeSetFreezeId=ProbeSetFreeze.Id + AND ProbeSetFreeze.public > 0 + ORDER BY species_name, inbredset_name, tissue_name, probesetfreeze_name, probeset_name + LIMIT 6000 + """ % (self.terms) + with Bench("Running query"): + re = g.db.execute(sql).fetchall() + self.trait_list = [] + with Bench("Creating trait objects"): + for line in re: + dataset = create_dataset(line[3], "ProbeSet", get_samplelist=False) + trait_id = line[4] + #with Bench("Building trait object"): + this_trait = GeneralTrait(dataset=dataset, name=trait_id, get_qtl_info=True, get_sample_info=False) + self.trait_list.append(this_trait) + + elif self.type == "phenotype": + sql = """ + SELECT + Species.`Name`, + InbredSet.`Name`, + PublishFreeze.`Name`, + PublishXRef.`Id`, + Phenotype.`Post_publication_description`, + Publication.`Authors`, + Publication.`Year`, + PublishXRef.`LRS`, + PublishXRef.`Locus`, + PublishXRef.`additive` + FROM Species,InbredSet,PublishFreeze,PublishXRef,Phenotype,Publication + WHERE PublishXRef.`InbredSetId`=InbredSet.`Id` + AND PublishFreeze.`InbredSetId`=InbredSet.`Id` + AND InbredSet.`SpeciesId`=Species.`Id` + AND PublishXRef.`PhenotypeId`=Phenotype.`Id` + AND PublishXRef.`PublicationId`=Publication.`Id` + AND (Phenotype.Post_publication_description REGEXP "[[:<:]]%s[[:>:]]" + OR Phenotype.Pre_publication_description REGEXP "[[:<:]]%s[[:>:]]" + OR Phenotype.Pre_publication_abbreviation REGEXP "[[:<:]]%s[[:>:]]" + OR Phenotype.Post_publication_abbreviation REGEXP "[[:<:]]%s[[:>:]]" + OR Phenotype.Lab_code REGEXP "[[:<:]]%s[[:>:]]" + OR Publication.PubMed_ID REGEXP "[[:<:]]%s[[:>:]]" + OR Publication.Abstract REGEXP "[[:<:]]%s[[:>:]]" + OR Publication.Title REGEXP "[[:<:]]%s[[:>:]]" + OR Publication.Authors REGEXP "[[:<:]]%s[[:>:]]" + OR PublishXRef.Id REGEXP "[[:<:]]%s[[:>:]]") + ORDER BY Species.`Name`, InbredSet.`Name`, PublishXRef.`Id` + LIMIT 6000 + """ % (self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms, self.terms) + re = g.db.execute(sql).fetchall() + self.trait_list = [] + with Bench("Creating trait objects"): + for line in re: + dataset = create_dataset(line[2], "Publish") + trait_id = line[3] + this_trait = GeneralTrait(dataset=dataset, name=trait_id, get_qtl_info=True, get_sample_info=False) + self.trait_list.append(this_trait) + + self.results = self.convert_to_json() + + def convert_to_json(self): + json_dict = {} + #json_dict['draw'] = self.draw, + json_dict['recordsTotal'] = len(self.trait_list), + json_dict['data'] = [] + + for i, trait in enumerate(self.trait_list): + trait_row = { "checkbox": "<INPUT TYPE=\"checkbox\" NAME=\"searchResult\" class=\"checkbox trait_checkbox\" style=\"transform: scale(1.5);\" VALUE=\"{}:{}\">".format(trait.name, trait.dataset.name), + "index": i+1, + "species": trait.dataset.group.species, + "group": trait.dataset.group.name, + "tissue": trait.dataset.tissue, + "dataset": trait.dataset.fullname, + "record": "<a href=\"/show_trait?trait_id=" + trait.name + "&dataset=" + trait.dataset.name + "\" target=\"_blank\">" + trait.name + "</a>", + "symbol": trait.symbol, + "description": trait.description_display, + "location": trait.location_repr, + "mean": trait.mean, + "max_lrs": trait.LRS_score_repr, + "max_lrs_location": trait.LRS_location_repr, + "additive_effect": trait.additive} + json_dict['data'].append(trait_row) + + json_results = json.dumps(json_dict) + return json_results + + + + diff --git a/wqflask/wqflask/views.py b/wqflask/wqflask/views.py index bd2fff50..7854b0df 100644 --- a/wqflask/wqflask/views.py +++ b/wqflask/wqflask/views.py @@ -34,6 +34,7 @@ from flask import (render_template, request, make_response, Response, from wqflask import search_results from wqflask import gsearch +from wqflask import update_search_results from wqflask import docs from wqflask import news from base.data_set import DataSet # Used by YAML in marker_regression @@ -169,6 +170,17 @@ def gsearchact(): return render_template("gsearch_gene.html", **result) elif type == "phenotype": return render_template("gsearch_pheno.html", **result) + +@app.route("/gsearch_updating", methods=('POST',)) +def gsearch_updating(): + print("REQUEST ARGS:", request.values) + result = update_search_results.GSearch(request.args).__dict__ + return result['results'] + # type = request.args['type'] + # if type == "gene": + # return render_template("gsearch_gene_updating.html", **result) + # elif type == "phenotype": + # return render_template("gsearch_pheno.html", **result) @app.route("/docedit") def docedit(): |