def set_tooltips_to_column(self, tooltips_col, target_col):
"""Hide a column with tooltips and connect it with a column.
:param str tooltips_col: column with your tooltips.
:param str target_col: column to connect.
"""
# hide tooltips
try:
self.datatable_columns[tooltips_col]['visible'] = 'false'
except KeyError:
logger.warning(
"KeyError: Column name '{0}' does not exist.".format(
target_col))
pass
# function to add tooltips
fct = """function(data, type, row, meta){{
return '<a href="#" data-toggle="tooltip" title="'+row.{0}+'">'+data+'</a>';
}}
""".format(tooltips_col)
try:
self.datatable_columns[target_col]['render'] = fct
except KeyError:
logger.warning(
"KeyError: Column name '{0}' does not exist.".format(
target_col))
pass
def set_links_to_column(self, link_col, target_col):
"""Hide a column with urls and connect it with a column.
:param str link_col: column with your URLs.
:param str target_col: column to connect.
"""
# hide the link column
try:
self.datatable_columns[link_col]['visible'] = 'false'
except KeyError:
logger.warning(
"KeyError: Column name '{0}' does not exist.".format(
target_col))
pass
# function to add link
fct = """function(data, type, row, meta){{
return '<a href="'+row.{0}+'" target="_blank">'+data+'</a>';
}}
""".format(link_col)
try:
self.datatable_columns[target_col]['render'] = fct
except KeyError:
logger.warning(
"KeyError: Column name '{0}' does not exist.".format(
target_col))
pass
def get_df(self):
import pandas as pd
data = {}
for sample, filename in zip(self.sample_names, self.filenames):
df = pd.read_csv(filename)
df = df.groupby("kingdom")['percentage'].sum()
# if a taxon is obsolete, the kingdom is empty.
# We will set the kingdom as Unclassified and raise a warning
# if the count is > 5%
if " " in df.index:
percent = df.loc[" "]
if percent > 5:
logger.warning(
"Found {}% of taxons in obsolete category".format(
percent))
if "Unclassified" in df.index:
df.loc['Unclassified'] += df.loc[' ']
df.drop(" ", inplace=True)
else:
df.loc['Unclassified'] = df.loc[' ']
df.drop(" ", inplace=True)
data[sample] = df
df = pd.DataFrame(data)
#df.to_json(output.data)
df = df.sort_index(ascending=False)
return df
def _parse_data(self):
"""Parse the YAML file to get the block content (comments)
before each top-level sections. See doc in the constructor
Removes all # so that the block of comments can be interpreted as
a standard docstring in Sequanix
"""
current_block = []
current_section = "docstring"
# if we get a line that starts with #, this is a new comment or
# part of a block comment. Otherwise, it means the current block
# comment has ended.
for this in self.data:
# Beginning of a new section at top level
if self.regex_section.findall(this):
name = self.regex_section.findall(this)[0]
current_section = name.strip(":")
self.sections[current_section] = "".join(current_block)
current_block = []
current_section = None
elif this.startswith('#'): # a comment at top level
current_block.append(this)
elif this.strip() == "": # an empty line
#this was the main comment, or an isolated comment
current_block = []
else: # a non-empty line to skip
current_block = []
for key in self._get_expected_sections():
if key not in self.sections.keys():
logger.warning("section %s not dealt by the parsing function" % key)
def _get_specials(self, section):
"""This method extracts data from the docstring
Lines such as ::
field_choice__ = ["a", "b"]
are extracted. Where _choice is a special keyword to be
found.
"""
if section not in self.sections.keys():
logger.warning("%s not found in the yaml " % section)
return
comments = self.sections[section]
specials = {}
for line in comments.split("\n"):
if "#############" in line:
pass
elif sum([this in line for this in self._specials]):
for special in self._specials:
line = line[2:]
key, value = line.split("=", 1)
key = key.strip().rstrip("__")
value = value.strip()
specials[key] = list(eval(value))
return specials
def _parse_data(self):
"""Parse the YAML file to get the block content (comments)
before each top-level sections. See doc in the constructor
Removes all # so that the block of comments can be interpreted as
a standard docstring in Sequanix
"""
current_block = []
current_section = "docstring"
# if we get a line that starts with #, this is a new comment or
# part of a block comment. Otherwise, it means the current block
# comment has ended.
for this in self.data:
# Beginning of a new section at top level
if self.regex_section.findall(this):
name = self.regex_section.findall(this)[0]
current_section = name.strip(":")
self.sections[current_section] = "".join(current_block)
current_block = []
current_section = None
elif this.startswith('#'): # a comment at top level
current_block.append(this)
elif this.strip() == "": # an empty line
#this was the main comment, or an isolated comment
current_block = []
else: # a non-empty line to skip
current_block = []
for key in self._get_expected_sections():
if key not in self.sections.keys():
logger.warning("section %s not dealt by the parsing function" %
key)
def save_significant_pathways(self,
mode,
cutoff=0.05,
nmax=20,
background=None): #pragma: no cover
"""mode should be up, down or all"""
if background is None:
background = self.background
# select the relevant pathways
df = self._enrichr(mode, background).results
df = self._get_final_df(df, cutoff=cutoff, nmax=nmax)
logger.warning("Found {} pathways to save".format(len(df)))
if len(df) == nmax:
logger.warning("Restricted pathways to {}".format(nmax))
logger.info("saving {} deregulated pathways".format(len(df)))
summaries = {}
# save them
for ID in df['Term']:
summary = self.save_pathway(ID,
filename="{}_{}.png".format(ID, mode))
summaries[ID] = summary
return summaries
def get_sequana_adapters(type_, direction):
"""Return path to a list of adapters in FASTA format
:param tag: PCRFree, Rubicon, Nextera
:param type_: fwd, rev, revcomp
:return: path to the adapter filename
"""
# search possible types
registered = _get_registered_adapters()
if type_ not in registered:
logger.error("This adapter type (%s) is not valid" % type_)
logger.error("choose one in %s types" % registered)
raise ValueError
directions = ["fwd", "rev", "revcomp"]
if direction not in directions:
logger.error("This kind of tag (%s) is not valid" % direction)
logger.error("choose one in %s " % directions)
raise ValueError
try:
this = sequana_data("adapters_%s_%s.fa" % (type_, direction))
logger.warning("Rename {} (remove the adapters_ prefix)".format(this))
return this
except:
return sequana_data("%s_%s.fa" % (type_, direction))
def df(self):
