def getSignigicantPhenotypes(type, value):
phenotypes = pandas.DataFrame()
# First we check the type of the input (rsID, Gene or Phenotype) and act accordingly
UserLogs.add('Victor Dupuy', '255.255.255.255',
'accessed the module : Tools - getPhenotypes',
'MySQL Database',
['assoc', 'marqueurs', 'phenotypes', 'experiment'])
treshold = "0.001"
if type == "rsID":
#phenotypes = connect.fetchData("select p.nom From marqueurs m join assoc a on a.rs_id_assoc = m.nom JOIN experiment xp on a.experiment = xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes where pvalue_assoc <"+treshold+" and m.nom in ('"+value+"') order by a.pvalue_assoc ASC")
phenotypes = connect.fetchData(
"select p.nom From marqueurs m join assoc a on a.rs_id_assoc = m.nom JOIN experiment xp on a.experiment = xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes where m.nom in ('"
+ value + "')")
elif type == "gene":
#phenotypes = connect.fetchData("select DISTINCT p.nom From marqueurs m join assoc a on a.rs_id_assoc = m.nom JOIN experiment xp on a.experiment = xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes where pvalue_assoc <"+treshold+" and (m.gene in ('"+value+"') or m.gene_before in ('"+value+"') or m.gene_after in ('"+value+"') ) order by a.pvalue_assoc ASC ")
phenotypes = connect.fetchData(
"select DISTINCT p.nom From marqueurs m join assoc a on a.rs_id_assoc = m.nom JOIN experiment xp on a.experiment = xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes where (m.gene in ('"
+ value + "') or m.gene_before in ('" + value +
"') or m.gene_after in ('" + value + "') ) ")
else:
phenotypes = pandas.DataFrame(data={'nom': value}, index=[0])
return phenotypes
def AutoCompletePhenotypes(request, text):
# First, we fetch the genes starting by the text value
sqlQuery = "select nom from phenotypes where nom like '" + text + "%' and type='binomial'"
phenotypesStarting = connect.fetchData(sqlQuery)
# Same for the end of the name
sqlQuery = "select nom from phenotypes where nom like '%" + text + "' and type='binomial'"
phenotypesEnding = connect.fetchData(sqlQuery)
# Finally, same with the name containing the text value
sqlQuery = "select nom from phenotypes where nom like '%" + text + "%' and type='binomial'"
phenotypesContaining = connect.fetchData(sqlQuery)
# We merge the dataframes
phenotypes = pandas.merge(
pandas.merge(phenotypesStarting, phenotypesEnding, on="nom", how="outer"),
phenotypesContaining,
on="nom",
how="outer",
)
# UserLogs.add(
# 'Victor Dupuy',
# '255.255.255.255',
# 'accessed the module : AutoComplete',
# 'MySQL Database',
# ['phenotypes']
# )
phenotypes = phenotypes.to_json(orient="records")
return HttpResponse(phenotypes)
def AutoCompletePhenotypes(request, text):
#First, we fetch the genes starting by the text value
sqlQuery = "select nom from phenotypes where nom like '" + text + "%' and type='binomial'"
phenotypesStarting = connect.fetchData(sqlQuery)
#Same for the end of the name
sqlQuery = "select nom from phenotypes where nom like '%" + text + "' and type='binomial'"
phenotypesEnding = connect.fetchData(sqlQuery)
#Finally, same with the name containing the text value
