def run(self):
if self.debug:
import pdb
pdb.set_trace()
db = Stock_250kDB(drivername=self.drivername, username=self.db_user,
password=self.db_passwd, hostname=self.hostname, database=self.dbname, schema=self.schema)
db.setup()
session = db.session
results_method_id_info = self.getResultsMethodIDInfo(db, self.call_method_id_ls, self.min_distance, self.get_closest, self.min_MAF)
results_method_id2gene_set = self.getResultsMethodID2GeneSet(db, results_method_id_info, self.results_directory, self.max_rank)
rdata = self.getDataMatrix(results_method_id2gene_set, results_method_id_info)
header = ['', ''] + results_method_id_info.results_method_id_label_ls
strain_acc_list = results_method_id_info.results_method_id_label_ls
category_list = results_method_id_info.results_method_id_ls
if SNPData.isDataMatrixEmpty(rdata.data_matrix):
sys.stderr.write("Nothing fetched from database.\n")
sys.exit(3)
if self.output_fname:
write_data_matrix(rdata.data_matrix, self.output_fname, header, strain_acc_list, category_list)
if self.fig_fname:
font = get_font(self.font_path, font_size=self.font_size) #2008-08-01
value2color_func = lambda x: Value2Color.value2HSLcolor(x, rdata.min_value, rdata.max_value)
im_legend = drawContinousLegend(rdata.min_value, rdata.max_value, self.no_of_ticks, value2color_func, font)
#im.save('%s_legend.png'%self.fig_fname_prefix)
im = drawMatrix(rdata.data_matrix, value2color_func, strain_acc_list,\
strain_acc_list, with_grid=1, font=font)
im = combineTwoImages(im, im_legend, font=font)
im.save(self.fig_fname)
def run(self):
if self.debug:
import pdb
pdb.set_trace()
db = Stock_250kDB.Stock_250kDB(drivername=self.drivername, username=self.db_user,
password=self.db_passwd, hostname=self.hostname, database=self.dbname, schema=self.schema)
db.setup(create_tables=False)
session = db.session
import MySQLdb
mysql_conn = MySQLdb.connect(db=self.dbname, host='banyan.usc.edu', user = self.db_user, passwd = self.db_passwd)
mysql_curs = mysql_conn.cursor()
from pymodule.utils import get_gene_id2gene_symbol
gene_id2gene_symbol = get_gene_id2gene_symbol(mysql_curs, 3702, table='genome.gene', upper_case_gene_symbol=0) #3702 is At's tax id
self.get_closest = 0
gene_id2score = self.get_results_by_gene(self.call_method_id, self.analysis_method_id, self.phenotype_method_id,\
self.min_distance, self.get_closest, self.min_MAF, self.results_directory, self.score_is_rank)
pathway_id2gene_id_ls = self.get_pathway_id2gene_id_ls()
min_score, max_score = self.get_min_max_score(pathway_id2gene_id_ls, gene_id2score)
value2color_func = lambda x: Value2Color.value2HSLcolor(x, min_score, max_score)
if not gene_id2score:
sys.stderr.write("gene_id2score is empty. exit.\n")
sys.exit(3)
gap = 0.05
axe_auto_x_offset = 0.05
axe_auto_y_offset = 0.7
axe_auto_width = 0.3
axe_auto_height = 0.2
axe_auto = pylab.axes([axe_auto_x_offset, axe_auto_y_offset, axe_auto_width, axe_auto_height])
axe_vern_x_offset = axe_auto_x_offset + axe_auto_width + gap
axe_vern_y_offset = axe_auto_y_offset
axe_vern_width = axe_auto_width*0.75
axe_vern_height = axe_auto_height
axe_vern = pylab.axes([axe_vern_x_offset, axe_vern_y_offset, axe_vern_width, axe_vern_height])
axe_light_x_offset = axe_vern_x_offset + axe_vern_width + gap
axe_light_y_offset = axe_vern_y_offset
axe_light_width = axe_auto_width
axe_light_height = axe_auto_height
axe_light = pylab.axes([axe_light_x_offset, axe_light_y_offset, axe_light_width, axe_light_height])
axe_flc_act_x_offset = 0.05
axe_flc_act_y_offset = 0.45
axe_flc_act_width = 0.3
axe_flc_act_height = 0.2
axe_flc_act = pylab.axes([axe_flc_act_x_offset, axe_flc_act_y_offset, axe_flc_act_width, axe_flc_act_height])
axe_repr_x_offset = 0.05
axe_repr_y_offset = 0.1
axe_repr_width = 0.3
axe_repr_height = 0.2
axe_repr = pylab.axes([axe_repr_x_offset, axe_repr_y_offset, axe_repr_width, axe_repr_height] )
axe_flc_x_offset = 0.4
axe_flc_y_offset = 0.5