# RG: ID read group ??
# np: number of passes
# rq ?
# rs: list 6 numbers ?
# za:
# zm ID of the ZMW
# sn: SNR how is this computed ?
# zs
# - sn: list of ACGT SNRs. A, C, G, T in that order
if self._df is not None:
return self._df
logger.info("Scanning input file. Please wait")
self.reset()
N = 0
all_results = []
# This takes 60% of the time...could use cython ?
for read in self.data:
tags = dict(read.tags) #11% of the time
res = []
# count reads
N += 1
if (N % 10000) == 0:
logger.info("Read %d sequences" %N)
# res[0] = read length
res.append(read.query_length) # also stored in tags["qe"] - tags["qs"]
# collections.counter is slow, let us do it ourself
res.append( 100. / read.qlen * sum(
[read.query_sequence.count(letter) if read.query_sequence
else 0 for letter in "CGcgSs"]))
# res[1:4] contains SNR stored in tags['sn'] in the order A, C, G, T
try:
snr = list(tags['sn'])
except:
snr = [None] * 4
res = res + snr
# res[6] = ZMW name, also stored in tags["zm"]
res.append(int(tags['zm']))
res.append(tags['np'])
# aggregate results
all_results.append(res)
self._df = pd.DataFrame(all_results,
columns=['read_length','GC_content','snr_A','snr_C','snr_G','snr_T','ZMW',
"nb_passes"])
self._df.ZMW = self._df.ZMW.astype(int)
if len(self._df.ZMW.unique()) != len(self._df):
logger.warning("Found non unique ZMW. This may not be a CCS but "
"a subread file. Consider using PacbioSubreads class")
self.reset()
return self._df
def _parse_data(self):
taxonomy = {}
logger.info("Reading kraken data")
columns = ["status", "taxon", "length"]
# we select only col 0,2,3 to save memoty, which is required on very
# large files
try:
# each call to concat in the for loop below
# will take time and increase with chunk position.
# for 15M reads, this has a big cost. So chunksize set to 1M
# is better than 1000 and still reasonable in memory
reader = pd.read_csv(self.filename,
sep="\t",
header=None,
usecols=[0, 2, 3],
chunksize=1000000)
except pd.parser.CParserError:
raise NotImplementedError # this section is for the case
#only_classified_output when there is no found classified read
self.unclassified = N # size of the input data set
self.classified = 0
self._df = pd.DataFrame([], columns=columns)
self._taxons = self._df.taxon
return
for chunk in reader:
try:
self._df
self._df = pd.concat([self._df, chunk])
except AttributeError:
self._df = chunk
self._df.columns = columns
count = sum(self._df.taxon == 1)
if count:
logger.warning("Found %s taxons with root ID (1)" % count)
# This gives the list of taxons as index and their amount
# above, we select only columns 0, 2, 3 the column are still labelled
# 0, 2, 3 in the df
self._taxons = self._df.groupby("taxon").size()
try:
self._taxons.drop(0, inplace=True)
except:
pass # 0 may not be there
self._taxons.sort_values(ascending=False, inplace=True)
category = self.df.groupby("status").size()
if 'C' in category.index:
self.classified = category['C']
else:
self.classified = 0
if 'U' in category.index:
self.unclassified = category['U']
else:
self.unclassified = 0
def _get_specials(self, section):
"""This method extracts data from the docstring
Lines such as ::
field_choice__ = ["a", "b"]
are extracted. Where _choice is a special keyword to be
found.
"""
if section not in self.sections.keys():
logger.warning("%s not found in the yaml " % section)
return
comments = self.sections[section]
specials = {}
for line in comments.split("\n"):
if "#############" in line:
pass
elif sum([this in line for this in self._specials]):
for special in self._specials:
line = line[2:]
key, value = line.split("=", 1)
key = key.strip().rstrip("__")
value = value.strip()
specials[key] = list(eval(value))
return specials
def _get_adapter_by_index(self, index_name, prefix):
"""Return adapter corresponding to the unique index
:param index_name: the unique index name to be found. If several
sequence do match, this is an error meaning the fasta file
with all adapters is not correctly formatted.
:return: an instance of :class:`Adapter` if index_name match an
adapter; returns None otherwise
::
from sequana import sequana_data, AdapterReader
filename = sequana_data("adapters_Nextera_fwd.fa")
ar = AdapterReader(filename)
ar.get_adapter_by_identifier("N712")
"""
# there should be only one
adapters = []
for this in self._data:
if prefix + str(index_name) in this.identifier.split("|"):
this_adapter = Adapter(identifier=this.identifier,
sequence=this.sequence,
comment=this.comment)
adapters.append(this_adapter)
if len(adapters) == 0:
return None
elif len(adapters) >= 2:
logger.warning(
"Found two adapters matching index {}. This may happen with Nextera adapters"
.format(index_name))
return adapters
def switch_header_to_gi(self, acc):
"""Kraken will only accept the GI from NCBI so we need to convert
the ENA accession to GI numbers"""
# Accession may have a version .1, .2 hence this try/except first
# without the version and then with the version.
# Note also that some accession are different from an earlier version.
# For instance, AF525933 is in the virus.txt list from ENA but
# the new updated accession ois AH012103 showing that the list and DB
# must not be fully synchronised.
# http://www.ebi.ac.uk/ena/data/search?query=AF525933
# In such case, the results attribute will be missing that accession,
# which needs to be searched for specifically. We cannot now its name
# before downloading the fasta.
if acc in self.results.keys():
res = self.results[acc]
else:
try:
res = self.results[acc.split(".")[0]]
except:
logger.warning(
"\nUnknown accession (%s). May be an updated version. Checking..."
