def get_out_search():
with open(sys.argv[-1]) as wdsp_f:
CUTOFF = 20
wdsp = Wdsp(wdsp_f)
pros,seqs,wdsps,hotspots = wdsp.pros,wdsp.seqs,wdsp.wdsps,wdsp.hotspots
best = {}
pro_num = 1000000
while pro_num > CUTOFF:
a = PatchSearchSpecific(pros,seqs,wdsps,hotspots,CUTOFF)
a.get_patches()
a.classify_patches()
shape,patch,pro_list,pro_num = a.get_best()
if not shape in best.keys():
best[shape] = {}
best[shape][patch] = pro_list
else:
best[shape][patch] = pro_list
for pro in pro_list:
if pro in seqs.keys():
pros.pop(pros.index(pro))
seqs.pop(pro)
wdsps.pop(pro)
hotspots.pop(pro)
with open(sys.argv[-1]) as wdsp_f:
wdsp = Wdsp(wdsp_f)
write_results(best,wdsp.pros,wdsp.seqs,wdsp.wdsps,wdsp.hotspots,CUTOFF)
def main():
with open(sys.argv[-2]) as wdsp_f:
tem_wdsp = Wdsp(wdsp_f)
tem_hots = tem_wdsp.hotspots
with open(sys.argv[-1]) as wdsp_f:
all_wdsp = Wdsp(wdsp_f)
all_hots = all_wdsp.hotspots
tem_all_hots = get_similar_hots(tem_hots, all_hots)
write_result(tem_all_hots, tem_hots, all_hots)
def main():
with open(sys.argv[-1]) as wdsp_f:
all_wdsp = Wdsp(wdsp_f)
all_hots = all_wdsp.hotspots
all_seqs = all_wdsp.seqs
clusters = cluster_topface(all_hots)
regressions = []
for pros in clusters:
c_size = len(pros)
if c_size > 10:
filename = str(c_size) + '_' + pros[0] + '_' + str(cutoff)
hots = [[pro, all_hots[pro]] for pro in pros]
seqs = [[pro, all_wdsp.seqs[pro]] for pro in pros]
hots_score = align_hots([hot[1] for hot in hots])
seqs_score = align_seqs([seq[1] for seq in seqs])
regressions.append(linregress(seqs_score, hots_score))
plot_scatter(seqs_score, hots_score, filename + '_scatter')
hots = adjust_hots(hots)
hots = [(pro, ''.join(hot)) for pro, hot in hots]
plotlogo(hots, filename + '_logo')
with open('regressions.txt', 'w') as w_f:
'slop,intercept,r-value,p-value,stderr'
for r in regressions:
print >> w_f, ';'.join(map(str, r))
def main():
with open('test.wdsp') as wdsp_f:
wdsp = Wdsp(wdsp_f)
a = PatchSearch(wdsp.pros,wdsp.seqs,wdsp.wdsps,wdsp.hotspots,1)
a.get_patches()
a.classify_patches()
a.write_results()
def main():
with open(sys.argv[1]) as wdsp_f:
CUTOFF = 20
wdsp = Wdsp(wdsp_f)
a = PatchSearch(wdsp.pros,wdsp.seqs,wdsp.wdsps,wdsp.hotspots,CUTOFF)
a.get_patches()
a.classify_patches()
write_results(a.shape_patch_pros,wdsp.pros,wdsp.seqs,wdsp.wdsps,wdsp.hotspots,CUTOFF)
def main():
with open(sys.argv[-2]) as wdsp_f:
tem_wdsp = Wdsp(wdsp_f)
tem_hots = tem_wdsp.hotspots
with open(sys.argv[-1]) as wdsp_f:
all_wdsp = Wdsp(wdsp_f)
all_hots = all_wdsp.hotspots
cutoff = 0.0
for cutoff in [0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]:
# for cutoff in [30,40,50,60,70,80,90]:
tem_all_hots = get_similar_hots(tem_hots, all_hots, cutoff)
for tem_pro, all_pros in tem_all_hots.iteritems():
hots = [[pro, all_hots[pro]] for pro, _ in all_pros]
seqs = [[pro, all_wdsp.seqs[pro]] for pro, _ in all_pros]
hots_score = get_hot_similarity(hots)
seqs_score = get_seq_similarity(seqs)
plot_scatter(seqs_score, hots_score, tem_pro + '_' + str(cutoff))
hots = adjust_hots(hots)
hots_len = len(hots)
hots = [(pro, ''.join(hot)) for pro, hot in hots]
plotlogo(hots, str(hots_len) + '_' + tem_pro + '_' + str(cutoff))
f, ax = plt.subplots()
fig = plt.figure(figsize=(5, 4))
ax = fig.add_subplot(111)
# sns.distplot(hots_score,hist=False,label='Topface',kde_kws={'linestyle':'-.'})
# sns.distplot(seqs_score,hist=False,label='Sequence',kde_kws={'linestyle':'--'})
sns.distplot(hots_score,
hist=False,
label='Topface',
kde_kws={'marker': ' '})
sns.distplot(seqs_score,
hist=False,
label='Sequence',
kde_kws={'marker': '*'})
ax.set(xlabel='Similarity',
ylabel='Frequency',
title='WD40 Protein Topface and Sequence Similarity')
