def __init__(self, sj, me, filepre, depth=500, maxcnt=10000):
MEGraph3.__init__(self, sj, me, depth, maxcnt)
self.pre = filepre
a = filepre + 'ex1.txt.gz'
b = filepre + 'ex2.txt.gz'
c = filepre + 'ov.txt.gz'
# calculate exon overlap to self
cols0 = ['chr', 'st', 'ed', 'strand', '_id']
# single cell data contains float in st,ed in ex ???
me = UT.check_int_nan(me)
a = UT.write_pandas(me[cols0], a, '')
b = UT.write_pandas(me[cols0], b, '')
c = BT.bedtoolintersect(a, b, c, wao=True)
cols1 = cols0 + ['b_' + x for x in cols0] + ['ovl']
self.ov = ov = UT.read_pandas(c, names=cols1)
# select same strand overlap to non-self
self.ov1 = ov1 = ov[(ov['_id'] != ov['b__id'])
& (ov['strand'] == ov['b_strand'])]
# make connected dictionary _id => [b__id's]
tmp = ov1.groupby('_id')['b__id'].apply(
lambda x: list(x)).reset_index()
if 'index' in tmp.columns:
tmp['_id'] = tmp['index']
#LOG.debug('graph.MEGraph4.__init__: tmp.columns={0}, len(tmp)={1}'.format(tmp.columns, len(tmp)))
self.eoe = dict(UT.izipcols(tmp, ['_id', 'b__id']))
# cleanup
os.unlink(a)
os.unlink(b)
os.unlink(c)
def __init__(self, mg, se=None):
self.mg = mg
self.ex = ex = mg.exons.set_index('_id')
#self.genes = genes = ex.groupby('_gidx')['_id'].groups # _gidx => [_id] dict
#WARNING above does not give the desired dict it maps to index instead of _id
self.genes = ex.groupby('_gidx').groups
self.se = se
self.i2g = dict(UT.izipcols(mg.exons, ['_id','gname']))
def calc_ecov_mp(covci, fname, np, blocksize=100):
"""
WARNING: this assumes _id is assinged according to sorted (chr,st,ed)
"""
LOG.debug('calc_ecov...')
chroms = sorted(covci['chr'].unique())
if 'name1' not in covci.columns:
covci['name1'] = covci['name'].astype(str).apply(
lambda x: [int(y) for y in x.split(',')])
if 'eidmax' not in covci.columns:
covci['eidmax'] = covci['name1'].apply(lambda x: max(x))
if 'eidmin' not in covci.columns:
covci['eidmin'] = covci['name1'].apply(lambda x: min(x))
args = [(covci[covci['chr'] == c].copy(), blocksize) for c in chroms]
e2cs = {}
if np == 1:
# for c,bwname,chrom,d in data:
for arg in args:
e2cs.update(calc_ecov_chrom(*arg))
else:
try:
p = multiprocessing.Pool(np)
rslts = p.map(mp_worker, zip(repeat(calc_ecov_chrom), args))
finally:
LOG.debug('closing pool')
p.close()
for x in rslts:
e2cs.update(x)
LOG.debug('writing rslts...')
