def outputClusterPlotsForAllSNPs(ordered_snpid_ls, snpid2name, snpid2peak1_peak2_color_ls, fig_output_dir, output_fname, need_draw_figure=0):
#from pymodule.latex import outputFigureInLatex
import os,sys
outf = open(output_fname, 'w')
for snpid in ordered_snpid_ls:
snp_name = snpid2name[snpid]
fig_fname = 'cluster_plot_%s'%(snp_name.replace(' ', '_')) #replace space with _
if need_draw_figure:
output_fname_prefix=os.path.join(fig_output_dir, fig_fname)
drawClusterPlotForOneSNP(snpid2peak1_peak2_color_ls[snpid], snp_name, output_fname_prefix)
#output the figure in latex
fig_fname = os.path.join('figures/', '%s.png'%fig_fname) #relative path of the figure to the latex file
fig_label = 'fl%s'%(snp_name.replace(' ', ''))
caption = 'cluster plot for %s.'%(snp_name)
outf.write(outputFigureInLatex(fig_fname, caption, fig_label, need_floatpackage=1, fig_width=0.5))
del outf
def OutputCliquesGroupedByRegion(curs, region2clique_id_ls, clique_id2size, clique2ecotype_table, component2clique_table, output_fname, fig_output_dir, stock_db='stock20071008', min_clique_size=3, need_to_draw_figures=0):
"""
2007-10-18
"""
from variation.src.common import draw_graph_on_map, get_ecotypeid2pos
from variation.src.common import get_pic_area
from pymodule.latex import outputFigureInLatex
import sys,os
ecotypeid2pos = get_ecotypeid2pos(curs, '%s.ecotype'%stock_db)
outf = open(output_fname, 'w')
no_of_regions = 0
clique_id2region = {} #for the cc2clique latex table
for region, clique_id_ls in region2clique_id_ls.iteritems():
region = ' and '.join(region) #region is tuple, convert it to string
sys.stderr.write("%s:\n"%region)
outf.write("\\subsection{%s}\n"%region)
size_clique_id_ls = []
for clique_id in clique_id_ls:
if clique_id2size[clique_id]>=min_clique_size:
size_clique_id_ls.append([clique_id2size[clique_id], clique_id])
outf.write('region %s has %s cliques in total but %s cliques above min size %s.\n'%(region, len(clique_id_ls), len(size_clique_id_ls), min_clique_size))
if len(size_clique_id_ls)==0:
continue
size_clique_id_ls.sort()
caption = 'region %s has %s cliques in total but %s cliques above min size %s.'%(region, len(clique_id_ls), len(size_clique_id_ls), min_clique_size)
no_of_regions += 1
table_label = 't_region_%s'%no_of_regions
clique_id_size_ls = [[row[1],row[0]] for row in size_clique_id_ls]
header_ls = ['clique_id', 'size']
outf.write(outputMatrixInLatexTable(clique_id_size_ls, caption, table_label, header_ls))
for size, clique_id in size_clique_id_ls:
sys.stderr.write("clique %s, size %s\n"%(clique_id, size))
#output the table and get ecotype_id_ls
fig_label = 'fclique%s'%clique_id
table_label = 'tclique%s'%clique_id
ecotype_id_ls = outputCliqueInLatexTable(curs, region, clique_id, table_label, fig_label, stock_db, clique2ecotype_table, outf)
#draw the clique
site_g, site2weight, site2pos = construct_site_graph_out_of_ecotype_id_ls(ecotype_id_ls, ecotypeid2pos)
fig_fname = 'haplotype_%s_size%s_%ssites_map'%(clique_id,len(ecotype_id_ls), len(site2pos)) #just the name without filetype extension, like png, eps etc.