# h.figure.subplots_adjust(top=0.9,bottom=0.05,left=0.18,right=0.98)
plt.savefig(tem_pro + '_' + str(cutoff) +
'hot_seq_similarity_dist.png',
dpi=300)
def main():
with open(sys.argv[1]) as wdsp_f:
CUTOFF = 10
wdsp = Wdsp(wdsp_f)
a = PatchSearchSpecific(wdsp.pros,wdsp.seqs,wdsp.wdsps,wdsp.hotspots,CUTOFF)
a.get_patches()
a.classify_patches()
# a.deredundant_patches()
a.write_results()
def main():
with open(sys.argv[-1]) as o_f:
w = Wdsp(o_f)
sims = OrderedDict()
for pro, repeats in w.repeats.iteritems():
sims[pro] = repeat_similarity(repeats)
with open('sims.txt', 'w') as w_f:
for k, v in sims.iteritems():
print >> w_f, '{0:<20}{1:<}'.format(k, v[0])
def main():
with open(sys.argv[-2]) as wdsp_f:
tem_wdsp = Wdsp(wdsp_f)
tem_hots = tem_wdsp.hotspots
tem_repeats_similarity = wdsp_repeat_similarity(tem_wdsp.repeats)
with open(sys.argv[-1]) as wdsp_f:
all_wdsp = Wdsp(wdsp_f)
all_hots = all_wdsp.hotspots
all_repeats_similarity = wdsp_repeat_similarity(all_wdsp.repeats)
tem_all_seq_similarity = seq_similarity(tem_wdsp, all_wdsp)
cutoff = 70
# for cutoff in [30,40,50,60,70,80,90]:
for cutoff in [10]:
tem_all_hots = get_similar_hots(tem_hots, all_hots, cutoff)
write_result(tem_all_hots, tem_hots, tem_wdsp, all_hots, all_wdsp,
tem_repeats_similarity, all_repeats_similarity,
tem_all_seq_similarity, cutoff)
def main():
with open(sys.argv[-2]) as o_f:
tem = Wdsp(o_f)
tem_seq = tem.seqs
with open(sys.argv[-1]) as o_f:
all1 = Wdsp(o_f)
all_seq = all1.seqs
similarity = OrderedDict()
for t_name, t_seq in tem_seq.iteritems():
sim = []
for a_name, a_seq in all_seq.iteritems():
sim.append((a_name, align(t_seq, a_seq)))
# sim = sorted(sim, key=operator.itemgetter(1),reverse=True)
similarity[t_name] = sim
for k, v in similarity.iteritems():
with lt.open_file(k) as w_f:
for a_name, a_identity in v:
print >> w_f, '{0:<15}{1:<}'.format(a_name, a_identity)
def main():
with open(sys.argv[-1]) as wdsp_f:
wdsp = Wdsp(wdsp_f)
a = PatchSearchSpecific(wdsp.pros,
wdsp.seqs,
wdsp.wdsps,
wdsp.hotspots,
cutoff=1)
a.get_patches()
a.classify_patches()
a.write_results()
def main():
with open('test.wdsp') as wdsp_f:
wdsp = Wdsp(wdsp_f)
a = PatchSearch(wdsp.pros,wdsp.seqs,wdsp.wdsps,wdsp.hotspots,2)
a.get_patches()
a.classify_patches()
write_results(a.shape_patch_pros,wdsp.pros,wdsp.seqs,wdsp.wdsps,wdsp.hotspots)
print 'shape_pro_patches'
print a.shape_pro_patches
print 'shape_patch_pros'
print a.shape_patch_pros
def classify_blade(wdsp_f):
with open(wdsp_f) as o_f:
wdsp = Wdsp(o_f)
hotspots = ' '.join(
[' '.join(v) for k, v in wdsp.hotspots.iteritems()]).split()
aa = {
'K', 'R', 'H', 'D', 'E', 'F', 'W', 'Y', 'S', 'T', 'N', 'Q', 'V',
'L', 'I', 'M', 'A', 'C', 'P', 'G', '*', 'X', 'B'
}
aa_combi = itertools.product(aa, repeat=3)
blades = {}
for c in aa_combi:
blades[''.join(c)] = 0
for hot in hotspots:
blades[hot] += 1
blades = [(v, k) for k, v in blades.iteritems()]
blades = sorted(blades, reverse=True)
return blades
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
select wd40s with repeats 6n,7n,8n
"""
from wdsp import Wdsp
import numpy as np
from numpy.random import randint
with open('wd648_uniprot_select_cd-hit_90_7.wdsp') as wdsp_f:
wdsp = Wdsp(wdsp_f)
pro_num = len(wdsp.pros)
for i in range(100):
with open('random_' + str(i) + '.fa', 'w') as w_f:
for j in range(2):
pro = wdsp.pros[randint(0, pro_num)]
seq = wdsp.seqs[pro]
print >> w_f, '> ', pro
for s in [seq[i:i + 80] for i in range(0, len(seq), 80)]:
print >> w_f, s
def read_hots(wdsp_f):
with open(sys.argv[-2]) as wdsp_f:
wdsp = Wdsp(wdsp_f)
hots = wdsp.hotspots
return hots
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
usage: python get_hotspot.py *.wdsp
output hotspots in following format
