def run(self):
lens1 = base.newlens(self.lens1, 'order')
lens2 = base.newlens(self.lens2, 'order')
gff1 = base.newgff(self.gff1)
gff2 = base.newgff(self.gff2)
gff1 = gff1[gff1['chr'].isin(lens1.index)]
gff2 = gff2[gff2['chr'].isin(lens2.index)]
blast = base.newblast(self.blast, int(self.score), float(self.evalue),
gff1, gff2, self.blast_reverse)
blast = self.deal_blast(blast, gff1, gff2, int(self.multiple),
int(self.repeat_number))
blast['loc1'] = blast[0].map(gff1.loc[:, self.position])
blast['loc2'] = blast[1].map(gff2.loc[:, self.position])
blast['chr1'] = blast[0].map(gff1.loc[:, 'chr'])
blast['chr2'] = blast[1].map(gff2.loc[:, 'chr'])
total = []
for (chr1, chr2), group in blast.groupby(['chr1', 'chr2']):
df = pd.DataFrame(np.zeros((lens1[chr1], lens2[chr2])))
for index, row in group.iterrows():
df.loc[row['loc1'], row['loc2']] = row['grading']
df = df.loc[:, df.sum(axis=0) != 0]
df = df.loc[df.sum(axis=1) != 0, :]
collinearity = improvedcollinearity.collinearity(self.options, df)
data = collinearity.run()
fp = self.dir + '/' + str(chr1) + '.vs.' + str(chr2) + '.blk'
gf1, gf2 = gff1[gff1['chr'] == chr1], gff2[gff2['chr'] == chr1]
self.write_block(fp, data, chr1, chr2, gff1, gff2)
args = ['cat', self.dir + '/*.blk', '>', self.savefile]
command = ' '.join([str(k) for k in args])
os.system(command)
shutil.rmtree(self.dir)
sys.exit(0)
def run(self):
lens1 = base.newlens(self.lens1, 'order')
lens2 = base.newlens(self.lens2, 'order')
lens1 = lens1[lens1 > 4]
lens2 = lens2[lens2 > 4]
gff1 = base.newgff(self.gff1)
gff2 = base.newgff(self.gff2)
gff1 = gff1[gff1['chr'].isin(lens1.index)]
gff2 = gff2[gff2['chr'].isin(lens2.index)]
blast = base.newblast(self.blast, int(self.score), float(self.evalue),
gff1, gff2)
df = self.deal_blast(blast, gff1, gff2, int(self.repnum))
for (chr1, chr2), group in df.groupby(['chr1', 'chr2']):
group = group.sort_values(by=['loc1', 'loc2'])
dir1 = './' + self.dir + '/pair/' + str(chr1) + '.vs.' + str(
chr2) + '.pair'
dir2 = './' + self.dir + '/block/' + str(chr1) + '.vs.' + str(
chr2) + '.blk'
group[[0, 'stand1', 'loc1', 1, 'stand2',
'loc2']].to_csv(dir1, sep=' ', index=None, header=None)
args = [
'blockscan', '-chr1len', lens1[str(chr1)], '-chr2len',
lens2[str(chr2)], '-mg1', self.mg[0], '-mg2', self.mg[1], dir1,
'>' + dir2
]
command = ' '.join([str(k) for k in args])
os.system(command)
args = [
'cat', self.dir + '/block/*.blk', '>', self.dir + '.block.old.txt'
]
command = ' '.join([str(k) for k in args])
os.system(command)
self.rewriteblock(blast, self.dir + '.block.old.txt',
self.dir + '.block.txt')
def run(self):
lens1 = base.newlens(self.lens1, self.position)
lens2 = base.newlens(self.lens2, self.position)
if re.search('\d', self.figsize):
self.figsize = [float(k) for k in self.figsize.split(',')]
else:
self.figsize = np.array(
[1, float(lens1.sum()) / float(lens2.sum())]) * 10
plt.rcParams['ytick.major.pad'] = 0
fig, ax = plt.subplots(figsize=self.figsize)
ax.xaxis.set_ticks_position('top')
step1 = 1 / float(lens1.sum())
step2 = 1 / float(lens2.sum())
base.dotplot_frame(fig, ax, lens1, lens2, step1, step2,
self.genome1_name, self.genome2_name)
gff1 = base.newgff(self.gff1)
gff2 = base.newgff(self.gff2)
gff1 = base.gene_location(gff1, lens1, step1, self.position)
gff2 = base.gene_location(gff2, lens2, step2, self.position)
blast = base.newblast(self.blast, int(self.score), float(self.evalue),
gff1, gff2)
df = self.pair_positon(blast, gff1, gff2, int(self.multiple),
int(self.repnum))
plt.scatter(df['loc2'],
df['loc1'],