sqlQuery = "select nom from phenotypes where nom like '%" + text + "%' and type='binomial'"
phenotypesContaining = connect.fetchData(sqlQuery)
#We merge the dataframes
phenotypes = pandas.merge(pandas.merge(phenotypesStarting,
phenotypesEnding,
on="nom",
how="outer"),
phenotypesContaining,
on="nom",
how="outer")
# UserLogs.add(
# 'Victor Dupuy',
# '255.255.255.255',
# 'accessed the module : AutoComplete',
# 'MySQL Database',
# ['phenotypes']
# )
phenotypes = phenotypes.to_json(orient='records')
return HttpResponse(phenotypes)
def autoCompleteGenes(request, text):
#First, we fetch the genes starting by the text value
sqlQuery = "select symbol from genes where symbol like '" + text + "%'"
genesStarting = connect.fetchData(sqlQuery)
#Same for the end of the name
sqlQuery = "select symbol from genes where symbol like '%" + text + "'"
genesEnding = connect.fetchData(sqlQuery)
#Finally, same with the name containing the text value
sqlQuery = "select symbol from genes where symbol like '%" + text + "%'"
genesContaining = connect.fetchData(sqlQuery)
#We merge the dataframes
genes = pandas.merge(pandas.merge(genesStarting,
genesEnding,
on="symbol",
how="outer"),
genesContaining,
on="symbol",
how="outer")
# UserLogs.add(
# 'Victor Dupuy',
# '255.255.255.255',
# 'accessed the module : AutoComplete',
# 'MySQL Database',
# ['genes']
# )
genes = genes.to_json(orient='records')
return HttpResponse(genes)
def getManhattanData(type, value):
# We'll need to get which phenotypes are selected.
significantPhenotypes = getPhenotypes.getSignigicantPhenotypes(type, value)
if significantPhenotypes.any()['nom']: # Check if there are significant phenotypes
phenotype = getPhenotypes.getSignigicantPhenotypes(type, value).iat[0,0] #This gets the name of the significant phenotype
# phenotype = "All cause death"
# First we define the query
sqlQuery="select distinct a.rs_id_assoc, a.chromosome,a.pos,a.info_assoc,a.pvalue_assoc,a.allele_A,a.allele_B,a.cohort_AA,a.cohort_BB,a.beta_assoc,a.maf, a.all_OR,xp.covariates,m.gene,m.gene_before,m.gene_after from assoc a join experiment xp on a.experiment=xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes join marqueurs m on a.rs_id_assoc=m.nom where p.nom='"+phenotype+"' and a.pvalue_assoc<=0.001"
# Then we fetch the data and store it in a dataframe
sorted_data = connect.fetchData(sqlQuery)
UserLogs.add(
'Victor Dupuy',
'255.255.255.255',
'accessed the module : Interactive Manhattan',
'MySQL Database',
['assoc', 'phenotypes', 'marqueurs']
)
sorted_data.drop_duplicates(subset='rs_id_assoc', inplace=True,keep='last')
sorted_data["log10"] = -numpy.log10(sorted_data.pvalue_assoc) #ADD COLUMN LOG10
sorted_data = sorted_data.sort(['chromosome', 'pos'])
sorted_data['even']=numpy.where(sorted_data['chromosome'] %2==0,sorted_data['log10'] , 'NaN')
sorted_data["odd"]=numpy.where(sorted_data['chromosome'] %2!=0,sorted_data['log10'] , 'NaN')
col=['rs_id_assoc', 'chromosome', 'pos', 'pvalue_assoc', 'allele_A', 'allele_B', 'covariates', 'cohort_BB', 'cohort_AA', 'beta_assoc', 'maf']
return sorted_data, phenotype
else:
print "There is no phenotype !"
return "No Phenotype !"
def getManhattanData(type, value):