axe_flc_width = 0.1
axe_flc_height = axe_flc_act_height/3
axe_flc = pylab.axes([axe_flc_x_offset, axe_flc_y_offset, axe_flc_width, axe_flc_height])
axe_co_x_offset = 0.6
axe_co_y_offset = axe_flc_y_offset
axe_co_width = axe_flc_width
axe_co_height = axe_flc_height
axe_co = pylab.axes([axe_co_x_offset, axe_co_y_offset, axe_co_width, axe_co_height])
axe_gibbe_x_offset = axe_co_x_offset + axe_co_width + gap
axe_gibbe_y_offset = axe_co_y_offset
axe_gibbe_width = axe_auto_width*0.75
axe_gibbe_height = axe_co_height
axe_gibbe = pylab.axes([axe_gibbe_x_offset, axe_gibbe_y_offset, axe_gibbe_width, axe_gibbe_height ])
axe_int_x_offset = 0.4
axe_int_y_offset = 0.3
axe_int_width = 0.3
axe_int_height = 0.05
axe_int = pylab.axes([axe_int_x_offset, axe_int_y_offset, axe_int_width, axe_int_height])
axe_meri_x_offset = 0.4
axe_meri_y_offset = 0.1
axe_meri_width = axe_int_width
axe_meri_height = axe_int_height
axe_meri = pylab.axes([axe_meri_x_offset, axe_meri_y_offset, axe_meri_width, axe_meri_height])
axe_flower = pylab.axes([axe_meri_x_offset, 0.01, axe_meri_width, 0.05])
axe_flower.text(0.5, 0.5, 'Flowering', \
horizontalalignment ='center', verticalalignment='center', size=8)
#iteration over pathway_id2axe will do. omit this section.
axe_flower.set_xticks([])
axe_flower.set_yticks([])
axe_score_legend = pylab.axes([axe_int_x_offset+axe_int_width+0.1, 0.01, 0.1, 0.3], frameon=False) #axes for the legend of LD
axe_score_legend.set_xticks([])
axe_score_legend.set_yticks([])
axe_cover = pylab.axes([0.03,0.01, 0.95,0.95],frameon=False)
axe_cover.set_xticks([])
axe_cover.set_yticks([])
pathway_id2axe = {1:axe_flc_act, 2: axe_repr, 3:axe_light, 4:axe_gibbe, 5:axe_vern, 6:axe_int, 7: axe_meri, \
8:axe_flc, 9:axe_co, 10:axe_auto}
#draw edges first to avoid the edges overwriting the pathway labels
pathway_edges = Stock_250kDB.FTPathwayRelationship.query.all()
for pathway_edge in pathway_edges:
axe_start = pathway_id2axe[pathway_edge.pathway1_id]
axe_stop = pathway_id2axe[pathway_edge.pathway2_id]
axe_cover.set_xlim([0,1])
axe_cover.set_ylim([0,1])
self.draw_edge_between_pathways(axe_cover, axe_start, axe_stop, edge_type=pathway_edge.relationship_type_id, \
arrow_start_point_loc =pathway_edge.arrow_start_point_loc,\
arrow_end_point_loc=pathway_edge.arrow_end_point_loc)
axe_cover.set_xlim([0,1])
axe_cover.set_ylim([0,1])
for pathway_id, axe in pathway_id2axe.iteritems():
axe.set_xticks([])
axe.set_yticks([])
pathway = Stock_250kDB.FTPathway.get(pathway_id)
axe.title.set_text(pathway.short_name)
gene_id_ls = pathway_id2gene_id_ls.get(pathway_id)
if gene_id_ls:
self.draw_pathway_box(gene_id_ls, axe, gene_id2gene_symbol, gene_id2score, value2color_func, x_span=1., y_span=1., \
gene_label_size=8)
#draw the final edge from axe_meri to axe_flower
self.draw_edge_between_pathways(axe_cover, axe_meri, axe_flower, edge_type=0, arrow_start_point_loc='bottom middle',\
arrow_end_point_loc='top middle')
phenotype_method = Stock_250kDB.PhenotypeMethod.get(self.phenotype_method_id)
analysis_method = Stock_250kDB.AnalysisMethod.get(self.analysis_method_id)
axe_cover.title.set_text('%s by %s'%(phenotype_method.short_name, analysis_method.short_name))
axe_cover.set_xlim([0,1])
axe_cover.set_ylim([0,1])
self.drawLegend(axe_score_legend, value2color_func, min_score, max_score, no_of_bands = 20, score_is_rank=self.score_is_rank)
png_output_fname = '%s.png'%self.output_fname_prefix
pylab.savefig(png_output_fname, dpi=600)
pylab.savefig('%s.svg'%self.output_fname_prefix)
def run(self):
if self.debug:
import pdb
pdb.set_trace()
db = StockDB.StockDB(drivername=self.drivername, username=self.db_user,
password=self.db_passwd, hostname=self.hostname, database=self.dbname, schema=self.schema)
db.setup(create_tables=False)
session = db.session
order_by_sentence = " order by c.longitude, c.latitude, e.longitude, e.latitude, e.nativename " #how to order strains.