% acc)
res = self.ena_id_to_gi_number([acc])
self.results.update(res)
res = res[acc]
logger.info('Found %s using GI number' % acc)
return ">" + res['identifier'] + " " + res['comment']
def exists(self, filename, exit_on_error=True, warning_only=False):
if os.path.exists(filename) is False:
if warning_only is False:
logger.error("{} file does not exists".format(filename))
if exit_on_error:
sys.exit(1)
elif warning_only is True:
logger.warning("{} file does not exists".format(filename))
def window_size(self, n):
if n % 2 == 0:
logger.warning("Window size must be an odd number.")
self._window_size = n + 1
logger.warning("{0} is incremented to {1}".format(
n, self._window_size))
else:
self._window_size = n
def run(self):
# To normalise one need to ignore the insertions since there
# are already included in the ACGT nucleotides
cols = ["A", "C", "G", "T", "N", "DEL"]
df = self.get_bases()
deletions = self.identify_deletions()
# consensus without deletions
dd = df.apply(lambda x: x.idxmax(), axis=1)
# check that deletions are consistent with the data
for d in deletions:
pos = int(d.resume["position"])
ref = d.resume["reference"]
# compare the reference of the deletions with the consensus
if "".join(dd.loc[pos:pos+len(ref)-1]) != ref:
logger.warning("reference string {} not found in consensus at position {}".format(ref, pos))
# Now, we insert the deletions removing the reference and then adding
# the alternate
# We aware that some deletions may overlap
for d in deletions:
pos = int(d.resume["position"])
ref = d.resume["reference"]
alt = d.resume["alternative"]
# the data up to the position of the reference/alternate SNP
# indices may not start at zero so we use loc instead of iloc
dfA = df.loc[0:pos-1]
# the alternate data needs a dummy dataframe. The indices are
# e.g. 0,1,2,3,4,5 and
# We reset the indices to start at the dfA last position and
# to be constant. For instance a dataframe dfB of 3 rows to be
# appended aftre position 1500 will have the indices 1500,1500,1500
# This garantee that the following dataframe dfC has indices > to
# those of dfB while allowing the next iteration to use the same
# consistance indices when searching for the next deletions
dfB = df.iloc[0:len(alt)].copy()
dfB.index = [pos] * len(dfB)
dfB *= 0
for i, nucleotide in enumerate(alt):
dfB.iloc[i][nucleotide] = 10000
# the rest of the data
dfC = df.loc[pos+len(ref):]
# !! do no reset indices !!! so that inserted dfB is still sorted
# and next accession with iloc/loc are still correctin the next
# iteration
df = dfA.append(dfB).append(dfC)#.reset_index(drop = True)
# now we can reset the indices
df.reset_index(drop=True, inplace=True)
dd = df.apply(lambda x: x.idxmax(), axis=1)
return dd
def download_taxonomic_file(self, overwrite=False):
"""Loads entire flat file from EBI
Do not overwrite the file by default.
"""
import ftplib
from sequana import sequana_config_path
if os.path.exists(self.database) and overwrite is False:
logger.info(
"Found taxonomy.dat file in sequana your path {}".format(
sequana_config_path))
return
else:
logger.info(
"Downloading and extracting the taxonomy file from the web. Please be patient."
)
if self.source == "ena":
url = 'ftp.ebi.ac.uk'
else:
url = 'ftp.ncbi.nlm.nih.gov'
self.ftp = ftplib.FTP(url)
self.ftp.login()
if self.source == "ena":
# for the EBI ftp only: self.ftp.cwd('databases')
self.ftp.cwd('pub')
self.ftp.cwd('databases')
self.ftp.cwd('taxonomy')
logger.warning(
'Downloading and saving in %s. This is from ebi and may be behind the NCBI taxonomy'
% self.database)
self.ftp.retrbinary('RETR taxonomy.dat',
open(self.database, 'wb').write)
ftp.close()
else:
self.ftp.cwd('pub')
self.ftp.cwd('taxonomy')
logger.warning('Downloading and saving in %s from ncbi ftp' %
self.database)
import tempfile
import shutil
with tempfile.TemporaryDirectory() as tmpdir:
filename = tmpdir + os.sep + "taxdump.tar.gz"
self.ftp.retrbinary('RETR taxdump.tar.gz',
open(filename, "wb").write)
import tarfile
tf = tarfile.open(filename)
assert "nodes.dmp" in tf.getnames()
assert "names.dmp" in tf.getnames()
tf.extract("nodes.dmp", tmpdir)
tf.extract("names.dmp", tmpdir)
ncbi = NCBITaxonomy(tmpdir + os.sep + "names.dmp",
tmpdir + os.sep + "nodes.dmp")
ncbi.create_taxonomy_file(tmpdir + os.sep + "taxonomy.dat")
shutil.move(tmpdir + os.sep + "taxonomy.dat", self.database)
self.ftp.close()
def _parse_data(self):
taxonomy = {}
logger.info("Reading kraken data")
columns = ["status", "taxon", "length"]
# we select only col 0,2,3 to save memoty, which is required on very
# large files
try:
# each call to concat in the for loop below
# will take time and increase with chunk position.
# for 15M reads, this has a big cost. So chunksize set to 1M
# is better than 1000 and still reasonable in memory
reader = pd.read_csv(self.filename, sep="\t", header=None,
usecols=[0,2,3], chunksize=1000000)
except pd.parser.CParserError:
raise NotImplementedError # this section is for the case
#only_classified_output when there is no found classified read
self.unclassified = N # size of the input data set
self.classified = 0
self._df = pd.DataFrame([], columns=columns)
self._taxons = self._df.taxon
return
for chunk in reader:
try:
self._df
self._df = pd.concat([self._df, chunk])
except AttributeError:
self._df = chunk
self._df.columns = columns
count = sum(self._df.taxon == 1)
if count:
logger.warning("Found %s taxons with root ID (1)" % count)
# This gives the list of taxons as index and their amount
# above, we select only columns 0, 2, 3 the column are still labelled
# 0, 2, 3 in the df
self._taxons = self._df.groupby("taxon").size()
try:
self._taxons.drop(0, inplace=True)
except:
pass # 0 may not be there
self._taxons.sort_values(ascending=False, inplace=True)
category = self.df.groupby("status").size()
if 'C' in category.index:
self.classified = category['C']
else:
self.classified = 0
if 'U' in category.index:
self.unclassified = category['U']
else:
self.unclassified = 0
def get_cond_from_sample(self, sample_name):
try:
candidates = [x for x in self.condition_names if sample_name.startswith(x)]
if len(candidates) == 1:
return candidates[0]
else:
raise ValueError("ambiguous sample name found in several conditions")
except:
logger.warning("{} not found".format(sample_name))
return None
def _check_if_joint(self):
try:
# Needs a try/except for empty VCF files
line = next(self)
self.rewind()
if len(line.samples) > 1:
return True
except:
logger.warning("Your input VCF may be empty")
return False
def copy_config_from_sequana(module, source="config.yaml",
target="config.yaml"):
# identify config name from the requested module
user_config = module.path + os.sep + source
if os.path.exists(user_config):
shutil.copy(user_config, target)
txt = "copied %s from sequana %s pipeline"
logger.info(txt % (source, module.name))
else:
logger.warning(user_config + "not found")
def get_roi(self):
"""Keep positions with zscore outside of the thresholds range.
:return: a dataframe from :class:`FilteredGenomeCov`
.. note:: depends on the :attr:`thresholds` low and high values.
"""
features = self.bed.feature_dict
try:
second_high = self.thresholds.high2
second_low = self.thresholds.low2
query = "zscore > @second_high or zscore < @second_low"
# in the genbank, the names appears as e.g. JB12345
# but in the fasta or BED files, it may be something like
# gi|269939526|emb|FN433596.1|
# so they do not match. We can try to guess it
alternative = None
if features:
if self.chrom_name not in features.keys():
msg = """Chromosome name (%s) not found
in the genbank. Make sure the chromosome names in
the BAM/BED files are compatible with the genbank
content. Genbank files contains the following keys """
for this in features.keys():
msg += "\n - %s" % this
alternative = [x for x in self.chrom_name.split("|") if x]
alternative = alternative[-1] # assume the accession is last
alternative = alternative.split('.')[0] # remove version
if alternative in features.keys():
msg += "\n Guessed the chromosome name to be: %s" % alternative
else:
features = None
logger.warning(msg % self.chrom_name)
if features:
if alternative:
return FilteredGenomeCov(self.df.query(query), self.thresholds,
features[alternative])
else:
return FilteredGenomeCov(self.df.query(query), self.thresholds,
features[self.chrom_name])
else:
return FilteredGenomeCov(self.df.query(query), self.thresholds)
except KeyError:
logger.error("Column zscore is missing in data frame.\n"
"You must run compute_zscore before get low coverage."
"\n\n", self.__doc__)
sys.exit(1)
def _get_files(self, pattern):
filenames = glob.glob(os.sep.join([self.directory, self.phix_directory,
pattern]))
if len(filenames) == 4:
mode = "pe"
elif len(filenames) == 2:
mode = "se"
elif len(filenames) == 0:
return
else:
logger.warning("PhixModule: more than 4 files "
"matched the pattern %s" % pattern)
return
return filenames, mode
def _get_files(self, pattern):
filenames = glob.glob(
os.sep.join([self.directory, self.phix_directory, pattern]))
if len(filenames) == 4:
mode = "pe"
elif len(filenames) == 2:
mode = "se"
elif len(filenames) == 0:
return
else:
logger.warning("PhixModule: more than 4 files "
"matched the pattern %s" % pattern)
return
return filenames, mode
def _download_minikraken(self, verbose=True):
dv = DevTools()
base = sequana_config_path + os.sep + ""
taxondir = base + os.sep + "taxonomy"
dv.mkdir(base)
dv.mkdir(taxondir)