# We'll need to get which phenotypes are selected.
significantPhenotypes = getPhenotypes.getSignigicantPhenotypes(type, value)
if significantPhenotypes.any()['nom']: # Check if there are significant phenotypes
phenotype = getPhenotypes.getSignigicantPhenotypes(type, value).iat[0,0] #This gets the name of the significant phenotype
# phenotype = "All cause death"
# First we define the query
sqlQuery="select distinct a.rs_id_assoc, a.chromosome,a.pos,a.info_assoc,a.pvalue_assoc,a.allele_A,a.allele_B,a.cohort_AA,a.cohort_BB,a.beta_assoc,a.maf, a.all_OR,xp.covariates,m.gene,m.gene_before,m.gene_after from assoc a join experiment xp on a.experiment=xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes join marqueurs m on a.rs_id_assoc=m.nom where p.nom='"+phenotype+"' and a.pvalue_assoc<=0.001 limit 1000"
# Then we fetch the data and store it in a dataframe
sorted_data = connect.fetchData(sqlQuery)
UserLogs.add(
'Victor Dupuy',
'255.255.255.255',
'accessed the module : Interactive Manhattan',
'MySQL Database',
['assoc', 'phenotypes', 'marqueurs']
)
sorted_data.drop_duplicates(subset='rs_id_assoc', inplace=True,keep='last')
sorted_data["log10"] = -numpy.log10(sorted_data.pvalue_assoc) #ADD COLUMN LOG10
sorted_data = sorted_data.sort(['chromosome', 'pos'])
sorted_data['even']=numpy.where(sorted_data['chromosome'] %2==0,sorted_data['log10'] , 'NaN')
sorted_data["odd"]=numpy.where(sorted_data['chromosome'] %2!=0,sorted_data['log10'] , 'NaN')
col=['rs_id_assoc', 'chromosome', 'pos', 'pvalue_assoc', 'allele_A', 'allele_B', 'covariates', 'cohort_BB', 'cohort_AA', 'beta_assoc', 'maf']
return sorted_data, phenotype
else:
print "There is no phenotype !"
return "No Phenotype !"
def getSignigicantPhenotypes(type, value):
phenotypes = pandas.DataFrame()
# First we check the type of the input (rsID, Gene or Phenotype) and act accordingly
UserLogs.add(
'Victor Dupuy',
'255.255.255.255',
'accessed the module : Tools - getPhenotypes',
'MySQL Database',
['assoc', 'marqueurs', 'phenotypes' , 'experiment']
)
treshold = "0.001"
if type == "rsID":
phenotypes = connect.fetchData("select p.nom From marqueurs m join assoc a on a.rs_id_assoc = m.nom JOIN experiment xp on a.experiment = xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes where pvalue_assoc <"+treshold+" and m.nom in ('"+value+"') order by a.pvalue_assoc ASC")
elif type == "gene":
phenotypes = connect.fetchData("select DISTINCT p.nom From marqueurs m join assoc a on a.rs_id_assoc = m.nom JOIN experiment xp on a.experiment = xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes where pvalue_assoc <"+treshold+" and (m.gene in ('"+value+"') or m.gene_before in ('"+value+"') or m.gene_after in ('"+value+"') ) order by a.pvalue_assoc ASC ")
else:
phenotypes = pandas.DataFrame(data={'nom': value}, index=[0])
return phenotypes
def autoCompleteGenes(request, text):
# First, we fetch the genes starting by the text value
sqlQuery = "select symbol from genes where symbol like '" + text + "%'"
genesStarting = connect.fetchData(sqlQuery)
# Same for the end of the name
sqlQuery = "select symbol from genes where symbol like '%" + text + "'"
genesEnding = connect.fetchData(sqlQuery)
# Finally, same with the name containing the text value
sqlQuery = "select symbol from genes where symbol like '%" + text + "%'"
genesContaining = connect.fetchData(sqlQuery)
# We merge the dataframes
genes = pandas.merge(
pandas.merge(genesStarting, genesEnding, on="symbol", how="outer"), genesContaining, on="symbol", how="outer"
)
# UserLogs.add(
# 'Victor Dupuy',
# '255.255.255.255',
# 'accessed the module : AutoComplete',
# 'MySQL Database',
# ['genes']
# )
genes = genes.to_json(orient="records")
return HttpResponse(genes)
def table(request, type, value):
print request, type, value
global data
UserLogs.add(
'Victor Dupuy',
'255.255.255.255',
'accessed the module : Web',
'MySQL Database',
['assoc', 'phenotypes', 'marqueurs']
)