if self.QC_method_id ==4:
sql_table_str = "from %s e, %s s, %s a, %s c"%(StockDB.Ecotype.table.name, StockDB.Site.table.name, StockDB.Address.table.name,\
StockDB.Country.table.name)
common_where_condition = "where e.siteid=s.id and s.addressid=a.id and a.countryid=c.id %s " + order_by_sentence
strain_where_condition = common_where_condition%(" and e.id=st.ecotypeid")
strain_id_info_query = "select distinct st.id as strainid, e.id as ecotypeid, e.nativename, s.name as sitename, c.abbr %s, %s st %s"%(sql_table_str, StockDB.Strain.table.name, strain_where_condition)
else:
sql_table_str = "from %s q, %s e, %s s, %s a, %s c"%(StockDB.QCCrossMatch.table.name, StockDB.Ecotype.table.name, StockDB.Site.table.name, StockDB.Address.table.name,\
StockDB.Country.table.name)
common_where_condition = "where e.siteid=s.id and s.addressid=a.id and a.countryid=c.id %s"+ " and q.qc_method_id=%s and q.no_of_non_NA_pairs>=%s and q.mismatch_rate<=%s "%\
(self.QC_method_id, self.min_no_of_non_NAs, self.max_mismatch_rate) + order_by_sentence
strain_where_condition = common_where_condition%(" and e.id=st.ecotypeid and st.id=q.strainid")
strain_id_info_query = "select distinct q.strainid, e.id as ecotypeid, e.nativename, s.name as sitename, c.abbr %s, %s st %s"%(sql_table_str, StockDB.Strain.table.name, strain_where_condition)
if self.how_to_group_strains==2 or self.how_to_group_strains==3:
plate_info = self.alignStrainsAccordingToSeqPlate(db)
id_set_data = PassingData()
id_set_data.strain_id_set = None
id_set_data.target_id_set = None
elif self.input_fname:
id_set_data = self.getStrainidTargetidFromFile(db, self.QC_method_id, self.input_fname, self.max_mismatch_rate, self.min_no_of_non_NAs)
else:
id_set_data = PassingData()
id_set_data.strain_id_set = None
id_set_data.target_id_set = None
if self.how_to_group_strains==2 or self.how_to_group_strains==3:
strain_id_info = self.getStrainInfoGivenPlateInfo(db, plate_info, strain_id_info_query, strain_id_set=None)
else:
strain_id_info = self.getStrainIDInfo(db, strain_id_info_query, id_set_data.strain_id_set)
if self.QC_method_id==4:
if self.how_to_group_strains==3:
#2008-09-15 column strain id is in country, strain-longitude order
target_id_info = self.getStrainIDInfo(db, strain_id_info_query, id_set_data.strain_id_set)
else:
target_id_info = strain_id_info
else:
target_where_condition = common_where_condition%(" and e.id=q.target_id")
target_id_info_query = "select distinct e.id as strainid, e.id as ecotypeid, e.nativename, s.name as sitename, c.abbr %s %s"%(sql_table_str, target_where_condition)
target_id_info = self.getStrainIDInfo(db, target_id_info_query)
if self.input_fname:
rdata = self.get_data_matrixFromFile(db, strain_id_info, target_id_info, self.QC_method_id, self.input_fname, self.max_mismatch_rate, self.min_no_of_non_NAs)
else:
rdata = self.get_data_matrix(db, strain_id_info, target_id_info, self.QC_method_id, self.max_mismatch_rate, self.min_no_of_non_NAs)
rdata.data_matrix = self.markDataMatrixBoundary(rdata.data_matrix, strain_id_info, target_id_info)
header = ['strain info', ''] + target_id_info.strain_label_ls
strain_acc_list = strain_id_info.strain_label_ls
category_list = [1]*len(strain_acc_list)
if SNPData.isDataMatrixEmpty(rdata.data_matrix):
sys.stderr.write("Nothing fetched from database.\n")
sys.exit(3)
if self.output_fname:
write_data_matrix(rdata.data_matrix, self.output_fname, header, strain_acc_list, category_list)
if self.fig_fname:
font = get_font(self.font_path, font_size=self.font_size) #2008-08-01
value2color_func = lambda x: Value2Color.value2HSLcolor(x, rdata.min_value, rdata.max_value)
im_legend = drawContinousLegend(rdata.min_value, rdata.max_value, self.no_of_ticks, value2color_func, font)
#im.save('%s_legend.png'%self.fig_fname_prefix)
im = drawMatrix(rdata.data_matrix, value2color_func, strain_id_info.strain_label_ls,\
target_id_info.strain_label_ls, with_grid=1, font=font)
im = combineTwoImages(im, im_legend, font=font)
im.save(self.fig_fname)
def run(self):
if self.debug:
import pdb
pdb.set_trace()
db = StockDB.StockDB(drivername=self.drivername,
username=self.db_user,
password=self.db_passwd,
hostname=self.hostname,
database=self.dbname,
schema=self.schema)
db.setup(create_tables=False)
session = db.session
order_by_sentence = " order by c.longitude, c.latitude, e.longitude, e.latitude, e.nativename " #how to order strains.