if fname is None:
return e2cs
ccf = UT.flattendf(covci, 'name1')
ccfg = ccf.groupby('name1')
e2chr = dict(
UT.izipcols(ccfg['chr'].first().reset_index(), ['name1', 'chr']))
e2st = dict(UT.izipcols(ccfg['st'].min().reset_index(), ['name1', 'st']))
e2ed = dict(UT.izipcols(ccfg['ed'].max().reset_index(), ['name1', 'ed']))
df = PD.DataFrame(e2cs, index=['ecov']).T
df.index.name = 'eid'
df = df.reset_index()
df['chr'] = [e2chr[x] for x in df['eid']]
df['st'] = [e2st[x] for x in df['eid']]
df['ed'] = [e2ed[x] for x in df['eid']]
UT.save_tsv_nidx_whead(df[['eid', 'chr', 'st', 'ed', 'ecov']], fname)
return df
def _gen():
for x in UT.izipcols(df, cols):
rec = [x[0], 0, 0, x[3], x[4], x[5], x[-1]]
bsizes = [int(y) for y in x[-3].split(',')]
bstarts = [int(y) for y in x[-2].split(',')]
for y, z in zip(bstarts, bsizes):
if y >= 0:
rec[1] = x[1] + y
rec[2] = x[1] + y + z
yield rec.copy()
def __init__(self, sjexpre, th=0.1):
self.sjexpre = sjexpre
self.th = th
self.sj = sj = UT.read_pandas(sjexpre+'.sj.txt.gz')
self.ex = ex = UT.read_pandas(sjexpre+'.ex.txt.gz')
self.mg = mg = GP.MEGraph3(sj,ex) # only consider splice junction connections
self.exg = ex.set_index('_gidx')
self.exi = ex.set_index('_id')
self.nullidx = UT.find_nullidx(self.ex)
self.e2c = dict(UT.izipcols(ex, ['_id', 'cat']))
self.precalc_branch_p()
def _egen():
for chrom,tname,strand,est,eed in UT.izipcols(kg,['chr','name','strand','_ests','_eeds']):
if len(est)==1:
yield (chrom,st,ed,tname,0,strand,'s')
else:
if strand=='+':
yield (chrom,est[0],eed[0],tname,0,strand,'5')
for st,ed in izip(est[1:-1],eed[1:-1]):
yield (chrom,st,ed,tname,0,strand,'i')
yield (chrom,est[-1],eed[-1],tname,0,strand,'3')
else: #'-'
yield (chrom,est[0],eed[0],tname,0,strand,'3')
for st,ed in izip(est[1:-1],eed[1:-1]):
yield (chrom,st,ed,tname,0,strand,'i')
yield (chrom,est[-1],eed[-1],tname,0,strand,'5')
def calc_glen(ex, cipath):
ci = GGB.read_bed(cipath) # 5 col bed, name:eids, sc1:cid
ci['len'] = ci['ed'] - ci['st']
ci['cid'] = ci['sc1']
c2l = dict(UT.izipcols(ci, ['cid', 'len']))
if 'cid' not in ex.columns:
e2c = {}
for i, name in ci[['cid', 'name']].values:
for eid in name.split(','):
e2c.setdefault(int(eid), []).append(i)
ex['cid'] = [e2c[x] for x in ex['_id']]
def _gen():
for g, cids in UT.izipcols(ex, ['_gidx', 'cid']):
for c in cids:
yield (c, g)
df = PD.DataFrame(list(set([x for x in _gen()])), columns=['cid', '_gidx'])
df['len'] = [c2l[x] for x in df['cid']]
glen = df.groupby('_gidx')['len'].sum()
return dict(zip(glen.index, glen.values))
def _gen():
for g, cids in UT.izipcols(ex, ['_gidx', 'cid']):
for c in cids:
yield (c, g)
def _set_df2prop(src, tgt, default):
dic = dict(UT.izipcols(df2, ['_gidx', src]))
df[tgt] = [dic.get(x, default) for x in df['_gidx']]
def calc_gcov(expath, cipath, bwpath, dstprefix, override=False, np=4):
"""Calculate gene coverages.
Args:
expath: merged ex
cipath: chopped interval for ex
bwpath: bigwig file (sample)
dstprefix: prefix for outputs
Outputs:
1. dstprefix+'.covci.txt.gz'
2. dstprefix+'.gcov.txt.gz' : DataFrame(col:_gidx,len,val,gcov,len2,gcov2,cids)
len2: calculate length from ci with cov > 0
(normal length = use entire ci's belonging to the gene)
gcov2 = val/len2
cids: cid with cov > for the gene ','.joined
"""
ex = UT.read_pandas(expath)
covcipath = dstprefix + 'covci.txt.gz'
gcovpath = dstprefix + 'gcov.txt.gz'
if UT.notstale([expath, cipath], covcipath, override):
cc = UT.read_pandas(covcipath)
else:
if UT.notstale(expath, cipath, False):
ci = UT.read_pandas(cipath,
names=['chr', 'st', 'ed', 'name', 'id'])
else:
ci = UT.chopintervals(ex, cipath, idcol='_id')
cc = calc_cov_mp(ci, bwpath, covcipath, np=np)