# We'll need to get which phenotypes are selected.
significantPhenotypes = getPhenotypes.getSignigicantPhenotypes(type, value)
if significantPhenotypes.any()['nom']: # Check if there are significant phenotypes
phenotype_selected = getPhenotypes.getSignigicantPhenotypes(type, value).iat[0,0] #This gets the name of the significant phenotype
print significantPhenotypes
# First we define the query
sqlQuery = "select distinct a.rs_id_assoc,a.chromosome,a.pos,a.pvalue_assoc,m.gene_before,m.gene,m.gene_after,a.experiment,a.info_assoc,a.allele_A,a.allele_B,a.cohort_AA,a.cohort_BB,a.beta_assoc,a.maf,a.all_OR,xp.covariates, p.nom, case when a.beta_assoc>0 and p.Risk_on_rise is true then a.allele_B when a.beta_assoc<0 and p.Risk_on_rise is true then a.allele_A when a.beta_assoc>0 and p.Risk_on_rise is false then a.allele_A when a.beta_assoc<0 and p.Risk_on_rise is false then a.allele_B end as risk_allele, case when a.beta_assoc>0 and p.Risk_on_rise is true then ( (2*a.cohort_BB) + a.cohort_AB ) / ((2*a.cohort_AA) + (2*a.cohort_AB) +(2*a.cohort_BB) ) when a.beta_assoc<0 and p.Risk_on_rise is true then ( (2*a.cohort_AA) + a.cohort_AB ) / ((2*a.cohort_AA) + (2*a.cohort_AB) +(2*a.cohort_BB) ) when a.beta_assoc>0 and p.Risk_on_rise is false then ( (2*a.cohort_AA) + a.cohort_AB ) / ((2*a.cohort_AA) + (2*a.cohort_AB) +(2*a.cohort_BB) ) when a.beta_assoc<0 and p.Risk_on_rise is false then ( (2*a.cohort_BB) + a.cohort_AB ) / ((2*a.cohort_AA) + (2*a.cohort_AB) +(2*a.cohort_BB) ) end as risk_af, case when a.beta_assoc>0 and p.Risk_on_rise is true then a.beta_assoc when a.beta_assoc<0 and p.Risk_on_rise is true then a.beta_assoc*(-1) when a.beta_assoc>0 and p.Risk_on_rise is false then a.beta_assoc*(-1) when a.beta_assoc<0 and p.Risk_on_rise is false then a.beta_assoc end as risk_allele_beta from assoc a join experiment xp on a.experiment=xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes join marqueurs m on a.rs_id_assoc=m.nom where p.nom='"+phenotype_selected+"' and a.pvalue_assoc<0.001"
# Then we fetch the data and store it in a dataframe
dataframe = connect.fetchData(sqlQuery)
# dataframe.rename(columns={'rs_id_assoc':'rs id assoc'}) #Self-explanatory
data = dataframe
dataframe.rename(columns = {
'rs_id_assoc': 'rs_ID',
'pos': 'Pos',
'chromosome': 'Chr',
'gene_before': 'GeneBefore',
'pvalue_assoc': 'P-value',
'allele_A': 'Allele A',
'allele_B': 'Allele B',
'cohort_AA' : 'cohort AA',
'cohort_BB' : 'cohort BB',
'beta_assoc' : 'Beta Assoc',
'covariates' : 'Covariates',
'risk_allele' : 'Risk Allele',
'risk_af' : 'Risk Af',
'risk_allele_beta' : 'Risk Allele Beta',
'gene': 'Gene',
'gene_after': 'GeneAfter',
'nom' : 'Phenotype'
}, inplace=True) #rename column to make them look nicer
jsonData = dataframe.to_json(orient='records')
significantPhenotypes.rename(columns={'nom' : 'Phenotypes'}, inplace=True);
# Then we pass the dataframe to the client with json format
response = json.dumps({
'noResult' : False,
'phenotypes': significantPhenotypes.to_json(orient='records'),
'data' : jsonData
},
sort_keys=True,
indent=4,
separators=(',', ': ')
)
return HttpResponse(response)
else:
print "There is no phenotype !"