if self.QC_method_id == 4:
sql_table_str = "from %s e, %s s, %s a, %s c"%(StockDB.Ecotype.table.name, StockDB.Site.table.name, StockDB.Address.table.name,\
StockDB.Country.table.name)
common_where_condition = "where e.siteid=s.id and s.addressid=a.id and a.countryid=c.id %s " + order_by_sentence
strain_where_condition = common_where_condition % (
" and e.id=st.ecotypeid")
strain_id_info_query = "select distinct st.id as strainid, e.id as ecotypeid, e.nativename, s.name as sitename, c.abbr %s, %s st %s" % (
sql_table_str, StockDB.Strain.table.name,
strain_where_condition)
else:
sql_table_str = "from %s q, %s e, %s s, %s a, %s c"%(StockDB.QCCrossMatch.table.name, StockDB.Ecotype.table.name, StockDB.Site.table.name, StockDB.Address.table.name,\
StockDB.Country.table.name)
common_where_condition = "where e.siteid=s.id and s.addressid=a.id and a.countryid=c.id %s"+ " and q.qc_method_id=%s and q.no_of_non_NA_pairs>=%s and q.mismatch_rate<=%s "%\
(self.QC_method_id, self.min_no_of_non_NAs, self.max_mismatch_rate) + order_by_sentence
strain_where_condition = common_where_condition % (
" and e.id=st.ecotypeid and st.id=q.strainid")
strain_id_info_query = "select distinct q.strainid, e.id as ecotypeid, e.nativename, s.name as sitename, c.abbr %s, %s st %s" % (
sql_table_str, StockDB.Strain.table.name,
strain_where_condition)
if self.how_to_group_strains == 2 or self.how_to_group_strains == 3:
plate_info = self.alignStrainsAccordingToSeqPlate(db)
id_set_data = PassingData()
id_set_data.strain_id_set = None
id_set_data.target_id_set = None
elif self.input_fname:
id_set_data = self.getStrainidTargetidFromFile(
db, self.QC_method_id, self.input_fname,
self.max_mismatch_rate, self.min_no_of_non_NAs)
else:
id_set_data = PassingData()
id_set_data.strain_id_set = None
id_set_data.target_id_set = None
if self.how_to_group_strains == 2 or self.how_to_group_strains == 3:
strain_id_info = self.getStrainInfoGivenPlateInfo(
db, plate_info, strain_id_info_query, strain_id_set=None)
else:
strain_id_info = self.getStrainIDInfo(db, strain_id_info_query,
id_set_data.strain_id_set)
if self.QC_method_id == 4:
if self.how_to_group_strains == 3:
#2008-09-15 column strain id is in country, strain-longitude order
target_id_info = self.getStrainIDInfo(
db, strain_id_info_query, id_set_data.strain_id_set)
else:
target_id_info = strain_id_info
else:
target_where_condition = common_where_condition % (
" and e.id=q.target_id")
target_id_info_query = "select distinct e.id as strainid, e.id as ecotypeid, e.nativename, s.name as sitename, c.abbr %s %s" % (
sql_table_str, target_where_condition)
target_id_info = self.getStrainIDInfo(db, target_id_info_query)
if self.input_fname:
rdata = self.get_data_matrixFromFile(db, strain_id_info,
target_id_info,
self.QC_method_id,
self.input_fname,
self.max_mismatch_rate,
self.min_no_of_non_NAs)
else:
rdata = self.get_data_matrix(db, strain_id_info, target_id_info,
self.QC_method_id,
self.max_mismatch_rate,
self.min_no_of_non_NAs)
rdata.data_matrix = self.markDataMatrixBoundary(
rdata.data_matrix, strain_id_info, target_id_info)
header = ['strain info', ''] + target_id_info.strain_label_ls
strain_acc_list = strain_id_info.strain_label_ls
category_list = [1] * len(strain_acc_list)
if SNPData.isDataMatrixEmpty(rdata.data_matrix):
sys.stderr.write("Nothing fetched from database.\n")
sys.exit(3)
if self.output_fname:
write_data_matrix(rdata.data_matrix, self.output_fname, header,
strain_acc_list, category_list)
if self.fig_fname:
font = get_font(self.font_path,
font_size=self.font_size) #2008-08-01
value2color_func = lambda x: Value2Color.value2HSLcolor(
x, rdata.min_value, rdata.max_value)
im_legend = drawContinousLegend(rdata.min_value, rdata.max_value,