logger.info("Downloading minikraken (4Gb)")
filename = base + os.sep + "minikraken.tgz"
if os.path.exists(filename) and md5(filename) == "30eab12118158d0b31718106785195e2":
logger.warning("%s already present" % filename)
else:
wget("https://ccb.jhu.edu/software/kraken/dl/minikraken.tgz", filename)
def _download_minikraken(self, verbose=True):
dv = DevTools()
base = sequana_config_path + os.sep + ""
taxondir = base + os.sep + "taxonomy"
dv.mkdir(base)
dv.mkdir(taxondir)
logger.info("Downloading minikraken (4Gb)")
filename = base + os.sep + "minikraken.tgz"
if os.path.exists(filename) and md5(filename) == "30eab12118158d0b31718106785195e2":
logger.warning("%s already present" % filename)
else:
wget("https://ccb.jhu.edu/software/kraken/dl/minikraken.tgz", filename)
def _get_files(self, pattern):
# !! need to sort the files so that R1 appears before R2
filenames = sorted(glob.glob(self.directory + os.sep + pattern))
if len(filenames) == 2:
mode = "pe"
elif len(filenames) == 1:
mode = "se"
elif len(filenames) == 0:
return
else:
logger.warning("FastQStatsModule: more than 2 files "
"matched the pattern %s" % pattern)
return
return filenames, mode
def _get_files(self, pattern):
# !! need to sort the files so that R1 appears before R2
filenames = sorted(glob.glob(self.directory + os.sep + pattern))
if len(filenames) == 2:
mode = "pe"
elif len(filenames) == 1:
mode = "se"
elif len(filenames) == 0:
return
else:
logger.warning("FastQStatsModule: more than 2 files "
"matched the pattern %s" % pattern)
return
return filenames, mode
def kraken_to_krona(self, output_filename=None, nofile=False):
"""
:return: status: True is everything went fine otherwise False
"""
if output_filename is None:
output_filename = self.filename + ".summary"
taxon_to_find = list(self.taxons.index)
if len(taxon_to_find) == 0:
logger.warning(
"No reads were identified. You will need a more complete database"
)
self.output_filename = output_filename
with open(output_filename, "w") as fout:
fout.write("%s\t%s" % (self.unclassified, "Unclassified"))
return False
# classified reads as root (1)
"""try:
logger.warning("Removing taxon 1 (%s values) " % self.taxons.iloc[1])
logger.info("Found %s taxons " % len(taxon_to_find))
taxon_to_find.pop(taxon_to_find.index(1))
except:
pass
"""
if len(taxon_to_find) == 0:
return False
df = self.get_taxonomy_db(taxon_to_find)
self.lineage = [";".join(this) for this in df[df.columns[0:-1]].values]
self.scnames = list(df['name'].values) # do we need a cast ?
# Now save the file
self.output_filename = output_filename
with open(output_filename, "w") as fout:
for i, this in enumerate(self.lineage):
taxon = taxon_to_find[i]
count = self.taxons.loc[taxon]
line = str(count) + "\t" + "\t".join(this.split(';'))
line += " " + self.scnames[i]
fout.write(line + '\n')
try:
fout.write("%s\t%s" % (self.unclassified, "Unclassified"))
except:
pass #unclassified may not exists if all classified
self._data_created = True
return True
def parse_cutadapt(self):
d = {}
# output
tobefound = self._get_data_tobefound()
adapters = []
data = self._rawdata.splitlines()
# some metadata to extract
for this in tobefound:
key, pattern = this
found = [line for line in data if line.startswith(pattern)]
if len(found) == 0:
logger.warning("ReportCutadapt: %s (not found)" % pattern)
elif len(found) == 1:
text = found[0].split(":", 1)[1].strip()
try:
this, percent = text.split()
self.jinja[key] = this
self.jinja[key+'_percent'] = percent
except:
self.jinja[key] = text
self.jinja[key+'_percent'] = "?"
dd = {}
positions = []
executable = "cutadapt"
for pos, this in enumerate(data):
if "This is Atropos" in this:
executable = "atropos"
if "Command line parameters: " in this:
cmd = this.split("Command line parameters: ")[1]
self.jinja['command'] = executable + " " + cmd
if this.startswith("=== ") and "Adapter" in this:
name = this.split("=== ")[1].split(" ===")[0].strip()
dd['name'] = name
continue
if this.startswith('Sequence:'):
info = this.split("Sequence:", 1)[1].strip()
info = info.split(";")
dd['info'] = {
'Sequence': info[0].strip(),
'Type': info[1].split(':',1)[1].strip(),
'Length': info[2].split(':',1)[1].strip(),
'Trimmed': info[3].split(':',1)[1].strip()
}
adapters.append(dd.copy())
self.data["adapters"] = adapters
def parse_cutadapt(self):
d = {}
# output
tobefound = self._get_data_tobefound()
adapters = []
data = self._rawdata.splitlines()
# some metadata to extract
for this in tobefound:
key, pattern = this
found = [line for line in data if line.startswith(pattern)]
if len(found) == 0:
logger.warning("ReportCutadapt: %s (not found)" % pattern)
elif len(found) == 1:
text = found[0].split(":", 1)[1].strip()
try:
this, percent = text.split()
self.jinja[key] = this
self.jinja[key + '_percent'] = percent
except:
self.jinja[key] = text
self.jinja[key + '_percent'] = "?"
