def read(handle, format):
alphabet = IUPAC.unambiguous_dna
counts = {}
if format == "pfm":
# reads the motif from Jaspar .pfm file
letters = "ACGT"
for letter, line in zip(letters, handle):
words = line.split()
#if there is a letter in the beginning, ignore it
if words[0] == letter:
words = words[1:]
counts[letter] = map(float, words)
motif = Motif(alphabet, counts=counts)
elif format == "sites":
# reads the motif from Jaspar .sites file
instances = []
for line in handle:
if not line.startswith(">"):
break
# line contains the header ">...."
# now read the actual sequence
line = handle.next()
instance = ""
for c in line.strip():
if c == c.upper():
instance += c
instance = Seq(instance, alphabet)
instances.append(instance)
motif = Motif(alphabet, instances=instances)
else:
raise ValueError("Unknown format %s" % format)
motif.mask = "*" * motif.length
return motif
def read(handle, format):
alphabet = IUPAC.unambiguous_dna
counts = {}
if format=="pfm":
# reads the motif from Jaspar .pfm file
letters = "ACGT"
for letter, line in zip(letters, handle):
words = line.split()
#if there is a letter in the beginning, ignore it
if words[0]==letter:
words = words[1:]
counts[letter] = map(float, words)
motif = Motif(alphabet, counts=counts)
elif format=="sites":
# reads the motif from Jaspar .sites file
instances = []
for line in handle:
if not line.startswith(">"):
break
# line contains the header ">...."
# now read the actual sequence
line = handle.next()
instance = ""
for c in line.strip():
if c==c.upper():
instance += c
instance = Seq(instance, alphabet)
instances.append(instance)
motif = Motif(alphabet, instances=instances)
else:
raise ValueError("Unknown format %s" % format)
motif.set_mask("*"*motif.length)
return motif
class AlignAceConsumer:
"""
The general purpose consumer for the AlignAceScanner (DEPRECATED).
Should be passed as the consumer to the feed method of the AlignAceScanner. After 'consuming' the file, it has the list of motifs in the motifs property.
This class is DEPRECATED; please use the read() function in this module
instead.
"""
def __init__(self):
import warnings
warnings.warn(
"Bio.Motif.Parsers.AlignAce.AlignAceConsumer is deprecated; please use the read() function in this module instead.",
Bio.BiopythonDeprecationWarning)
self.motifs = []
self.current_motif = None
self.param_dict = None
def parameters(self, line):
self.param_dict = {}
def parameter(self, line):
par_name = line.split("=")[0].strip()
par_value = line.split("=")[1].strip()
self.param_dict[par_name] = par_value
def sequences(self, line):
self.seq_dict = []
def sequence(self, line):
seq_name = line.split("\t")[1]
self.seq_dict.append(seq_name)
def motif(self, line):
self.current_motif = Motif()
self.motifs.append(self.current_motif)
self.current_motif.alphabet = IUPAC.unambiguous_dna
def motif_hit(self, line):
seq = Seq(line.split("\t")[0], IUPAC.unambiguous_dna)
self.current_motif.add_instance(seq)
def motif_score(self, line):
self.current_motif.score = float(line.split()[-1])
def motif_mask(self, line):
self.current_motif.set_mask(line.strip("\n\c"))
def noevent(self, line):
pass
def version(self, line):
self.ver = line
def command_line(self, line):
self.cmd_line = line
class AlignAceConsumer(object):
"""
The general purpose consumer for the AlignAceScanner (DEPRECATED).
Should be passed as the consumer to the feed method of the AlignAceScanner. After 'consuming' the file, it has the list of motifs in the motifs property.
This class is DEPRECATED; please use the read() function in this module
instead.