self.no_of_ticks, value2color_func,
font)
#im.save('%s_legend.png'%self.fig_fname_prefix)
im = drawMatrix(rdata.data_matrix, value2color_func, strain_id_info.strain_label_ls,\
target_id_info.strain_label_ls, with_grid=1, font=font)
im = combineTwoImages(im, im_legend, font=font)
im.save(self.fig_fname)
def run(self):
if self.debug:
import pdb
pdb.set_trace()
db = Stock_250kDB.Stock_250kDB(drivername=self.drivername, username=self.db_user,
password=self.db_passwd, hostname=self.hostname, database=self.dbname, schema=self.schema)
db.setup()
session = db.session
if self.test_result_type==1:
test_result_class_table = CandidateGeneRankSumTestResult.table.name
test_result_class_table = 'candidate_gene_rank_sum_test_result_2008_09_15'
elif self.test_result_type==2:
test_result_class_table = CandidateGeneTopSNPTest.table.name
elif self.test_result_type==3:
test_result_class_table = Stock_250kDB.CandidateGeneRankSumTestResultMethod.table.name
else:
sys.stderr.write(" test_result_type %s not supported.\n"%(self.test_result_type))
sys.exit(2)
#the condition for min_MAF is tricky because of the floating precision.
if self.test_result_type==1:
where_condition = "%s r, %s c, %s g where g.id=c.list_type_id and r.analysis_method_id is not null \
and c.results_id=r.id and c.get_closest=%s and c.min_distance=%s and abs(c.min_MAF-%s)<0.00001"\
%(ResultsMethod.table.name, test_result_class_table, GeneListType.table.name, self.get_closest, self.min_distance, self.min_MAF)
elif self.test_result_type==2:
where_condition = "%s r, %s rg, %s c, %s g where g.id=c.list_type_id and r.analysis_method_id is not null and r.id=rg.results_method_id \
and c.results_id=rg.id and c.get_closest=%s and c.min_distance=%s and abs(c.min_MAF-%s)<0.00001"\
%(ResultsMethod.table.name, ResultsByGene.table.name, test_result_class_table, GeneListType.table.name, self.get_closest, self.min_distance, self.min_MAF)
elif self.test_result_type==3:
where_condition = "%s r, %s c, %s g where g.id=c.list_type_id and r.analysis_method_id is not null \
and c.results_id=r.id and c.get_closest=%s and c.min_distance=%s and abs(c.min_MAF-%s)<0.00001"\
%(ResultsMethod.table.name, test_result_class_table, GeneListType.table.name, self.get_closest, self.min_distance, self.min_MAF)
if self.call_method_id_ls:
where_condition += " and r.call_method_id in (%s)"%self.call_method_id_ls
if self.analysis_method_id_ls:
where_condition += " and r.analysis_method_id in (%s)"%self.analysis_method_id_ls
if self.super_type_id:
where_condition += " and g.super_type_id=%s"%self.super_type_id
if self.test_type:
where_condition += " and c.test_type=%s"%self.test_type
if self.test_result_type==1:
pass
where_condition += " and c.max_pvalue_per_gene=%s"%(self.max_pvalue_per_gene)
elif self.test_result_type==2:
where_condition += " and c.no_of_top_snps=%s"%(self.no_of_top_snps)
list_type_id_ls = self.getListTypeInfo(db, where_condition)
analysis_method_id_ls = self.getAnalysisMethodInfo(db, where_condition)
list_type_analysis_method_info = self.orderListTypeAnalysisMethodID(list_type_id_ls, analysis_method_id_ls)
phenotype_info = self.getPhenotypeInfo(db, where_condition)
rdata = self.get_data_matrix(db, phenotype_info, list_type_analysis_method_info, where_condition)
rdata.data_matrix = self.markDataMatrixBoundary(rdata.data_matrix, phenotype_info, list_type_analysis_method_info)
header = ['list_type_analysis_method', ''] + phenotype_info.phenotype_method_label_ls
strain_acc_list = list_type_analysis_method_info.list_type_analysis_method_label_ls
category_list = list_type_analysis_method_info.list_type_id_analysis_method_id_ls
if SNPData.isDataMatrixEmpty(rdata.data_matrix):
sys.stderr.write("Nothing fetched from database.\n")