dd = {}
positions = []
executable = "cutadapt"
for pos, this in enumerate(data):
if "This is Atropos" in this:
executable = "atropos"
if "Command line parameters: " in this:
cmd = this.split("Command line parameters: ")[1]
self.jinja['command'] = executable + " " + cmd
if this.startswith("=== ") and "Adapter" in this:
name = this.split("=== ")[1].split(" ===")[0].strip()
dd['name'] = name
continue
if this.startswith('Sequence:'):
info = this.split("Sequence:", 1)[1].strip()
info = info.split(";")
dd['info'] = {
'Sequence': info[0].strip(),
'Type': info[1].split(':', 1)[1].strip(),
'Length': info[2].split(':', 1)[1].strip(),
'Trimmed': info[3].split(':', 1)[1].strip()
}
adapters.append(dd.copy())
self.data["adapters"] = adapters
def __init__(self, design_filename, adapters):
""".. rubric:: Constructor
:param str design_filename: a CSV file that is compatible
with our :class:`sequana.expdesign.ExpDesignAdapter`
:param adapters: the type of adapters (PCRFree, Nextera,
Rubicon, TruSeq, SMARTer, Small)
The files of adapters are stored in Sequana and accessible with the
sequana_data function. So, for instance if adapters is set to Nextera,
the following file is used to identify the adapters::
sequana_data("adapters_Nextera_fwd.fa")
New adapters files can be added on request. See resources/data/adapters
for the full list. You can also use::
from sequana.adapters import _get_registered_adapters
_get_registered_adapters()
"""
from sequana.expdesign import ExpDesignAdapter
self.design = ExpDesignAdapter(design_filename)
if self.design.df.index.name == "Sample_ID" or \
"Sample_ID" in self.design.df.columns:
self.design.df.set_index("Sample_ID", inplace=True)
else:
raise ValueError("Incorrect design file. Missing Sample_ID field")
self.adapters = adapters
try:
file1 = sequana_data("adapters_%s_fwd.fa" % adapters)
logger.warning("rename your file removing prefix adatper")
except:
file1 = sequana_data("%s_fwd.fa" % adapters)
try:
file2 = sequana_data("adapters_%s_revcomp.fa" % adapters)
logger.warning("rename your file removing prefix adatper")
except:
file2 = sequana_data("%s_revcomp.fa" % adapters)
self._adapters_fwd = AdapterReader(file1)
self._adapters_revc = AdapterReader(file2) # !!! revcomp
def _block2docstring(self, section):
if section not in self.sections.keys():
logger.warning("%s not found in the yaml " % section)
return
comments = self.sections[section]
docstring = []
for line in comments.split("\n"):
if "#############" in line:
pass
elif sum([this in line for this in self._specials]):
pass
else:
if len(line) < 2: # an empty line (to keep)
docstring.append("")
else:
docstring.append(line[2:]) # strip the "# "characters
docstring = "\n".join(docstring).strip()
return docstring
def _block2docstring(self, section):
if section not in self.sections.keys():
logger.warning("%s not found in the yaml " % section)
return
comments = self.sections[section]
docstring = []
for line in comments.split("\n"):
if "#############" in line:
pass
elif sum([this in line for this in self._specials]):
pass
else:
if len(line)<2: # an empty line (to keep)
docstring.append("")
else:
docstring.append(line[2:]) # strip the "# "characters
docstring = "\n".join(docstring).strip()
return docstring
def scanner(self):
data = {}
# shlex removes all white lines and split by return carriage
# strip is also applied
rawdata = shlex.split(open(self.filename, "r"))
for line in rawdata:
# sometimes, IEM will store the ;;; at the end
# so we can get [HEADER];;;;;;;;;;;
if line.startswith('[') and "]" in line:
line = line.strip(";").strip(",").strip()
currentkey = line.replace("[", "").replace("]", "")
data[currentkey] = []
else:
data[currentkey].append(line)
for key in data.keys():
data[key] = "\n".join(data[key])
for this in ["Header", "Reads", "Settings", "Data"]:
if this not in data.keys():
logger.warning("%s not found in the DesignExpMiSeq file" %
this)
self.data = data
self.df = pd.read_csv(io.StringIO(data["Data"]))
ncols = [8, 9, 10, 12]
if self.df.shape[1] not in ncols:
self.df = pd.read_csv(io.StringIO(data["Data"]), ";")
if self.df.shape[1] not in ncols:
logger.warning(
"Data section must have 10 or 12 columns. Check the samplesheet"
)
# Fixes https://github.com/sequana/sequana/issues/507
self.df["Sample_ID"] = self.df["Sample_ID"].astype(str)
self.df.rename(columns={
"I7_Index_ID": "Index1_ID",
"index": "Index1_Seq",
"I5_Index_ID": "Index2_ID",
"index2": "Index2_Seq"
},
inplace=True)
def splitter_mapped_unmapped(self, filename, prefix):
# helpful resources:
# https://broadinstitute.github.io/picard/explain-flags.html
logger.info("Creating 2 files (mapped and unmapped reads)")
data = SAM(filename)
results = {"flags": [], "mapped": 0, "unmapped": 0, "bad": 0}
logger.info("Please wait while creating output files")
with open("{}/{}.unmapped.fastq".format(self.outdir, prefix),
"w") as fnosirv:
with open("{}/{}.mapped.fastq".format(self.outdir, prefix),
"w") as fsirv:
for a in data:
if a.flag & 2048: # suppl
# a bad read, we can just drop it
results['bad'] += 1
elif a.flag & 1024: # PCR duplicate
results['bad'] += 1
elif a.flag & 256: # secondary alignment
results["bad"] += 1
elif a.flag & 16: # mapped
read = "@{}\n{}\n+\n{}\n".format(
a.qname, a.query_sequence, a.qual)
assert len(a.query_sequence) == len(a.qual)
fsirv.write(read)
results["mapped"] += 1
elif a.flag & 4: # unmapped