pro xxx xxx xxx xxx xxx xxx
"""
import os
import sys
import lt
from wdsp import Wdsp
with open(sys.argv[-1]) as wdsp_f:
w = Wdsp(wdsp_f)
with lt.open_file(file_suffix='hotspots') as w_f:
for pro, hots in w.hotspots.iteritems():
print >> w_f, '{0:<25}{1:<}'.format(pro, ' '.join(hots))
# visual_style['vertex_label'] = labels
# visual_style['vertex_label_size'] = 2
visual_style['layout'] = graph.layout('kk')
filename = 'sim_top_seq_' + str(cluster_num) + '_' + str(label)
igraph.plot(graph, filename + '_graph.png', **visual_style)
# slope,intercept,rvalue,pvalue,stderr = linregress(seqs_score,hots_score)
# return [slope,intercept,rvalue,pvalue,stderr]
# plot_scatter(seqs_score,hots_score,filename+'_scatter')
# hots = adjust_hots(nr_hots)
# hots = [(pro,''.join(hot)) for pro,hot in hots]
# plotlogo(hots,filename+'_logo')
with open(sys.argv[-1]) as wdsp_f:
all_wdsp = Wdsp(wdsp_f)
all_hots = all_wdsp.hotspots
all_seqs = all_wdsp.seqs
pros = all_wdsp.pros
import lt
@lt.run_time
def main():
for cluster_num in range(10, 100, 50):
clusters = []
for i in range(10):
clusters.append(
[[pros[randint(0, len(pros))] for i in range(cluster_num)], i])
def main():
fname = os.path.split(sys.argv[-1])[1].split('.')[0]
with open(sys.argv[-1]) as wdsp_f:
wdsp = Wdsp(wdsp_f)
pros = wdsp.pros
hots = wdsp.hotspots
seqs = wdsp.seqs
parameters = []
for i1, pro1 in enumerate(pros):
for i2, pro2 in enumerate(pros):
if i2 > i1:
parameters.append([
pro1, pro2, hots[pro1], hots[pro2], seqs[pro1],
seqs[pro2]
])
p = Pool(6)
result = p.map(top_seq_align, parameters)
p.close()
# # result = []
# # for p in parameters:
# # r = top_seq_align(p)
# # result.append(r)
hots_score = [r[2][1] for r in result]
seqs_score = [r[3][1] for r in result]
pickle.dump([hots_score, seqs_score],
open('hots_seqs_score.pickle', 'w'))
hots_score, seqs_score = pickle.load(open('hots_seqs_score.pickle'))
plot_scatter(seqs_score, hots_score, fname + '_scatter')
hots = [[pro, hot] for pro, hot in hots.iteritems()]
hots = adjust_hots(hots)
hots = [(pro, ''.join(hot)) for pro, hot in hots]
plotlogo(hots, fname + '_logo')
regression = linregress(seqs_score, hots_score)
with open(fname + '_regression.txt', 'w') as w_f:
'slop,intercept,r-value,p-value,stderr'
print >> w_f, ';'.join(map(str, regression))
f, ax = plt.subplots()
fig = plt.figure(figsize=(5, 4))
ax = fig.add_subplot(111)
# sns.distplot(hots_score,hist=False,label='Topface',kde_kws={'linestyle':'-.'})
# sns.distplot(seqs_score,hist=False,label='Sequence',kde_kws={'linestyle':'--'})
sns.distplot(hots_score,
hist=False,
label='Topface',
kde_kws={'marker': ' '})
sns.distplot(seqs_score,
hist=False,
label='Sequence',
kde_kws={'marker': '*'})
ax.set(xlabel='Similarity',
ylabel='Frequency',
title='WD40 Protein Topface and Sequence Similarity')
# h.figure.subplots_adjust(top=0.9,bottom=0.05,left=0.18,right=0.98)
plt.savefig(fname + 'hot_seq_similarity_dist.png', dpi=300)
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
calculate statistics for WDSP output file
usage: python wdsp_sta.py wdsp_f
"""
import lt
import sys
import os
from wdsp import Wdsp
with open(sys.argv[-1]) as o_f:
wdsp = Wdsp(o_f)
scores_sta = lt.lis_sta(wdsp.scores.values())
with lt.open_file(file_suffix='total_score_sta') as w_f:
for num, freq in scores_sta:
print >> w_f, '{0:<10}{1}'.format(num, freq)
tetrad_sta = [
len([vi for vi in v if vi >= 44.0])
for k, v in wdsp.blade_scores.iteritems()
]
tetrad_sta = lt.lis_sta(tetrad_sta)
with lt.open_file(file_suffix='tetrad_num_sta') as w_f:
for num, freq in tetrad_sta:
print >> w_f, '{0:<5}{1}'.format(num, freq)
blades_sta = [len(blades) for pro, blades in wdsp.blades.iteritems()]
blades_sta = lt.lis_sta(blades_sta)