sys.exit(3)
if self.output_fname:
write_data_matrix(rdata.data_matrix, self.output_fname, header, strain_acc_list, category_list)
if self.fig_fname:
font = get_font(self.font_path, font_size=self.font_size) #2008-08-01
value2color_func = lambda x: Value2Color.value2HSLcolor(x, rdata.min_value, rdata.max_value)
im_legend = drawContinousLegend(rdata.min_value, rdata.max_value, self.no_of_ticks, value2color_func, font)
#im.save('%s_legend.png'%self.fig_fname_prefix)
im = drawMatrix(rdata.data_matrix, value2color_func, list_type_analysis_method_info.list_type_analysis_method_label_ls,\
phenotype_info.phenotype_method_label_ls, with_grid=1, font=font)
im = combineTwoImages(im, im_legend, font=font)
im.save(self.fig_fname)
def run(self):
if self.debug:
import pdb
pdb.set_trace()
db = Stock_250kDB.Stock_250kDB(drivername=self.drivername,
username=self.db_user,
password=self.db_passwd,
hostname=self.hostname,
database=self.dbname,
schema=self.schema)
db.setup(create_tables=False)
session = db.session
import MySQLdb
mysql_conn = MySQLdb.connect(db=self.dbname,
host='banyan.usc.edu',
user=self.db_user,
passwd=self.db_passwd)
mysql_curs = mysql_conn.cursor()
from pymodule.utils import get_gene_id2gene_symbol
gene_id2gene_symbol = get_gene_id2gene_symbol(
mysql_curs, 3702, table='genome.gene',
upper_case_gene_symbol=0) #3702 is At's tax id
self.get_closest = 0
gene_id2score = self.get_results_by_gene(self.call_method_id, self.analysis_method_id, self.phenotype_method_id,\
self.min_distance, self.get_closest, self.min_MAF, self.results_directory, self.score_is_rank)
pathway_id2gene_id_ls = self.get_pathway_id2gene_id_ls()
min_score, max_score = self.get_min_max_score(pathway_id2gene_id_ls,
gene_id2score)
value2color_func = lambda x: Value2Color.value2HSLcolor(
x, min_score, max_score)
if not gene_id2score:
sys.stderr.write("gene_id2score is empty. exit.\n")
sys.exit(3)
gap = 0.05
axe_auto_x_offset = 0.05
axe_auto_y_offset = 0.7
axe_auto_width = 0.3
axe_auto_height = 0.2
axe_auto = pylab.axes([
axe_auto_x_offset, axe_auto_y_offset, axe_auto_width,
axe_auto_height
])
axe_vern_x_offset = axe_auto_x_offset + axe_auto_width + gap
axe_vern_y_offset = axe_auto_y_offset
axe_vern_width = axe_auto_width * 0.75
axe_vern_height = axe_auto_height
axe_vern = pylab.axes([
axe_vern_x_offset, axe_vern_y_offset, axe_vern_width,
axe_vern_height
])
axe_light_x_offset = axe_vern_x_offset + axe_vern_width + gap
axe_light_y_offset = axe_vern_y_offset
axe_light_width = axe_auto_width
axe_light_height = axe_auto_height
axe_light = pylab.axes([
axe_light_x_offset, axe_light_y_offset, axe_light_width,
axe_light_height
])
axe_flc_act_x_offset = 0.05
axe_flc_act_y_offset = 0.45
axe_flc_act_width = 0.3
axe_flc_act_height = 0.2
axe_flc_act = pylab.axes([
axe_flc_act_x_offset, axe_flc_act_y_offset, axe_flc_act_width,
axe_flc_act_height
])
axe_repr_x_offset = 0.05
axe_repr_y_offset = 0.1
axe_repr_width = 0.3
axe_repr_height = 0.2
axe_repr = pylab.axes([
axe_repr_x_offset, axe_repr_y_offset, axe_repr_width,
axe_repr_height
])
axe_flc_x_offset = 0.4
axe_flc_y_offset = 0.5
axe_flc_width = 0.1
axe_flc_height = axe_flc_act_height / 3
axe_flc = pylab.axes([
axe_flc_x_offset, axe_flc_y_offset, axe_flc_width, axe_flc_height
])
axe_co_x_offset = 0.6
axe_co_y_offset = axe_flc_y_offset
axe_co_width = axe_flc_width
axe_co_height = axe_flc_height
axe_co = pylab.axes(
[axe_co_x_offset, axe_co_y_offset, axe_co_width, axe_co_height])
axe_gibbe_x_offset = axe_co_x_offset + axe_co_width + gap
axe_gibbe_y_offset = axe_co_y_offset
axe_gibbe_width = axe_auto_width * 0.75
axe_gibbe_height = axe_co_height
axe_gibbe = pylab.axes([
axe_gibbe_x_offset, axe_gibbe_y_offset, axe_gibbe_width,
axe_gibbe_height
])