read = "@{}\n{}\n+\n{}\n".format(
a.qname, a.query_sequence, a.qual)
assert len(a.query_sequence) == len(a.qual)
fnosirv.write(read)
results["unmapped"] += 1
elif a.flag == 0: # mapped
read = "@{}\n{}\n+\n{}\n".format(
a.qname, a.query_sequence, a.qual)
assert len(a.query_sequence) == len(a.qual)
fsirv.write(read)
results["mapped"] += 1
else:
logger.warning("{} flag not handled".format(a.flag))
results["flags"].append(a.flag)
return results
def _scanner(self):
current_section = None
data = collections.defaultdict(list)
with open(self.filename, "r") as fin:
for line in fin.readlines():
line = self._line_cleaner(line)
if len(line) == 0:
continue
if line.startswith("[") and line.endswith("]"):
name = line.lstrip("[").rstrip("]")
current_section = name
else:
data[current_section] += [line]
if "Header" not in data.keys():
logger.warning("Input file must contain [Header]")
if "Data" not in data.keys():
logger.warning("Input file must contain [Data]")
self.data = data
def _plot(self, Xr, pca=None, pc1=0, pc2=1, colors=None, show_labels=True):
if colors is None:
colors = [self.colors[k] for k in self.labels]
if len(colors) != len(Xr):
colors = ["r"] * len(Xr[:,0])
else:
for k in self.labels:
if k not in colors.keys():
logger.warning("No key color for this sample: {}. Set to red".format(k))
colors[k] = "r"
colors = [colors[k] for k in self.labels]
pylab.scatter(Xr[:,pc1], Xr[:,pc2], c=colors)
ax = pylab.gca()
X1, X2 = pylab.xlim()
dX = X2 - X1
pylab.xlim([X1 + X1*0.05, X2 + X2*0.05])
Y1, Y2 = pylab.ylim()
dY = Y2 - Y1
pylab.ylim([Y1 + Y1*0.05, Y2 + Y2*0.05])
count = 0
if show_labels:
for x,y in zip(Xr[:,pc1], Xr[:,pc2]):
x += dX / 40
y += dY / 40
ax.annotate(self.labels[count], (x,y))
count += 1
if count > 100:
break
if pca:
pylab.xlabel("PC{} ({}%)".format(pc1+1,
round(pca.explained_variance_ratio_[pc1]*100, 2)))
pylab.ylabel("PC{} ({}%)".format(pc2+1,
round(pca.explained_variance_ratio_[pc2]*100, 2)))
pylab.grid(True)
def _download_kraken_toydb(self, verbose=True):
"""Download the kraken DB toy example from sequana_data into
.config/sequana directory
Checks the md5 checksums. About 32Mb of data
"""
dv = DevTools()
base = sequana_config_path + os.sep + "kraken_toydb"
taxondir = base + os.sep + "taxonomy"
dv.mkdir(base)
dv.mkdir(taxondir)
baseurl = "https://github.com/sequana/data/raw/master/"
# download only if required
logger.info("Downloading the database into %s" % base)
md5sums = [
"28661f8baf0514105b0c6957bec0fc6e",
"97a39d44ed86cadea470352d6f69748d",
"d91a0fcbbc0f4bbac918755b6400dea6",
"c8bae69565af2170ece194925b5fdeb9"]
filenames = [
"database.idx",
"database.kdb",
"taxonomy/names.dmp",
"taxonomy/nodes.dmp"]
for filename, md5sum in zip(filenames, md5sums):
url = baseurl + "kraken_toydb/%s" % filename
filename = base + os.sep + filename
if os.path.exists(filename) and md5(filename) == md5sum:
logger.warning("%s already present" % filename)
else:
logger.info("Downloading %s" % url)
wget(url, filename)
def _download_kraken_toydb(self, verbose=True):
"""Download the kraken DB toy example from sequana_data into
.config/sequana directory
Checks the md5 checksums. About 32Mb of data
"""
dv = DevTools()
base = sequana_config_path + os.sep + "kraken_toydb"
taxondir = base + os.sep + "taxonomy"
dv.mkdir(base)
dv.mkdir(taxondir)
baseurl = "https://github.com/sequana/data/raw/master/"
# download only if required
logger.info("Downloading the database into %s" % base)
md5sums = [
"28661f8baf0514105b0c6957bec0fc6e",
"97a39d44ed86cadea470352d6f69748d",
"d91a0fcbbc0f4bbac918755b6400dea6",
"c8bae69565af2170ece194925b5fdeb9"
]
filenames = [
"database.idx", "database.kdb", "taxonomy/names.dmp",
"taxonomy/nodes.dmp"
]
for filename, md5sum in zip(filenames, md5sums):
url = baseurl + "kraken_toydb/%s" % filename
filename = base + os.sep + filename
if os.path.exists(filename) and md5(filename) == md5sum:
logger.warning("%s already present" % filename)
else:
logger.info("Downloading %s" % url)
wget(url, filename)
def set_links_to_column(self, link_col, target_col):
"""Hide a column with urls and connect it with a column.
:param str link_col: column with your URLs.
:param str target_col: column to connect.
"""
# hide the link column
try:
self.datatable_columns[link_col]['visible'] = 'false'
except KeyError:
logger.warning("KeyError: Column name '{0}' does not exist."
.format(target_col))
pass
# function to add link
fct = """function(data, type, row, meta){{
return '<a href="'+row.{0}+'" target="_blank">'+data+'</a>';
}}
""".format(link_col)
try:
self.datatable_columns[target_col]['render'] = fct
except KeyError:
logger.warning("KeyError: Column name '{0}' does not exist."
.format(target_col))
pass
def set_tooltips_to_column(self, tooltips_col, target_col):
"""Hide a column with tooltips and connect it with a column.
:param str tooltips_col: column with your tooltips.
:param str target_col: column to connect.
"""
# hide tooltips
try:
self.datatable_columns[tooltips_col]['visible'] = 'false'
except KeyError:
logger.warning("KeyError: Column name '{0}' does not exist."
.format(target_col))
pass
# function to add tooltips
fct = """function(data, type, row, meta){{
return '<a href="#" data-toggle="tooltip" title="'+row.{0}+'">'+data+'</a>';
}}
""".format(tooltips_col)
try:
self.datatable_columns[target_col]['render'] = fct
except KeyError:
logger.warning("KeyError: Column name '{0}' does not exist."