"""
def __init__(self):
import warnings
warnings.warn("Bio.Motif.Parsers.AlignAce.AlignAceConsumer is deprecated; please use the read() function in this module instead.", Bio.BiopythonDeprecationWarning)
self.motifs=[]
self.current_motif=None
self.param_dict = None
def parameters(self,line):
self.param_dict={}
def parameter(self,line):
par_name = line.split("=")[0].strip()
par_value = line.split("=")[1].strip()
self.param_dict[par_name]=par_value
def sequences(self,line):
self.seq_dict=[]
def sequence(self,line):
seq_name = line.split("\t")[1]
self.seq_dict.append(seq_name)
def motif(self,line):
self.current_motif = Motif()
self.motifs.append(self.current_motif)
self.current_motif.alphabet=IUPAC.unambiguous_dna
def motif_hit(self,line):
seq = Seq(line.split("\t")[0],IUPAC.unambiguous_dna)
self.current_motif.add_instance(seq)
def motif_score(self,line):
self.current_motif.score = float(line.split()[-1])
def motif_mask(self,line):
self.current_motif.set_mask(line.strip("\n\c"))
def noevent(self,line):
pass
def version(self,line):
self.ver = line
def command_line(self,line):
self.cmd_line = line
class AlignAceConsumer:
"""
The general purpose consumer for the AlignAceScanner.
Should be passed as the consumer to the feed method of the AlignAceScanner. After 'consuming' the file, it has the list of motifs in the motifs property.
"""
def __init__(self):
self.motifs = []
self.current_motif = None
self.param_dict = None
def parameters(self, line):
self.param_dict = {}
def parameter(self, line):
par_name = line.split("=")[0].strip()
par_value = line.split("=")[1].strip()
self.param_dict[par_name] = par_value
def sequences(self, line):
self.seq_dict = []
def sequence(self, line):
seq_name = line.split("\t")[1]
self.seq_dict.append(seq_name)
def motif(self, line):
self.current_motif = Motif()
self.motifs.append(self.current_motif)
self.current_motif.alphabet = IUPAC.unambiguous_dna
def motif_hit(self, line):
seq = Seq(line.split("\t")[0], IUPAC.unambiguous_dna)
self.current_motif.add_instance(seq)
def motif_score(self, line):
self.current_motif.score = float(line.split()[-1])
def motif_mask(self, line):
self.current_motif.set_mask(line.strip("\n\c"))
def noevent(self, line):
pass
def version(self, line):
self.ver = line
def command_line(self, line):
self.cmd_line = line
class AlignAceConsumer:
"""
The general purpose consumer for the AlignAceScanner.
Should be passed as the consumer to the feed method of the AlignAceScanner. After 'consuming' the file, it has the list of motifs in the motifs property.
"""
def __init__(self):
self.motifs = []
self.current_motif = None
self.param_dict = None
def parameters(self, line):
self.param_dict = {}
def parameter(self, line):
par_name = line.split("=")[0].strip()
par_value = line.split("=")[1].strip()
self.param_dict[par_name] = par_value
def sequences(self, line):
self.seq_dict = []
def sequence(self, line):
seq_name = line.split("\t")[1]
self.seq_dict.append(seq_name)
def motif(self, line):
self.current_motif = Motif()
self.motifs.append(self.current_motif)
self.current_motif.alphabet = IUPAC.unambiguous_dna
def motif_hit(self, line):
seq = Seq(line.split("\t")[0], IUPAC.unambiguous_dna)
self.current_motif.add_instance(seq)
def motif_score(self, line):
self.current_motif.score = float(line.split()[-1])
def motif_mask(self, line):
self.current_motif.set_mask(line.strip("\n\c"))
def noevent(self, line):
pass
def version(self, line):
self.ver = line
def command_line(self, line):
self.cmd_line = line
def construct_weblogo(self, weblogo_filename, weblogo_revcompl_filename):
self.weblogo_basename = os.path.basename(weblogo_filename)