axe_int_x_offset = 0.4
axe_int_y_offset = 0.3
axe_int_width = 0.3
axe_int_height = 0.05
axe_int = pylab.axes([
axe_int_x_offset, axe_int_y_offset, axe_int_width, axe_int_height
])
axe_meri_x_offset = 0.4
axe_meri_y_offset = 0.1
axe_meri_width = axe_int_width
axe_meri_height = axe_int_height
axe_meri = pylab.axes([
axe_meri_x_offset, axe_meri_y_offset, axe_meri_width,
axe_meri_height
])
axe_flower = pylab.axes(
[axe_meri_x_offset, 0.01, axe_meri_width, 0.05])
axe_flower.text(0.5, 0.5, 'Flowering', \
horizontalalignment ='center', verticalalignment='center', size=8)
#iteration over pathway_id2axe will do. omit this section.
axe_flower.set_xticks([])
axe_flower.set_yticks([])
axe_score_legend = pylab.axes(
[axe_int_x_offset + axe_int_width + 0.1, 0.01, 0.1, 0.3],
frameon=False) #axes for the legend of LD
axe_score_legend.set_xticks([])
axe_score_legend.set_yticks([])
axe_cover = pylab.axes([0.03, 0.01, 0.95, 0.95], frameon=False)
axe_cover.set_xticks([])
axe_cover.set_yticks([])
pathway_id2axe = {1:axe_flc_act, 2: axe_repr, 3:axe_light, 4:axe_gibbe, 5:axe_vern, 6:axe_int, 7: axe_meri, \
8:axe_flc, 9:axe_co, 10:axe_auto}
#draw edges first to avoid the edges overwriting the pathway labels
pathway_edges = Stock_250kDB.FTPathwayRelationship.query.all()
for pathway_edge in pathway_edges:
axe_start = pathway_id2axe[pathway_edge.pathway1_id]
axe_stop = pathway_id2axe[pathway_edge.pathway2_id]
axe_cover.set_xlim([0, 1])
axe_cover.set_ylim([0, 1])
self.draw_edge_between_pathways(axe_cover, axe_start, axe_stop, edge_type=pathway_edge.relationship_type_id, \
arrow_start_point_loc =pathway_edge.arrow_start_point_loc,\
arrow_end_point_loc=pathway_edge.arrow_end_point_loc)
axe_cover.set_xlim([0, 1])
axe_cover.set_ylim([0, 1])
for pathway_id, axe in pathway_id2axe.iteritems():
axe.set_xticks([])
axe.set_yticks([])
pathway = Stock_250kDB.FTPathway.get(pathway_id)
axe.title.set_text(pathway.short_name)
gene_id_ls = pathway_id2gene_id_ls.get(pathway_id)
if gene_id_ls:
self.draw_pathway_box(gene_id_ls, axe, gene_id2gene_symbol, gene_id2score, value2color_func, x_span=1., y_span=1., \
gene_label_size=8)
#draw the final edge from axe_meri to axe_flower
self.draw_edge_between_pathways(axe_cover, axe_meri, axe_flower, edge_type=0, arrow_start_point_loc='bottom middle',\
arrow_end_point_loc='top middle')
phenotype_method = Stock_250kDB.PhenotypeMethod.get(
self.phenotype_method_id)
analysis_method = Stock_250kDB.AnalysisMethod.get(
self.analysis_method_id)
axe_cover.title.set_text(
'%s by %s' %
(phenotype_method.short_name, analysis_method.short_name))
axe_cover.set_xlim([0, 1])
axe_cover.set_ylim([0, 1])
self.drawLegend(axe_score_legend,
value2color_func,
min_score,
max_score,
no_of_bands=20,
score_is_rank=self.score_is_rank)
png_output_fname = '%s.png' % self.output_fname_prefix
pylab.savefig(png_output_fname, dpi=600)
pylab.savefig('%s.svg' % self.output_fname_prefix)
def run(self):
if self.debug:
import pdb
pdb.set_trace()
db = Stock_250kDB.Stock_250kDB(drivername=self.drivername,
username=self.db_user,
password=self.db_passwd,
hostname=self.hostname,
database=self.dbname,
schema=self.schema)
db.setup()
session = db.session
if self.test_result_type == 1:
test_result_class_table = CandidateGeneRankSumTestResult.table.name
test_result_class_table = 'candidate_gene_rank_sum_test_result_2008_09_15'
elif self.test_result_type == 2:
test_result_class_table = CandidateGeneTopSNPTest.table.name
elif self.test_result_type == 3:
test_result_class_table = Stock_250kDB.CandidateGeneRankSumTestResultMethod.table.name
else:
sys.stderr.write(" test_result_type %s not supported.\n" %
(self.test_result_type))
sys.exit(2)
#the condition for min_MAF is tricky because of the floating precision.