# if override or (not os.path.exists(covcipath)):
# # calc covci
# if not os.path.exists(cipath):
# ci = UT.chopintervals(ex, cipath, idcol='_id')
# else:
# ci = UT.read_pandas(cipath, names=['chr','st','ed','name','id'])
# cc = calc_cov_mp(ci, bwpath, covcipath, np=np)
# else:
# cc = UT.read_pandas(covcipath)
if 'id' not in cc.columns:
cc['id'] = cc['sc1']
if 'eid' not in cc.columns:
cc['eid'] = cc['name'].astype(str).apply(
lambda x: [int(y) for y in x.split(',')])
cc['len'] = cc['ed'] - cc['st']
cc['val'] = cc['cov'] * cc['len']
ccf = UT.flattendf(cc[['id', 'eid', 'len', 'val', 'st', 'ed']], 'eid')
e2g = dict(UT.izipcols(ex, ['_id', '_gidx']))
ccf['_gidx'] = [e2g[x] for x in ccf['eid']]
# for normal gcov: take unique combination of (gid, id) (id=cid)
# for gocv2 : first select ccf with val>0
ccf2 = ccf[ccf['val'] > 0].groupby(['_gidx', 'id']).first().reset_index()
ccf2g = ccf2.groupby('_gidx')
df2 = ccf2g[['len', 'val']].sum()
df2['gcov2'] = df2['val'] / df2['len']
df2['cids'] = ccf2g['id'].apply(lambda x: ','.join([str(y) for y in x]))
df2['gst2'] = ccf2g['st'].min()
df2['ged2'] = ccf2g['ed'].max()
df2['glen2'] = df2['ged2'] - df2['gst2']
df2 = df2.reset_index()
ccf1 = ccf.groupby(['_gidx', 'id']).first().reset_index()
ccf1g = ccf1.groupby('_gidx')
df = ccf1g[['len', 'val']].sum()
df['gcov'] = df['val'] / df['len']
df['st'] = ccf1g['st'].min()
df['ed'] = ccf1g['ed'].max()
df['glen'] = df['ed'] - df['st']
df = df.reset_index()
g2chr = dict(UT.izipcols(ex, ['_gidx', 'chr']))
df['chr'] = [g2chr[x] for x in df['_gidx']]
def _set_df2prop(src, tgt, default):
dic = dict(UT.izipcols(df2, ['_gidx', src]))
df[tgt] = [dic.get(x, default) for x in df['_gidx']]
_set_df2prop('gcov2', 'gcov2', 0)
_set_df2prop('len', 'len2', 0)
_set_df2prop('cids', 'cids', '')
_set_df2prop('gst2', 'st2', -1)
_set_df2prop('ged2', 'ed2', -1)
_set_df2prop('glen2', 'glen2', 0)
cols = [
'_gidx', 'chr', 'st', 'ed', 'len', 'val', 'gcov', 'glen', 'len2',
'gcov2', 'cids', 'st2', 'ed2', 'glen2'
]
cols = ['_gidx', 'gcov']
df = df[cols]
UT.save_tsv_nidx_whead(df, gcovpath)
return df
def ex_d_e2p(self,eid):
try:
return dict(UT.izipcols(self.d2ep.get_group(eid), ['e_id_a','p']))
except:
return {}
def _igen():
for chrom,tname,strand,est,eed in UT.izipcols(kg,['chr','name','strand','_ests','_eeds']):
for st,ed in izip(eed[:-1],est[1:]):
yield (chrom,st+1,ed,tname,0,strand,'j')
def find_genes4(sj,
ae,
filepre,
cachename=None,
np=1,
override=False,
depth=500,
separatese=True):
"""
Adds _gidx column to ae
Connection: 1) by junctions, 2) by overlap in the same strand
Returns genes [set([_id,..]), ...]
"""
if '_id' not in ae.columns:
LOG.info('setting ex _id...')
UT.set_ids(ae)
if '_id' not in sj.columns:
LOG.info('setting sj _id...')
UT.set_ids(sj)
if 'cat' not in ae.columns:
UT.set_exon_category(sj, ae)
if 'a_id' not in ae.columns:
UT.set_ad_info(sj, ae)
### FIND GENES
if cachename and os.path.exists(cachename) and not override:
LOG.info('loading cached genes (connected components)...')
genes = pickle.load(open(cachename, 'rb'))
else:
LOG.info('finding genes (connected components)...')
_sttime = time.time()
if separatese:
me, se = UT.mese(ae)
genes = mcore_allcomponents4(sj, me, filepre, np, depth=depth)
# SE genes
genes += [set([x]) for x in se['_id']]
else:
genes = mcore_allcomponents4(sj, ae, filepre, np, depth=depth)
# version 4 graph: uses overlaps in addition to junctions to connect
# genes = [set([_id's]),...]
if cachename:
UT.makedirs(os.path.dirname(cachename))
pickle.dump(genes, open(cachename, 'wb'))
LOG.info(' time: {0:.3f}s'.format(time.time() - _sttime))
### WRITE EXONS W/ GENE number
LOG.info('assigning gidx...')
