def main():
main_html = HtmlWriter('res/fba.html')
main_html.write('<h1>Flux Balance Analysis</h1>\n')
carbon_sources = {}
BM_lower_bound = 0.01
ko_reactions = ''
ki_reactions = ''
PPP_reaction = ''
# full model carbon sources: glc, fru, xyl__D, succ, ac, rib__D, pyr
#########################################
# Core model for testing glycolysis KOs #
#########################################
#model = init_wt_model('core', {'ac' : -10}); ko_reactions = 'PGM'; ki_reactions = 'RED';
#model = init_wt_model('core', {'glc' : -10}); ko_reactions = 'PFK'; ki_reactions = 'RED';
#######################################
# Full model Rubisco optimization KOs #
#######################################
#model = init_wt_model('full', {'ac' : -10}); ko_reactions = 'PGM,TRSARr,HPYRRx,HPYRRy'; ki_reactions = 'RED';
#model = init_wt_model('full', {'rib_D' : -10}); ko_reactions = 'TKT1'; ki_reactions = 'RBC,PRK'; PPP_reaction = 'RBC';
model = init_wt_model('full', {'xyl_D': -10})
ko_reactions = 'RPI'
ki_reactions = 'RBC,PRK'
PPP_reaction = 'RBC'
#model = init_wt_model('full', {'xyl_D' : -10}); ko_reactions = 'G6PDH2r,PFK,F6PA,FRUK,PFK_3,DRPA'; ki_reactions = 'RBC,PRK';
#model = init_wt_model('full', {'fru' : -10, 'rib_D' : -10}); ko_reactions = 'G6PDH2r,PFK,F6PA,FRUK,PFK_3,DRPA,TKT1,TKT2'; ki_reactions = 'PKT';
###############################
# Shikimate generating strain #
###############################
#model = init_wt_model('full', {'fru' : -10});
#ko_reactions = 'G6PDH2r,TALA'; ki_reactions = 'EX_3dhsk_c'; PPP_reaction = 'EX_3dhsk_c';
#ko_reactions = 'G6PDH2r,PFK,F6PA,FRUK,PFK_3,DRPA'; ki_reactions = 'PKT'; PPP_reaction = 'PKT';
##########################################################
# Testing no growth when electrons are provided directly #
##########################################################
#model = init_wt_model('full', {}); ki_reactions = 'RED';
#ko_reactions = "POR5,MCITL2";
#ko_reactions = "POR5,FTHFLi,GART,RPI";
models = {'WT': model}
if ko_reactions:
for k in models.keys():
m = deepcopy(models[k])
knockout_reactions(m, ko_reactions)
models[k + ' -%s' % ko_reactions] = m
if ki_reactions:
for k in models.keys():
m = deepcopy(models[k])
knockin_reactions(m, ki_reactions)
models[k + ' +%s' % ki_reactions] = m
# Run the optimization for the objective reaction and medium composition
# set in the file.
main_html.write('<table border="1">\n')
main_html.write(
'<tr><td><b>Model Name</b></td><td><b>Growth Yield</b></td></tr>\n')
growths = {}
for name, model in sorted(models.iteritems()):
print "solving %50s model" % name,
ok = OptKnock(model)
ok.prepare_FBA_primal()
ok.solve()
growths[name] = ok.get_objective_value()
if growths[name] is None:
main_html.write('<tr><td>%s</td><td>infeasible</td></tr>\n' % name)
else:
print ': f = %.3g' % growths[name]
main_html.write('<tr><td>')
html = main_html.branch(name)
main_html.write('</td><td>%.3g</td></tr>\n' % growths[name])
html.write('<h1>Model name: %s</h1>\n' % name)
html.write('<h2>Growth Yield: %.3g</h2>\n' % growths[name])
ok.draw_svg(html)
ok.model_summary(html)
main_html.write('</table>\n')
if PPP_reaction:
print 'Calculating Phenotypic Phase Plane for phosphoketolase ...'