if self.test_result_type == 1:
where_condition = "%s r, %s c, %s g where g.id=c.list_type_id and r.analysis_method_id is not null \
and c.results_id=r.id and c.get_closest=%s and c.min_distance=%s and abs(c.min_MAF-%s)<0.00001" \
%(ResultsMethod.table.name, test_result_class_table, GeneListType.table.name, self.get_closest, self.min_distance, self.min_MAF)
elif self.test_result_type == 2:
where_condition = "%s r, %s rg, %s c, %s g where g.id=c.list_type_id and r.analysis_method_id is not null and r.id=rg.results_method_id \
and c.results_id=rg.id and c.get_closest=%s and c.min_distance=%s and abs(c.min_MAF-%s)<0.00001" \
%(ResultsMethod.table.name, ResultsByGene.table.name, test_result_class_table, GeneListType.table.name, self.get_closest, self.min_distance, self.min_MAF)
elif self.test_result_type == 3:
where_condition = "%s r, %s c, %s g where g.id=c.list_type_id and r.analysis_method_id is not null \
and c.results_id=r.id and c.get_closest=%s and c.min_distance=%s and abs(c.min_MAF-%s)<0.00001" \
%(ResultsMethod.table.name, test_result_class_table, GeneListType.table.name, self.get_closest, self.min_distance, self.min_MAF)
if self.call_method_id_ls:
where_condition += " and r.call_method_id in (%s)" % self.call_method_id_ls
if self.analysis_method_id_ls:
where_condition += " and r.analysis_method_id in (%s)" % self.analysis_method_id_ls
if self.super_type_id:
where_condition += " and g.super_type_id=%s" % self.super_type_id
if self.test_type:
where_condition += " and c.test_type=%s" % self.test_type
if self.test_result_type == 1:
pass
where_condition += " and c.max_pvalue_per_gene=%s" % (
self.max_pvalue_per_gene)
elif self.test_result_type == 2:
where_condition += " and c.no_of_top_snps=%s" % (
self.no_of_top_snps)
list_type_id_ls = self.getListTypeInfo(db, where_condition)
analysis_method_id_ls = self.getAnalysisMethodInfo(db, where_condition)
list_type_analysis_method_info = self.orderListTypeAnalysisMethodID(
list_type_id_ls, analysis_method_id_ls)
phenotype_info = self.getPhenotypeInfo(db, where_condition)
rdata = self.get_data_matrix(db, phenotype_info,
list_type_analysis_method_info,
where_condition)
rdata.data_matrix = self.markDataMatrixBoundary(
rdata.data_matrix, phenotype_info, list_type_analysis_method_info)
header = ['list_type_analysis_method', ''
] + phenotype_info.phenotype_method_label_ls
strain_acc_list = list_type_analysis_method_info.list_type_analysis_method_label_ls
category_list = list_type_analysis_method_info.list_type_id_analysis_method_id_ls
if SNPData.isDataMatrixEmpty(rdata.data_matrix):
sys.stderr.write("Nothing fetched from database.\n")
sys.exit(3)
if self.output_fname:
write_data_matrix(rdata.data_matrix, self.output_fname, header,
strain_acc_list, category_list)
if self.fig_fname:
font = get_font(self.font_path,
font_size=self.font_size) #2008-08-01
value2color_func = lambda x: Value2Color.value2HSLcolor(
x, rdata.min_value, rdata.max_value)
im_legend = drawContinousLegend(rdata.min_value, rdata.max_value,
self.no_of_ticks, value2color_func,
font)
#im.save('%s_legend.png'%self.fig_fname_prefix)
im = drawMatrix(rdata.data_matrix, value2color_func, list_type_analysis_method_info.list_type_analysis_method_label_ls,\
phenotype_info.phenotype_method_label_ls, with_grid=1, font=font)
im = combineTwoImages(im, im_legend, font=font)
im.save(self.fig_fname)