self.weblogo_revcompl_basename = os.path.basename(weblogo_revcompl_filename)
motif = Motif(alphabet=IUPAC.unambiguous_dna)
for kmer in self.kmer_lst:
motif.add_instance(Seq(kmer, motif.alphabet))
logowidth_normal = self.construct_weblogo_helper(weblogo_filename, motif)
#reverse complement
motif_revcompl = motif.reverse_complement()
logowidth_revcompl = self.construct_weblogo_helper(weblogo_revcompl_filename, motif_revcompl)
self.logowidth = max(self.logowidth, logowidth_normal, logowidth_revcompl)
def read(handle):
"""read(handle)"""
record = Record()
record.ver = next(handle)
record.cmd_line = next(handle)
for line in handle:
if line.strip() == "":
pass
elif line[:4] == "Para":
record.param_dict = {}
elif line[0] == "#":
seq_name = line.split("\t")[1]
record.seq_dict.append(seq_name)
elif "=" in line:
par_name = line.split("=")[0].strip()
par_value = line.split("=")[1].strip()
record.param_dict[par_name] = par_value
elif line[:5] == "Input":
record.seq_dict = []
elif line[:5] == "Motif":
record.current_motif = Motif()
record.motifs.append(record.current_motif)
record.current_motif.alphabet = IUPAC.unambiguous_dna
elif line[:3] == "MAP":
record.current_motif.score = float(line.split()[-1])
elif len(line.split("\t")) == 4:
seq = Seq(line.split("\t")[0], IUPAC.unambiguous_dna)
record.current_motif.add_instance(seq)
elif "*" in line:
record.current_motif.set_mask(line.strip("\n\c"))
else:
raise ValueError(line)
return record
def construct_weblogo(self, weblogo_filename, weblogo_revcompl_filename):
self.weblogo_basename = os.path.basename(weblogo_filename)
self.weblogo_revcompl_basename = os.path.basename(
weblogo_revcompl_filename)
motif = Motif(alphabet=IUPAC.unambiguous_dna)
for kmer in self.kmer_lst:
motif.add_instance(Seq(kmer, motif.alphabet))
logowidth_normal = self.construct_weblogo_helper(
weblogo_filename, motif)
#reverse complement
motif_revcompl = motif.reverse_complement()
logowidth_revcompl = self.construct_weblogo_helper(
weblogo_revcompl_filename, motif_revcompl)
self.logowidth = max(self.logowidth, logowidth_normal,
logowidth_revcompl)
def read(handle):
"""read(handle)"""
record = Record()
line = handle.next()
record.version = line.strip()
line = handle.next()
record.command = line.strip()
for line in handle:
line = line.strip()
if line == "":
pass
elif line[:4] == "Para":
record.parameters = {}
elif line[0] == "#":
seq_name = line.split("\t")[1]
record.sequences.append(seq_name)
elif "=" in line:
par_name, par_value = line.split("=")
par_name = par_name.strip()
par_value = par_value.strip()
record.parameters[par_name] = par_value
elif line[:5] == "Input":
record.sequences = []
elif line[:5] == "Motif":
words = line.split()
assert words[0] == "Motif"
number = int(words[1])
instances = []
elif line[:3] == "MAP":
motif = Motif(IUPAC.unambiguous_dna, instances)
motif.score = float(line.split()[-1])
motif.number = number
motif.set_mask(mask)
record.motifs.append(motif)
elif len(line.split("\t")) == 4:
seq = Seq(line.split("\t")[0], IUPAC.unambiguous_dna)
instances.append(seq)
elif "*" in line:
mask = line.strip("\r\n")
else:
raise ValueError(line)
return record
def read(handle):
"""read(handle)"""
record = Record()
line = handle.next()
record.version = line.strip()
line = handle.next()
record.command = line.strip()
for line in handle:
line = line.strip()
if line == "":
pass
elif line[:4] == "Para":
record.parameters = {}
elif line[0] == "#":
seq_name = line.split("\t")[1]
record.sequences.append(seq_name)
elif "=" in line:
par_name, par_value = line.split("=")
par_name = par_name.strip()
par_value = par_value.strip()
record.parameters[par_name] = par_value
elif line[:5] == "Input":
record.sequences = []
elif line[:5] == "Motif":