response = json.dumps({
'noResult' : True,
'phenotypes': [],
'data' : []
},
sort_keys=True,
indent=4,
separators=(',', ': ')
)
return HttpResponse(response)
def getAll():
phenotypes = connect.fetchData("select nom from phenotypes where type='binomial';")
return phenotypes
def comparison(phenotypes):
phenotype1=connect.fetchData("select distinct a.rs_id_assoc,a.chromosome, a.pos, a.allele_A, a.allele_B,m.gene_before,m.gene, m.gene_after from assoc a join marqueurs m on a.rs_id_assoc=m.nom JOIN experiment xp on a.experiment = xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes where a. pvalue_assoc<0.001 and p.nom = '"+phenotype1+"'")
phenotype2=connect.fetchData("select distinct a.rs_id_assoc,a.chromosome, a.pos, a.allele_A, a.allele_B,m.gene_before,m.gene, m.gene_after from assoc a join marqueurs m on a.rs_id_assoc=m.nom JOIN experiment xp on a.experiment = xp.idexperiment join phenotypes p on xp.phenotype=p.idphenotypes where a. pvalue_assoc<0.001 and p.nom = '"+phenotype2+"'")
data = pd.merge(chrarea1,chrarea2,on='rs_id_assoc',how='inner')
def getAreaSelectionData(position_min, position_max, chromosome, phenotype):
#We query the database for the relevant snps around our selected position
sqlQuery = "select distinct a.rs_id_assoc , a.chromosome,a.pos,a.info_assoc,a.pvalue_assoc,a.allele_A,a.allele_B,a.cohort_AA,a.cohort_BB, a.cohort_AB, a.beta_assoc,a.maf, a.all_OR,xp.covariates,p.Risk_on_rise,dat.name from assoc a join experiment xp on a.experiment=xp.idexperiment join dataset dat on xp.dataset=dat.iddataset join phenotypes p on xp.phenotype=p.idphenotypes where p.nom='"+phenotype+"' and a.chromosome="+chromosome+" limit 1000000;"
snps = connect.fetchData(sqlQuery)
UserLogs.add(
'Victor Dupuy',
'255.255.255.255',
'accessed the module : AreaSelection',
'MySQL Database',
['assoc', 'phenotypes', 'experiment']
)
snps = snps[(snps['pos']>position_min) & (snps['pos']<position_max)]
snps["risk_allele"] = numpy.where(\
numpy.logical_or(\
numpy.logical_and(snps['beta_assoc']>0, snps['Risk_on_rise']==1),\
numpy.logical_and(snps['beta_assoc']<0, snps['Risk_on_rise']==0)\
),\
snps['allele_B'],\
snps['allele_A']\
)
# snps["risk_allele"]=numpy.where((snps['beta_assoc'] > 0 & snps['Risk_on_rise']==1)|(snps['beta_assoc'] < 0 & snps['Risk_on_rise']==0), snps['allele_B'], snps['allele_A']) # select risk allele
snps["risk_af"] = numpy.where(\
numpy.logical_or(\
numpy.logical_and(snps['beta_assoc'] > 0,snps['Risk_on_rise']==1),\
numpy.logical_and(snps['beta_assoc'] < 0,snps['Risk_on_rise']==0)\
),\
((2*snps["cohort_BB"])+snps["cohort_AB"])/((2*snps["cohort_AA"])+(2*snps["cohort_AB"])+(2*snps["cohort_BB"])),\
((2*snps["cohort_AA"])+snps["cohort_AB"])/((2*snps["cohort_AA"])+(2*snps["cohort_AB"])+(2*snps["cohort_BB"]))\
)