_sttime = time.time()
i2g = {} # eid => _gidx
i2gn = {} # eidt => gname
g2gn = {}
i2s = dict(UT.izipcols(ae, ['_id', 'strand'])) # eid => strand
#i2c = dict(UT.izipcols(ae, ['_id','cat'])) # eid => category
s2n = {'+': 'P', '-': 'N', '.': '', '.+': '', '.-': ''}
c2n = {'s': 'S', 'i': 'G', '5': 'G', '3': 'G'}
for i, ids in enumerate(genes):
gid = i + 1
strand = s2n[i2s[list(ids)[0]]]
cat = 'S' if len(ids) == 1 else 'G'
if strand == 'N': # negative strand
gid = -gid
gname = 'J{0}{1}{2}'.format(strand, cat, abs(gid))
g2gn[gid] = gname
for x in ids:
i2g[x] = gid
i2gn[x] = gname
ae['_gidx'] = [i2g[x] for x in ae['_id']]
ae['gname'] = [i2gn[x] for x in ae['_id']]
## set sj _gidx, use acceptor=>_gidx map (exon a_id, sj a_id)
a2g = dict(UT.izipcols(ae, ['a_id', '_gidx']))
d2g = dict(UT.izipcols(ae, ['d_id', '_gidx']))
sj['_gidx'] = [
a2g.get(x, d2g.get(y, 0))
for x, y in UT.izipcols(sj, ['a_id', 'd_id'])
]
sj['gname'] = [g2gn.get(x, '') for x in sj['_gidx']]
# This shouldn't happen
nidx = ae['_gidx'] == 0
if N.sum(nidx) > 0:
LOG.warning(
'###### WARNING!!!!!! exons with no gene assignment:{0}'.format(
N.sum(nidx)))
#ae.loc[nidx, '_gidx'] = N.arange(len(ae),len(ae)+N.sum(nidx))
return genes
def __init__(self,
ex,
sj,
xmargin=None,
ymargin=0.25,
compress=True,
ecov='ecov',
ucnt='ucnt',
mcnt='mcnt',
minlw=1,
drawscalebar=True,
ecovth=None,
jcntth=None,
origin=None,
sortexby=None,
fontsize=7):
self.ymargin = ymargin
self.ecov = ecov
self.ucnt = ucnt
self.jcnt = jcnt = 'jcnt'
self.mcnt = mcnt
self.minlw = minlw
self.drawscalebar = drawscalebar
self.ecovth = ecovth
self.jcntth = jcntth
self.ex = ex = ex.copy()
self.sj = sj = sj.copy()
self.compress = compress
self.fontsize = fontsize
if sortexby is None:
self.sortexby = ecov
else:
self.sortexby = sortexby # when plotting multiple and comparing, you want to use same sorting
# start and end, strand
if ex.iloc[0]['strand'] == '+':
if origin is None:
origin = ex['st'].min()
ex['xst'] = ex['st'] - origin
ex['xed'] = ex['ed'] - origin
self.strand = '+'
self.origin = origin
else:
if origin is None:
origin = ex['ed'].max()
ex['xst'] = origin - ex['ed']
ex['xed'] = origin - ex['st']
self.strand = '-'
self.origin = origin
# fix old a_id null
if (ex['a_id'].min() == -1) and (N.sum(ex['a_id'] == 0) == 0):
ex.loc[ex['a_id'] == -1, 'a_id'] = 0
ex.loc[ex['d_id'] == -1, 'd_id'] = 0
sj.loc[sj['a_id'] == -1, 'a_id'] = 0
sj.loc[sj['d_id'] == -1, 'd_id'] = 0
ex['len'] = ex['xed'] - ex['xst']
if xmargin is None:
xmargin = int(ex['len'].mean())
self.xmargin = xmargin
if ecov not in ex.columns:
ex[ecov] = 1
if (ucnt not in sj.columns) or (mcnt not in sj.columns):
sj[jcnt] = 1
sj[jcnt + '_ls'] = 'solid'
else:
# sj uniq, mult
sj[jcnt] = [x or y for x, y in sj[[ucnt, mcnt]].values]
sj[jcnt + '_ls'] = ['solid' if x else 'dashed' for x in sj[ucnt]]
if ecovth is not None:
self.ex = ex = ex[ex[ecov] > ecovth].copy()
if jcntth is not None:
self.sj = sj = sj[sj[jcnt] > jcntth].copy()
if len(ex) == 0:
return
# find exon groups
if 'asize' not in ex.columns:
a2size = dict(
UT.izipcols(
ex.groupby('a_id').size().reset_index(), ['a_id', 0]))
d2size = dict(
UT.izipcols(
ex.groupby('d_id').size().reset_index(), ['d_id', 0]))
a2size[0] = 0
d2size[0] = 0
ex['asize'] = [a2size[x] for x in ex['a_id']]
ex['dsize'] = [d2size[x] for x in ex['d_id']]
ex['group'] = [
'a{0}'.format(ai) if (a != 0 and a > d) else 'd{0}'.format(di)
for a, ai, d, di in ex[['asize', 'a_id', 'dsize', 'd_id']].values
]
# find exon group st, ed
exg = ex.groupby('group')
g2st = dict(
UT.izipcols(exg['xst'].min().reset_index(), ['group', 'xst']))
g2ed = dict(
UT.izipcols(exg['xed'].max().reset_index(), ['group', 'xed']))
g2size = dict(UT.izipcols(exg.size().reset_index(), ['group', 0]))
ex['gst'] = [g2st[x] for x in ex['group']]
ex['ged'] = [g2ed[x] for x in ex['group']]
ex['gsize'] = [g2size[x] for x in ex['group']]
#self.ex = ex = ex.sort_values(['group',ecov]) #'gst','ged','xst','xed'])
self.ex = ex = ex.sort_values(['group', self.sortexby
]) #'gst','ged','xst','xed'])
# find exon y pos within group
def _eypos(gs):
g0, s0 = gs[0] # first g
cnt = 0
yield cnt - (s0 - 1) / 2.
for g1, s1 in gs[1:]:
if g1 == g0:
cnt += 1
else:
cnt = 0
yield cnt - (s1 - 1) / 2.
g0 = g1
ex['eypos'] = [x for x in _eypos(ex[['group', 'gsize']].values)]
# find group y center pos
self.gr = gr = ex.groupby('group')[['gst', 'ged',
'gsize']].first().sort_values(
['gst', 'ged'])
gr['len'] = gr['ged'] - gr['gst']
def _gypos(gr):
side = 1
r0 = gr.iloc[0]
h = r0['gsize'] / 2.
ged0 = r0['ged']
gy0 = {1: h, -1: -h} # remember filled height both side (1,-1)
yield 0 # first one gets center
for gst1, ged1, gsiz1 in gr[['gst', 'ged', 'gsize']].values[1:]:
h = gsiz1 / 2.
if ged0 <= gst1: # no overlap
gy0 = {1: h, -1: -h}
yield 0
else:
gy1 = gy0[side] + side * gsiz1 / 2.
gy0[side] = gy0[side] + side * gsiz1
side = -1 * side # flip side
yield gy1
gst0 = gst1
ged0 = max(ged0, ged1)
gr['gypos'] = [x for x in _gypos(gr)]
# compress x coord
if compress:
def _gxst(gr):
r0 = gr.iloc[0]
delta = 0
yield r0['gst'] - delta # 0
ged0 = r0['ged']
for i, r1 in gr.iloc[1:].iterrows():
gst1 = r1['gst']
if gst1 - ged0 > self.xmargin:
delta += (gst1 - ged0 - self.xmargin)
yield gst1 - delta
ged0 = r1['ged']
gr['cst'] = [x for x in _gxst(gr)]
else:
gr['cst'] = gr['gst']
#gr['ced'] = gr['cst']+gr['len']
ex['cst0'] = [
gr['cst'].ix[g] + (xst - gst)
for g, xst, gst in ex[['group', 'xst', 'gst']].values
]
ex['ced0'] = ex['cst0'] + ex['len']
if self.strand == '+':
ex['cst'] = origin + ex['cst0']
ex['ced'] = origin + ex['ced0']
else:
ex['cst'] = origin - ex['ced0']
ex['ced'] = origin - ex['cst0']
ex['ey'] = [
ey + gr['gypos'].ix[g] for ey, g in ex[['eypos', 'group']].values
]