words = line.split()
assert words[0] == "Motif"
number = int(words[1])
instances = []
elif line[:3] == "MAP":
motif = Motif(IUPAC.unambiguous_dna, instances)
motif.score = float(line.split()[-1])
motif.number = number
motif.mask = mask
record.motifs.append(motif)
elif len(line.split("\t")) == 4:
seq = Seq(line.split("\t")[0], IUPAC.unambiguous_dna)
instances.append(seq)
elif "*" in line:
mask = line.strip("\r\n")
else:
raise ValueError(line)
return record
def read(handle):
"""read(handle)"""
record = Record()
line = handle.next()
record.version = line.strip()
line = handle.next()
record.command = line.strip()
for line in handle:
line = line.strip()
if line == "":
pass
elif line[:4] == "Para":
record.parameters = {}
elif line[0] == "#":
seq_name = line.split("\t")[1]
record.sequences.append(seq_name)
elif "=" in line:
par_name, par_value = line.split("=")
par_name = par_name.strip()
par_value = par_value.strip()
record.parameters[par_name] = par_value
elif line[:5] == "Input":
record.sequences = []
elif line[:5] == "Motif":
current_motif = Motif()
current_motif.alphabet = IUPAC.unambiguous_dna
record.motifs.append(current_motif)
elif line[:3] == "MAP":
current_motif.score = float(line.split()[-1])
elif len(line.split("\t")) == 4:
seq = Seq(line.split("\t")[0], IUPAC.unambiguous_dna)
current_motif.add_instance(seq)
elif "*" in line:
current_motif.set_mask(line.strip("\n\c"))
else:
raise ValueError(line)
return record
def read(handle):
"""read(handle)"""
record = Record()
line = handle.next()
record.version = line.strip()
line = handle.next()
record.command = line.strip()
for line in handle:
line = line.strip()
if line=="":
pass
elif line[:4]=="Para":
record.parameters={}
elif line[0]=="#":
seq_name = line.split("\t")[1]
record.sequences.append(seq_name)
elif "=" in line:
par_name, par_value = line.split("=")
par_name = par_name.strip()
par_value = par_value.strip()
record.parameters[par_name]=par_value
elif line[:5]=="Input":
record.sequences=[]
elif line[:5]=="Motif":
current_motif = Motif()
current_motif.alphabet=IUPAC.unambiguous_dna
record.motifs.append(current_motif)
elif line[:3]=="MAP":
current_motif.score = float(line.split()[-1])
elif len(line.split("\t"))==4:
seq = Seq(line.split("\t")[0],IUPAC.unambiguous_dna)
current_motif.add_instance(seq)
elif "*" in line:
current_motif.set_mask(line.strip("\n\c"))
else:
raise ValueError(line)
return record
def __init__(self):
BaseMotif.__init__(self)
self.references = []
def __init__(self):
Motif.__init__(self)
self.evalue = 0.0
def motif(self,line):
self.current_motif = Motif()
self.motifs.append(self.current_motif)
self.current_motif.alphabet=IUPAC.unambiguous_dna
def __init__ (self):
Motif.__init__(self)
self.evalue = 0.0
def motif(self, line):
self.current_motif = Motif()
self.motifs.append(self.current_motif)
self.current_motif.alphabet = IUPAC.unambiguous_dna
def __init__(self, alphabet=None, instances=None):
Motif.__init__(self, alphabet, instances)
self.evalue = 0.0
def __init__ (self, alphabet=None, instances=None):
Motif.__init__(self, alphabet, instances)
self.evalue = 0.0
def graph_logo(
alignment,
columns,
filename=None,
dpi=None, edgecolor='k', figsize=None, format='pdf', labels=None, linewidth=0., transparent=True,
refidx=-1
):
if filename is None:
fd, filename = mkstemp(); close(fd)
if figsize is None:
figsize = (3, 3)
if labels is None:
labels = ['%d' % (idx + 1) for idx in columns]
if refidx >= 0:
msa = alignment
alignment = msa[:refidx]
alignment.extend(msa[refidx + 1:])
M = len(alignment)
N = len(columns)
alph = None
for _alph in (_DNA_ALPHABET, _RNA_ALPHABET, _AMINO_ALPHABET):
for r in alignment:
r.seq.alphabet = _alph