fig, ax = plt.subplots(1, figsize=(6, 6))
plot_multi_PPP(models, PPP_reaction, ax)
ax.set_title('Phenotypic Phase Plane')
main_html.embed_matplotlib_figure(fig, width=400, height=400)
def main():
main_html = HtmlWriter('res/fba.html')
main_html.write('<h1>Flux Balance Analysis</h1>\n')
carbon_sources = {}
BM_lower_bound = 0.01
ko_reactions = ''
ki_reactions = ''
PPP_reaction = ''
# full model carbon sources: glc, fru, xyl__D, succ, ac, rib__D, pyr
#########################################
# Core model for testing glycolysis KOs #
#########################################
#model = init_wt_model('core', {'ac' : -10}); ko_reactions = 'PGM'; ki_reactions = 'RED';
#model = init_wt_model('core', {'glc' : -10}); ko_reactions = 'PFK'; ki_reactions = 'RED';
#######################################
# Full model Rubisco optimization KOs #
#######################################
#model = init_wt_model('full', {'ac' : -10}); ko_reactions = 'PGM,TRSARr,HPYRRx,HPYRRy'; ki_reactions = 'RED';
#model = init_wt_model('full', {'rib_D' : -10}); ko_reactions = 'TKT1'; ki_reactions = 'RBC,PRK'; PPP_reaction = 'RBC';
model = init_wt_model('full', {'xyl_D' : -10}); ko_reactions = 'RPI'; ki_reactions = 'RBC,PRK'; PPP_reaction = 'RBC';
#model = init_wt_model('full', {'xyl_D' : -10}); ko_reactions = 'G6PDH2r,PFK,F6PA,FRUK,PFK_3,DRPA'; ki_reactions = 'RBC,PRK';
#model = init_wt_model('full', {'fru' : -10, 'rib_D' : -10}); ko_reactions = 'G6PDH2r,PFK,F6PA,FRUK,PFK_3,DRPA,TKT1,TKT2'; ki_reactions = 'PKT';
###############################
# Shikimate generating strain #
###############################
#model = init_wt_model('full', {'fru' : -10});
#ko_reactions = 'G6PDH2r,TALA'; ki_reactions = 'EX_3dhsk_c'; PPP_reaction = 'EX_3dhsk_c';
#ko_reactions = 'G6PDH2r,PFK,F6PA,FRUK,PFK_3,DRPA'; ki_reactions = 'PKT'; PPP_reaction = 'PKT';
##########################################################
# Testing no growth when electrons are provided directly #
##########################################################
#model = init_wt_model('full', {}); ki_reactions = 'RED';
#ko_reactions = "POR5,MCITL2";
#ko_reactions = "POR5,FTHFLi,GART,RPI";
models = {'WT' : model}
if ko_reactions:
for k in models.keys():
m = deepcopy(models[k])
knockout_reactions(m, ko_reactions)
models[k + ' -%s' % ko_reactions] = m
if ki_reactions:
for k in models.keys():
m = deepcopy(models[k])
knockin_reactions(m, ki_reactions)
models[k + ' +%s' % ki_reactions] = m
# Run the optimization for the objective reaction and medium composition
# set in the file.
main_html.write('<table border="1">\n')
main_html.write('<tr><td><b>Model Name</b></td><td><b>Growth Yield</b></td></tr>\n')
growths = {}
for name, model in sorted(models.iteritems()):
print "solving %50s model" % name,
ok = OptKnock(model)
ok.prepare_FBA_primal()
ok.solve()
growths[name] = ok.get_objective_value()
if growths[name] is None:
main_html.write('<tr><td>%s</td><td>infeasible</td></tr>\n' % name)
else:
print ': f = %.3g' % growths[name]
main_html.write('<tr><td>')
html = main_html.branch(name)
main_html.write('</td><td>%.3g</td></tr>\n' % growths[name])
html.write('<h1>Model name: %s</h1>\n' % name)
html.write('<h2>Growth Yield: %.3g</h2>\n' % growths[name])
ok.draw_svg(html)
ok.model_summary(html)
main_html.write('</table>\n')
if PPP_reaction:
print 'Calculating Phenotypic Phase Plane for phosphoketolase ...'
fig, ax = plt.subplots(1, figsize=(6,6))
plot_multi_PPP(models, PPP_reaction, ax)
ax.set_title('Phenotypic Phase Plane')
main_html.embed_matplotlib_figure(fig, width=400, height=400)