# snps["risk_af"]=numpy.where((snps['beta_assoc'] > 0 & snps['Risk_on_rise']==1)|(snps['beta_assoc'] < 0 & snps['Risk_on_rise']==0), ((2*snps["cohort_BB"])+snps["cohort_AB"])/((2*snps["cohort_AA"])+(2*snps["cohort_AB"])+(2*snps["cohort_BB"])), ((2*snps["cohort_AA"])+snps["cohort_AB"])/((2*snps["cohort_AA"])+(2*snps["cohort_AB"])+(2*snps["cohort_BB"]))) #calculate allele frequency for each allele
snps["risk_allele_beta"] = numpy.where(\
numpy.logical_or(\
numpy.logical_and(snps['beta_assoc'] > 0,snps['Risk_on_rise']==1),\
numpy.logical_and(snps['beta_assoc'] < 0,snps['Risk_on_rise']==0)\
),\
snps['beta_assoc'],\
snps['beta_assoc']*-1\
)
# snps["risk_allele_beta"]=numpy.where((snps['beta_assoc'] > 0 & snps['Risk_on_rise']==1)|(snps['beta_assoc'] < 0 & snps['Risk_on_rise']==0), snps['beta_assoc'], snps['beta_assoc']*-1) #update beta according to risk allele result
snps.rename(columns = {'nom' : 'rs_id_assoc'}, inplace=True)
sqlQuery2 = ("select m.nom, m.gene, m.gene_before, m.gene_after, m.end_gen_after,m.end_gen,m.start_gen,m.end_gen_before,m.func,m.position,m.start_gen_after,m.start_gen_before, m.observed "
+" from marqueurs m where m.chromosome="+chromosome+" and position between "+str(position_min)+" and "+str(position_max))
lastSnps= connect.fetchData(sqlQuery2)
UserLogs.add(
'Victor Dupuy',
'255.255.255.255',
'accessed the module : AreaSelection',
'MySQL Database',
['marqueurs']
)
lastSnps.rename(columns = {'nom':'rs_id_assoc'}, inplace=True)
#We sort the dataframe by the snp's position
lastSnps.sort_values(by="position")
# print lastSnps
# Then we create the Plot Title :
title = "CHROMOSOME :"+str(chromosome)+" \nBETWEEN POSITIONS "+str(position_min)+" AND "+str(position_max)
# Then we merge the two dataframes together.
pandas.set_option('display.max_colwidth',-1) #important to make links appear in pandas dataframe
snps = pandas.merge(snps,lastSnps,on='rs_id_assoc',how='outer')
del snps["pos"]
return snps
def getAreaSelectionData(position_min, position_max, chromosome, phenotype):
#We query the database for the relevant snps around our selected position
sqlQuery = "select distinct a.rs_id_assoc , a.chromosome,a.pos,a.info_assoc,a.pvalue_assoc,a.allele_A,a.allele_B,a.cohort_AA,a.cohort_BB, a.cohort_AB, a.beta_assoc,a.maf, a.all_OR,xp.covariates,p.Risk_on_rise,dat.name from assoc a join experiment xp on a.experiment=xp.idexperiment join dataset dat on xp.dataset=dat.iddataset join phenotypes p on xp.phenotype=p.idphenotypes where p.nom='" + phenotype + "' and a.chromosome=" + chromosome + " limit 1000000;"
snps = connect.fetchData(sqlQuery)
UserLogs.add('Victor Dupuy', '255.255.255.255',
'accessed the module : AreaSelection', 'MySQL Database',
['assoc', 'phenotypes', 'experiment'])
snps = snps[(snps['pos'] > position_min) & (snps['pos'] < position_max)]
snps["risk_allele"] = numpy.where(\
numpy.logical_or(\
numpy.logical_and(snps['beta_assoc']>0, snps['Risk_on_rise']==1),\