if all([_verify_alphabet(r.seq.upper()) for r in alignment]):
alph = _alph
break
if alph is None:
raise RuntimeError("sequences with indeterminable alphabet provided")
motif = Motif(alphabet=alph)
instances = (''.join(z).upper() for z in zip(*[alignment[:, i] for i in columns]))
for instance in instances:
motif.add_instance(Seq(instance, alph))
# set laplace = True to include the backgrounds
pwm = _fix_ambigs(motif.pwm(laplace=False), alph)
# heuristic to determine whether nucleotide or protein alphabet
# need to use either base 4 or 20 depending
alphlen, _alphkeys = max(((len(pwm[i]), pwm[i].keys()) for i in range(N)), key=itemgetter(0))
s, colors = (4, _DNA_COLORS) if alphlen < 20 else (20, _AMINO_COLORS)
alphkeys = ['']
alphkeys.extend(_alphkeys)
alphmap = dict(zip(alphkeys, range(len(alphkeys))))
# compute the information content at each position
maxbits = np.log2(s)
e_n = (s - 1) / (2. * np.log(2) * M)
R = maxbits * np.ones((N,), dtype=float)
R -= [-sum(v * np.log2(v) for _, v in pwm[i].items() if v > 0.) for i in range(N)]
R -= e_n
heights = np.zeros((alphlen, N), dtype=float)
identities = np.zeros((alphlen, N), dtype=int)
for j in range(N):
i = 0
for k, v in sorted(pwm[j].items(), key=itemgetter(1)):
heights[i, j] = R[j] * v
identities[i, j] = alphmap[k]
i += 1
font = Basefont(join(_HY454_FONT_PATHS[0], 'Roboto-Black.ttf'))
fig = plt.figure(figsize=figsize, dpi=dpi)
# make each column a vertical golden rect
rect = 0.2, 0.2, 0.382 * N, 0.618
ax = fig.add_axes(rect)
_adjust_spines_outward(ax, ('left',), 9)
ax.set_ylabel('bits', fontproperties=_ROBOTO_REGULAR)
if figsize is None:
fig.set_figwidth(N)
if transparent:
fig.patch.set_alpha(0.)
ax.patch.set_alpha(0.)
# remove the top and right ticks
for tick in ax.xaxis.get_major_ticks() + ax.yaxis.get_major_ticks():
tick.tick2On = False
# remove the bottom ticks
for tick in ax.xaxis.get_major_ticks():
tick.tick1On = False
# rotate the x-axis labels by 45 degrees to enhance packing
for label in ax.xaxis.get_ticklabels():
label.set_rotation(45)
# set font properties
for label in ax.xaxis.get_ticklabels() + ax.yaxis.get_ticklabels():
label.set_fontproperties(_ROBOTO_REGULAR)
# disable top and right spines, we don't need them
ax.spines['bottom'].set_visible(False)
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)
def format_xlabel(x, pos=None):
idx = np.clip(int(x)-1, 0, N-1)
return labels[idx]
ax.xaxis.set_major_formatter(FuncFormatter(format_xlabel))
# avoid too much precision
ax.yaxis.set_major_formatter(FormatStrFormatter('%1.1f'))
# set the ticks
ysep = 0.5 if alphlen < 20 else 1.0
yticks = np.arange(0, maxbits, ysep, dtype=float)
if maxbits - yticks[-1] < ysep:
yticks[-1] = maxbits
else:
yticks = np.append(yticks, maxbits)
ax.set_yticks(yticks)
ax.set_xticks(np.arange(1, N+1, dtype=float) + 0.5)
# set the axes limits here AFTER the ticks, otherwise borkage
ax.set_xlim((1, N+1))
ax.set_ylim((0, maxbits))
idxs = np.arange(1, N+1)
bottoms = np.zeros((N,), dtype=float)
for i in range(alphlen):
bars = ax.bar(idxs, heights[i, :], width=1., bottom=bottoms)
bottoms += heights[i, :]
for j, bar in enumerate(bars):
if identities[i, j]:
l = alphkeys[identities[i, j]]
glyph = font[l]
ax.add_patch(glyph)
glyph.set_transform(bar.get_transform())
bar.set_visible(False)
glyph.set_edgecolor(edgecolor)
glyph.set_facecolor(colors[l])
glyph.set_linewidth(linewidth)
glyph.set_zorder(-1)
# set the remaining spine to show the maximum value
ax.spines['left'].set_bounds(0, max(bottoms))
fig.savefig(filename, format=format, transparent=transparent, bbox_inches='tight', pad_inches=0.25)
return filename
def __init__(self):
BaseMotif.__init__(self)
self.references = []