def __init__(self,
infasta=None,
outfile=None,
k=6,
binary=True,
mean=True,
std=True,
leave=True,
silent=False,
label=False):
self.infasta = infasta
self.seqs = None
if infasta is not None:
self.reader = Reader(infasta)
self.seqs = self.reader.get_seqs()
self.outfile = outfile
self.k = k
self.binary = binary
self.mean = mean
if isinstance(mean, str):
self.mean = np.load(mean)
self.std = std
if isinstance(std, str):
self.std = np.load(std)
self.leave = leave
self.silent = silent
self.label = label
self.counts = None
self.kmers = [''.join(i) for i in product('AGTC', repeat=k)]
self.map = {k: i for k, i in zip(self.kmers, range(4**k))}
def save(self, names=None):
"""Saves the counts appropriately based on current settings.
There are four output methods for the counts:
1. Binary. This saves just the counts as a binary numpy array.
2. No labels. Saves in plain text, but without any labels.
3. Default names. If no names are provided, fasta headers will be used as labels.
4. Custom names. Provide a list of names if you want to label lncRNAs with your own names.
Parameters
----------
names : [str] (default=None)
Unique names for rows of the Dataframe.
"""
assert not (self.binary and self.label
), 'You cannot label a binary file. Set only one as True.'
assert self.outfile is not None, 'Please provide an outfile location.'
if self.binary:
np.save(self.outfile, self.counts)
elif self.label:
if names is None:
if self.reader is None:
self.reader = Reader(self.infasta)
names = self.reader.get_headers()
df = DataFrame(data=self.counts, index=names, columns=self.kmers)
df.to_csv(self.outfile)
else:
np.savetxt(self.outfile, self.counts, delimiter=',', fmt='%1.6f')
def __init__(self, infasta=None, outfile=None, k=6,
binary=True, mean=True, std=True,
leave=True, silent=False, label=False, percentage=True):
self.infasta = infasta
self.seqs = None
if infasta is not None:
self.seqs = Reader(infasta).get_seqs()
self.outfile = outfile
self.k = k
self.binary = binary
self.mean = mean
if isinstance(mean, str):
self.mean = np.load(mean)
self.std = std
if isinstance(std, str):
self.std = np.load(std)
self.leave = leave
self.silent = silent
self.label = label
self.perc = percentage
self.counts = None
self.kmers = [''.join(i) for i in product('AGTC', repeat=k)]
self.map = {k:i for k,i in zip(self.kmers, range(4**k))}
if len(self.seqs) == 1 and self.std is True:
err = ('You cannot standardize a single sequence. '
'Please pass the path to an std. dev. array, '
'or use raw counts by setting std=False.')
raise ValueError(err)
def __init__(self, infasta=None, outfile=None, k=6):
self.seqs = None
if infasta is not None:
self.data, self.names, self.seqs = Reader(infasta).get_data()
self.outfile = outfile
self.k = k
self.counts = None
self.kmers = [''.join(i) for i in list(product('AGTC', repeat=k))]
class BasicCounter:
"""Generates overlapping kmer counts for a fasta file
Parameters
----------
infasta : str (default=None)
Full path to fasta file to be counted
outfile : str (default=None)
Full path to the counts file to be saved
k : int (default=6)
Size of kmer to be counted
binary : bool (default=True)
Saves as numpy array if True, else saves as csv
mean : bool, np.array, str (default=True)
Set the mean to 0 for each kmer/column of the count matrix.
If str, provide path to a previously calculated mean array.
std : bool or str (default=True)
Set the std. dev. to 1 for each kmer/column of the count matrix
If str, provide path to a previously calculated std array.
leave : bool (default=True)
Set to False if get_counts is used within another tqdm loop
silent : bool (default=False)
Set to True to turn off tqdm progress bar
Attributes
----------
counts : None
Stores the ndarray of kmer counts
kmers : list
str elements of all kmers of size k
map : dict
Mapping of kmers to column values
"""
def __init__(self,
infasta=None,
outfile=None,
k=6,
binary=True,
mean=True,
std=True,
leave=True,
silent=False,
label=False):
self.infasta = infasta
self.seqs = None
if infasta is not None:
self.reader = Reader(infasta)
self.seqs = self.reader.get_seqs()
self.outfile = outfile
self.k = k
self.binary = binary
self.mean = mean
if isinstance(mean, str):
self.mean = np.load(mean)
self.std = std
if isinstance(std, str):
self.std = np.load(std)
self.leave = leave
self.silent = silent
self.label = label
self.counts = None
self.kmers = [''.join(i) for i in product('AGTC', repeat=k)]
self.map = {k: i for k, i in zip(self.kmers, range(4**k))}
def occurrences(self, row, seq):
"""Counts kmers on a per kilobase scale"""
counts = defaultdict(int)
length = len(seq)
increment = 1000 / length
for c in range(length - self.k + 1):
kmer = seq[c:c + self.k]
counts[kmer] += increment
for kmer, n in counts.items():
if kmer in self.map:
row[self.map[kmer]] = n
return row
def _progress(self):
"""Determine which iterator to loop over for counting."""
if self.silent:
return self.seqs
if not self.leave:
tqdm_seqs = my_tqdm()(self.seqs, desc='Kmers', leave=False)
else:
tqdm_seqs = my_tqdm()(self.seqs)
return tqdm_seqs
def center(self):
"""mean center counts by column"""
if self.mean is True:
self.mean = np.mean(self.counts, axis=0)
self.counts -= self.mean
def standardize(self):
"""divide out the standard deviations from columns of the count matrix"""
if self.std is True:
self.std = np.std(self.counts, axis=0)
self.counts /= self.std
def get_counts(self):
"""Generates kmer counts for a fasta file"""
self.counts = np.zeros([len(self.seqs), 4**self.k], dtype=np.float32)
seqs = self._progress()
for i, seq in enumerate(seqs):
self.counts[i] = self.occurrences(self.counts[i], seq)
if self.mean is not False:
self.center()
if self.std is not False:
self.standardize()
def save(self, names=None):
"""Saves the counts appropriately based on current settings.
There are four output methods for the counts:
1. Binary. This saves just the counts as a binary numpy array.
2. No labels. Saves in plain text, but without any labels.
3. Default names. If no names are provided, fasta headers will be used as labels.
4. Custom names. Provide a list of names if you want to label lncRNAs with your own names.
Parameters
----------
names : [str] (default=None)
Unique names for rows of the Dataframe.
"""
assert not (self.binary and self.label
), 'You cannot label a binary file. Set only one as True.'
assert self.outfile is not None, 'Please provide an outfile location.'
if self.binary:
np.save(self.outfile, self.counts)
elif self.label:
if names is None:
if self.reader is None:
self.reader = Reader(self.infasta)
names = self.reader.get_headers()
df = DataFrame(data=self.counts, index=names, columns=self.kmers)
df.to_csv(self.outfile)
else:
np.savetxt(self.outfile, self.counts, delimiter=',', fmt='%1.6f')
def make_count_file(self, names=None):
"""Wrapper function for the most common way to generate count files.
Given a numpy file name, it will save a numpy file where counts have been:
cast as a dense array, centered, and standardized.
Parameters
----------
names : [str] (default=None)
lncRNA names to pass to self.save
"""
self.get_counts()
if self.outfile is not None:
self.save(names)
return self.counts