s=float(self.markersize),
c=df['color'],
alpha=0.5,
edgecolors=None,
linewidths=0,
marker='o')
plt.subplots_adjust(left=0.07, right=0.97, top=0.93, bottom=0.03)
plt.savefig(self.savefile, dpi=500)
sys.exit(0)
def run(self):
bkinfo = pd.read_csv(self.blockinfo, index_col='id')
bkinfo['chr1'] = bkinfo['chr1'].astype(str)
bkinfo['chr2'] = bkinfo['chr2'].astype(str)
if self.blockinfo_reverse == True or self.blockinfo_reverse.upper(
) == 'TRUE':
bkinfo[['chr1', 'chr2']] = bkinfo[['chr2', 'chr1']]
bkinfo[['block1', 'block2']] = bkinfo[['block2', 'block1']]
bkinfo = bkinfo[bkinfo['length'] > int(self.block_length)]
bkinfo['class1'] = ''
bkinfo['col1'] = ''
bkinfo['class2'] = ''
bkinfo['col2'] = ''
gff1 = base.newgff(self.gff1)
gff2 = base.newgff(self.gff2)
lens = base.newlens(self.the_other_lens, self.position)
blast = base.newblast(self.blast, int(self.score), float(self.evalue),
gff1, gff2, self.blast_reverse)
index = [
group.sort_values(
by=11,
ascending=False)[:int(self.repeat_number)].index.tolist()
for name, group in blast.groupby([0])
]
blast = blast.loc[np.concatenate(
np.
array([k[:int(self.repeat_number)] for k in index], dtype=object)),
[0, 1]]
pairs = self.colinear_gene_pairs(bkinfo, gff1, gff2)
data = []
for k, v in lens.items():
for i in range(1, v + 1):
data.append([k, i, 0])
df = pd.DataFrame(data)
gff1['color'] = np.nan
gff2['color'] = np.nan
gff1['classification'] = np.nan
gff2['classification'] = np.nan
if hasattr(self, 'ancestor_top'):
ancestor = base.read_calassfication(self.ancestor_top)
data = self.karyotype_top(pairs, ancestor, gff1, gff2)
if hasattr(self, 'ancestor_left'):
ancestor = base.read_calassfication(self.ancestor_left)
data = self.karyotype_left(pairs, ancestor, gff1, gff2)
the_other_ancestor_file = self.karyotype_map(data, lens)
the_other_ancestor_file = self.new_ancestor(the_other_ancestor_file,
gff1, gff2, blast)
the_other_ancestor_file.to_csv(self.the_other_ancestor_file,
sep='\t',
header=False,
index=False)
def run(self):
lens1 = base.newlens(self.lens1, self.position)
lens2 = base.newlens(self.lens2, self.position)
gff1 = base.newgff(self.gff1)
gff2 = base.newgff(self.gff2)
gff1 = gff1[gff1['chr'].isin(lens1.index)]
gff2 = gff2[gff2['chr'].isin(lens2.index)]
blast = base.newblast(self.blast, int(self.score), float(self.evalue),
gff1, gff2)
blast = self.blast_homo(blast, gff1, gff2, int(self.repnum))
blast.index = blast[0] + ',' + blast[1]
colinearity = base.read_colinearscan(self.colinearity)
ks = base.read_ks(self.ks, self.ks_col)
data = self.block_position(colinearity, blast, gff1, gff2, ks)
def run(self):
lens1 = base.newlens(self.lens1, 'order')
lens2 = base.newlens(self.lens2, 'order')
gff1 = base.newgff(self.gff1)
gff2 = base.newgff(self.gff2)
gff1 = gff1[gff1['chr'].isin(lens1.index)]
gff2 = gff2[gff2['chr'].isin(lens2.index)]
blast = base.newblast(self.blast, int(self.score), float(self.evalue),
gff1, gff2, self.blast_reverse)
blast = self.deal_blast(blast, gff1, gff2, int(self.multiple),
int(self.repeat_number))
blast['loc1'] = blast[0].map(gff1.loc[:, self.position])
blast['loc2'] = blast[1].map(gff2.loc[:, self.position])
blast['chr1'] = blast[0].map(gff1.loc[:, 'chr'])
blast['chr2'] = blast[1].map(gff2.loc[:, 'chr'])
total = []
for (chr1, chr2), group in blast.groupby(['chr1', 'chr2']):
total.append([chr1, chr2, group])
del blast, group
gc.collect()
n = int(np.ceil(len(total) / float(self.process)))
result, data = '', []
try:
pool = Pool(self.process)
for i in range(0, len(total), n):
data.append(
pool.apply_async(self.single_pool,