if need_to_draw_figures:
draw_graph_on_map(site_g, site2weight, site2pos, 'clique %s, size %s, %s sites of %s'%(clique_id, len(ecotype_id_ls), len(site2pos), region), pic_area=get_pic_area(site2pos.values(), 30), output_fname_prefix=os.path.join(fig_output_dir, fig_fname))
#output the figure in latex
fig_fname = os.path.join('figures/', '%s.png'%fig_fname) #relative path of the figure to the latex file
caption = 'map of haplotype clique %s has %s ecotypes from %s sites of %s. check Table~\\ref{%s}.'%(clique_id, len(ecotype_id_ls), len(site2pos), region, table_label)
outf.write(outputFigureInLatex(fig_fname, caption, fig_label))
clique_id2region[clique_id] = region
del site_g, site2pos, site2weight, ecotype_id_ls
outf.write("\\subsection{which component the clique is from}\n")
cc_id_clique_id_region_ls = []
cc_id2clique_id_ls = {}
for clique_id, region in clique_id2region.iteritems():
curs.execute("select cc_id from %s where clique_id=%s"%(component2clique_table, clique_id))
rows = curs.fetchall()
cc_id = rows[0][0]
cc_id_clique_id_region_ls.append([cc_id, clique_id, region])
if cc_id not in cc_id2clique_id_ls:
cc_id2clique_id_ls[cc_id] = []
cc_id2clique_id_ls[cc_id].append(clique_id)
cc_id_clique_id_region_ls.sort()
caption = '%s components versus %s cliques'%(len(cc_id2clique_id_ls), len(cc_id_clique_id_region_ls))
table_label = 't_cc2clique_1'
header_ls = ['component id', 'clique id', 'region']
outf.write(outputMatrixInLatexTable(cc_id_clique_id_region_ls, caption, table_label, header_ls))
del outf
def OutputCliquesGroupedByRegion(curs,
region2clique_id_ls,
clique_id2size,
clique2ecotype_table,
component2clique_table,
output_fname,
fig_output_dir,
stock_db='stock20071008',
min_clique_size=3,
need_to_draw_figures=0):
"""
2007-10-18
"""
from variation.src.common import draw_graph_on_map, get_ecotypeid2pos
from variation.src.common import get_pic_area
from pymodule.latex import outputFigureInLatex
import sys, os
ecotypeid2pos = get_ecotypeid2pos(curs, '%s.ecotype' % stock_db)
outf = open(output_fname, 'w')
no_of_regions = 0
clique_id2region = {} #for the cc2clique latex table
for region, clique_id_ls in region2clique_id_ls.iteritems():
region = ' and '.join(region) #region is tuple, convert it to string
sys.stderr.write("%s:\n" % region)
outf.write("\\subsection{%s}\n" % region)
size_clique_id_ls = []
for clique_id in clique_id_ls:
if clique_id2size[clique_id] >= min_clique_size:
size_clique_id_ls.append(
[clique_id2size[clique_id], clique_id])
outf.write(
'region %s has %s cliques in total but %s cliques above min size %s.\n'
% (region, len(clique_id_ls), len(size_clique_id_ls),
min_clique_size))
if len(size_clique_id_ls) == 0:
continue
size_clique_id_ls.sort()
caption = 'region %s has %s cliques in total but %s cliques above min size %s.' % (
region, len(clique_id_ls), len(size_clique_id_ls), min_clique_size)
no_of_regions += 1
table_label = 't_region_%s' % no_of_regions
clique_id_size_ls = [[row[1], row[0]] for row in size_clique_id_ls]
header_ls = ['clique_id', 'size']
outf.write(
outputMatrixInLatexTable(clique_id_size_ls, caption, table_label,
header_ls))
for size, clique_id in size_clique_id_ls:
sys.stderr.write("clique %s, size %s\n" % (clique_id, size))
#output the table and get ecotype_id_ls
fig_label = 'fclique%s' % clique_id
table_label = 'tclique%s' % clique_id
ecotype_id_ls = outputCliqueInLatexTable(curs, region, clique_id,
table_label, fig_label,
stock_db,
clique2ecotype_table,
outf)
#draw the clique
site_g, site2weight, site2pos = construct_site_graph_out_of_ecotype_id_ls(
ecotype_id_ls, ecotypeid2pos)
fig_fname = 'haplotype_%s_size%s_%ssites_map' % (
clique_id, len(ecotype_id_ls), len(site2pos)
) #just the name without filetype extension, like png, eps etc.
if need_to_draw_figures:
draw_graph_on_map(
site_g,
site2weight,
site2pos,
'clique %s, size %s, %s sites of %s' %
(clique_id, len(ecotype_id_ls), len(site2pos), region),
pic_area=get_pic_area(site2pos.values(), 30),
output_fname_prefix=os.path.join(fig_output_dir,
fig_fname))
#output the figure in latex
fig_fname = os.path.join(
'figures/', '%s.png' %
fig_fname) #relative path of the figure to the latex file
caption = 'map of haplotype clique %s has %s ecotypes from %s sites of %s. check Table~\\ref{%s}.' % (
clique_id, len(ecotype_id_ls), len(site2pos), region,
table_label)
outf.write(outputFigureInLatex(fig_fname, caption, fig_label))
clique_id2region[clique_id] = region
del site_g, site2pos, site2weight, ecotype_id_ls
outf.write("\\subsection{which component the clique is from}\n")
cc_id_clique_id_region_ls = []
cc_id2clique_id_ls = {}
for clique_id, region in clique_id2region.iteritems():
curs.execute("select cc_id from %s where clique_id=%s" %
(component2clique_table, clique_id))
rows = curs.fetchall()
cc_id = rows[0][0]
cc_id_clique_id_region_ls.append([cc_id, clique_id, region])
if cc_id not in cc_id2clique_id_ls:
cc_id2clique_id_ls[cc_id] = []
cc_id2clique_id_ls[cc_id].append(clique_id)
cc_id_clique_id_region_ls.sort()
caption = '%s components versus %s cliques' % (
len(cc_id2clique_id_ls), len(cc_id_clique_id_region_ls))
table_label = 't_cc2clique_1'
header_ls = ['component id', 'clique id', 'region']
outf.write(
outputMatrixInLatexTable(cc_id_clique_id_region_ls, caption,
table_label, header_ls))
del outf