numpy.logical_and(snps['beta_assoc']<0, snps['Risk_on_rise']==0)\
),\
snps['allele_B'],\
snps['allele_A']\
)
# snps["risk_allele"]=numpy.where((snps['beta_assoc'] > 0 & snps['Risk_on_rise']==1)|(snps['beta_assoc'] < 0 & snps['Risk_on_rise']==0), snps['allele_B'], snps['allele_A']) # select risk allele
snps["risk_af"] = numpy.where(\
numpy.logical_or(\
numpy.logical_and(snps['beta_assoc'] > 0,snps['Risk_on_rise']==1),\
numpy.logical_and(snps['beta_assoc'] < 0,snps['Risk_on_rise']==0)\
),\
((2*snps["cohort_BB"])+snps["cohort_AB"])/((2*snps["cohort_AA"])+(2*snps["cohort_AB"])+(2*snps["cohort_BB"])),\
((2*snps["cohort_AA"])+snps["cohort_AB"])/((2*snps["cohort_AA"])+(2*snps["cohort_AB"])+(2*snps["cohort_BB"]))\
)
# snps["risk_af"]=numpy.where((snps['beta_assoc'] > 0 & snps['Risk_on_rise']==1)|(snps['beta_assoc'] < 0 & snps['Risk_on_rise']==0), ((2*snps["cohort_BB"])+snps["cohort_AB"])/((2*snps["cohort_AA"])+(2*snps["cohort_AB"])+(2*snps["cohort_BB"])), ((2*snps["cohort_AA"])+snps["cohort_AB"])/((2*snps["cohort_AA"])+(2*snps["cohort_AB"])+(2*snps["cohort_BB"]))) #calculate allele frequency for each allele
snps["risk_allele_beta"] = numpy.where(\
numpy.logical_or(\
numpy.logical_and(snps['beta_assoc'] > 0,snps['Risk_on_rise']==1),\
numpy.logical_and(snps['beta_assoc'] < 0,snps['Risk_on_rise']==0)\
),\
snps['beta_assoc'],\
snps['beta_assoc']*-1\
)
# snps["risk_allele_beta"]=numpy.where((snps['beta_assoc'] > 0 & snps['Risk_on_rise']==1)|(snps['beta_assoc'] < 0 & snps['Risk_on_rise']==0), snps['beta_assoc'], snps['beta_assoc']*-1) #update beta according to risk allele result
snps.rename(columns={'nom': 'rs_id_assoc'}, inplace=True)
sqlQuery2 = (
"select m.nom, m.gene, m.gene_before, m.gene_after, m.end_gen_after,m.end_gen,m.start_gen,m.end_gen_before,m.func,m.position,m.start_gen_after,m.start_gen_before, m.observed "
+ " from marqueurs m where m.chromosome=" + chromosome +
" and position between " + str(position_min) + " and " +
str(position_max))
lastSnps = connect.fetchData(sqlQuery2)
UserLogs.add('Victor Dupuy', '255.255.255.255',
'accessed the module : AreaSelection', 'MySQL Database',
['marqueurs'])
lastSnps.rename(columns={'nom': 'rs_id_assoc'}, inplace=True)
#We sort the dataframe by the snp's position
lastSnps.sort_values(by="position")
# print lastSnps
# Then we create the Plot Title :
title = "CHROMOSOME :" + str(chromosome) + " \nBETWEEN POSITIONS " + str(
position_min) + " AND " + str(position_max)
# Then we merge the two dataframes together.
pandas.set_option('display.max_colwidth',
-1) #important to make links appear in pandas dataframe
snps = pandas.merge(snps, lastSnps, on='rs_id_assoc', how='outer')
del snps["pos"]
return snps
def getAll():
phenotypes = connect.fetchData(
"select nom from phenotypes where type='binomial';")
return phenotypes