args=(total[i:i + n], gff1, gff2, lens1,
lens2)))
pool.close()
pool.join()
except:
pool.terminate()
for k in data:
result += k.get()
result = re.split('\n', result)
fout = open(self.savefile, 'w')
num = 1
for line in result:
if re.match(r"# Alignment", line):
s = '# Alignment ' + str(num) + ':'
fout.writelines(s + line.split(':')[1] + '\n')
num += 1
continue
if len(line) > 0:
fout.writelines(line + '\n')
fout.close()
sys.exit(0)
def run(self):
lens1 = base.newlens(self.lens1, self.position)
lens2 = base.newlens(self.lens2, self.position)
gff1 = base.newgff(self.gff1)
gff2 = base.newgff(self.gff2)
gff1 = gff1[gff1['chr'].isin(lens1.index)]
gff2 = gff2[gff2['chr'].isin(lens2.index)]
blast = base.newblast(self.blast, int(self.score), float(self.evalue),
gff1, gff2, self.blast_reverse)
blast = self.blast_homo(blast, gff1, gff2, int(self.repeat_number))
blast.index = blast[0] + ',' + blast[1]
collinearity = self.auto_file(gff1, gff2)
ks = base.read_ks(self.ks, self.ks_col)
data = self.block_position(collinearity, blast, gff1, gff2, ks)
data['class1'] = 0
data['class2'] = 0
data.to_csv(self.savefile, index=None)
def run(self):
length = 1
axis = [0, 1, 1, 0]
left, right, top, bottom = 0.07, 0.97, 0.93, 0.03
lens1 = base.newlens(self.lens1, self.position)
lens2 = base.newlens(self.lens2, self.position)
step1 = 1 / float(lens1.sum())
step2 = 1 / float(lens2.sum())
if self.ancestor_left != None:
axis[0] = -0.02
lens_ancestor_left = pd.read_csv(self.ancestor_left,
sep="\t",
header=None)
lens_ancestor_left[0] = lens_ancestor_left[0].astype(str)
lens_ancestor_left[3] = lens_ancestor_left[3].astype(str)
lens_ancestor_left[4] = lens_ancestor_left[4].astype(int)
lens_ancestor_left[
4] = lens_ancestor_left[4] / lens_ancestor_left[4].max()
lens_ancestor_left = lens_ancestor_left[lens_ancestor_left[0].isin(
lens1.index)]
if self.ancestor_top != None:
axis[3] = -0.02
lens_ancestor_top = pd.read_csv(self.ancestor_top,
sep="\t",
header=None)
lens_ancestor_top[0] = lens_ancestor_top[0].astype(str)
lens_ancestor_top[3] = lens_ancestor_top[3].astype(str)
lens_ancestor_top[4] = lens_ancestor_top[4].astype(int)
lens_ancestor_top[
4] = lens_ancestor_top[4] / lens_ancestor_top[4].max()
lens_ancestor_top = lens_ancestor_top[lens_ancestor_top[0].isin(
lens2.index)]
if re.search('\d', self.figsize):
self.figsize = [float(k) for k in self.figsize.split(',')]
else:
self.figsize = np.array(
[1, float(lens1.sum()) / float(lens2.sum())]) * 10
plt.rcParams['ytick.major.pad'] = 0
fig, ax = plt.subplots(figsize=self.figsize)
ax.xaxis.set_ticks_position('top')
base.dotplot_frame(fig, ax, lens1, lens2, step1, step2,
self.genome1_name, self.genome2_name,
[axis[0], axis[3]])
gff1 = base.newgff(self.gff1)
gff2 = base.newgff(self.gff2)
gff1 = base.gene_location(gff1, lens1, step1, self.position)
gff2 = base.gene_location(gff2, lens2, step2, self.position)
if self.ancestor_top != None:
top = top
self.ancestor_posion(ax, gff2, lens_ancestor_top, 'top')
if self.ancestor_left != None:
left = left
self.ancestor_posion(ax, gff1, lens_ancestor_left, 'left')
blast = base.newblast(self.blast, int(self.score), float(self.evalue),
gff1, gff2, self.blast_reverse)
df = self.pair_positon(blast, gff1, gff2, int(self.multiple),
int(self.repeat_number))
ax.scatter(df['loc2'],
df['loc1'],
s=float(self.markersize),
c=df['color'],
alpha=0.5,
edgecolors=None,
linewidths=0,
marker='o')
ax.axis(axis)
plt.subplots_adjust(left=left, right=right, top=top, bottom=bottom)
plt.savefig(self.savefig, dpi=500)
plt.show()
sys.exit(0)