.format(target_col))
pass
def kraken_to_krona(self, output_filename=None, mode=None, nofile=False):
"""
:return: status: True is everything went fine otherwise False
"""
if output_filename is None:
output_filename = self.filename + ".summary"
taxon_to_find = list(self.taxons.index)
if len(taxon_to_find) == 0:
logger.warning("No reads were identified. You will need a more complete database")
self.output_filename = output_filename
with open(output_filename, "w") as fout:
fout.write("%s\t%s" % (self.unclassified, "Unclassified"))
return False
# classified reads as root (1)
"""try:
logger.warning("Removing taxon 1 (%s values) " % self.taxons.ix[1])
logger.info("Found %s taxons " % len(taxon_to_find))
taxon_to_find.pop(taxon_to_find.index(1))
except:
pass
"""
if len(taxon_to_find) == 0:
return False
if mode != "adapters":
df = self.get_taxonomy_biokit(taxon_to_find)
self.lineage = [";".join(this) for this in df[df.columns[0:-1]].values]
self.scnames = list(df['name'].values) # do we need a cast ?
else:
# Let us get the known adapters and their identifiers
from sequana.adapters import AdapterDB
adapters = AdapterDB()
adapters.load_all()
self.scnames = []
for taxon in self.taxons.index:
if str(taxon) in [1, "1"]:
self.scnames.append('unknown')
continue
if str(taxon) not in list(adapters.df.identifier):
self.scnames.append('unknown')
continue
self.scnames.append(adapters.get_name(taxon))
self.lineage = ["Adapters;%s"% x for x in self.scnames]
assert len(self.lineage) == len(self.taxons)
assert len(self.scnames) == len(self.taxons)
# Now save the file
self.output_filename = output_filename
with open(output_filename, "w") as fout:
for i, this in enumerate(self.lineage):
taxon = taxon_to_find[i]
count = self.taxons.loc[taxon]
line = str(count)+"\t"+"\t".join(this.split(';'))
line += " " +self.scnames[i]
fout.write(line+'\n')
try:
fout.write("%s\t%s" % (self.unclassified, "Unclassified"))
except:
pass #unclassified may not exists if all classified
self._data_created = True
return True
def __len__(self):
if self._N is None:
logger.warning("Scanning the BAM. Please wait")
self._N = sum(1 for _ in self._data)
self.reset()
return self._N
def run_analysis(chrom, options, feature_dict):
logger.info("Computing some metrics")
if chrom.DOC < 8:
logger.warning("The depth of coverage is below 8. sequana_coverage is"
" not optimised for such depth. You may want to "
" increase the threshold to avoid too many false detections")
logger.info(chrom.__str__())
if options.w_median > len(chrom.df) / 4:
NW = int(len(chrom.df) / 4)
if NW % 2 == 0:
NW += 1
logger.warning("median window length is too long. \n"
" Setting the window length automatically to a fifth of\n"
" the chromosome length ({})".format(NW))
options.w_median = NW
# compute the running median, zscore and ROIs for each chunk summarizing the
# results in a ChromosomeCovMultiChunk instane
logger.info('Using running median (w=%s)' % options.w_median)
logger.info("Number of mixture models %s " % options.k)
results = chrom.run(options.w_median, options.k,
circular=options.circular, binning=options.binning,
cnv_delta=options.cnv_clustering)
# Print some info related to the fitted mixture models
try:
mu = results.data[0][0].as_dict()['data']['fit_mu']
sigma = results.data[0][0].as_dict()['data']['fit_sigma']
pi = results.data[0][0].as_dict()['data']['fit_pi']
logger.info("Fitted central distribution (first chunk): mu=%s, sigma=%s, pi=%s" %
(round(mu,3), round(sigma,3), round(pi,3)))
except:
pass
# some information about the ROIs found
high = chrom.thresholds.high2
low = chrom.thresholds.low2
logger.info("Searching for ROIs (threshold=[{},{}] ; double =[{},{}])".format(
chrom.thresholds.low, chrom.thresholds.high, low, high))
ROIs = results.get_rois() # results is a ChromosomeCovMultiChunk instane
logger.info("Number of ROIs found: {}".format(len(ROIs.df)))
logger.info(" - below average: {}".format(len(ROIs.get_low_rois())))
logger.info(" - above average: {}".format(len(ROIs.get_high_rois())))
# Create directory and save ROIs
directory = options.output_directory
directory += os.sep + "coverage_reports"
directory += os.sep + chrom.chrom_name
mkdirs(directory)
ROIs.df.to_csv("{}/rois.csv".format(directory))
# save summary and metrics
logger.info("Computing extra metrics")
summary = results.get_summary()
summary.to_json(directory + os.sep + "sequana_summary_coverage.json")
logger.info("Evenness: {}".format(summary.data['evenness']))
logger.info("Centralness (3 sigma): {}".format(summary.data['C3']))
logger.info("Centralness (4 sigma): {}".format(summary.data['C4']))
if options.skip_html:
return
logger.info("Creating report in %s. Please wait" % config.output_dir)
if chrom._mode == "chunks":
logger.warning(("This chromosome is large. "
"Plots in the HTML reports are skipped"))
datatable = CoverageModule.init_roi_datatable(ROIs)
ChromosomeCoverageModule(chrom, datatable,
options={"W": options.w_median,
"k": options.k,
"ROIs": ROIs,
"circular": options.circular},
command=" ".join(["sequana_coverage"] + sys.argv[1:]))
def main(args=None):
if args is None:
args = sys.argv[:]
user_options = Options(prog="sequana")
# If --help or no options provided, show the help
if len(args) == 1:
user_options.parse_args(["prog", "--help"])
else:
options = user_options.parse_args(args[1:])
logger.level = options.logging_level
if options.download_reference:
logger.info("Downloading reference %s from %s\n" %
(options.download_reference, options.database))
from bioservices.apps import download_fasta as df
df.download_fasta(options.download_reference, method=options.database)
if options.download_genbank is None:
return
if options.download_genbank:
logger.info("Downloading genbank %s from %s\n" %
(options.download_genbank, options.database))
from sequana.snpeff import download_fasta_and_genbank
download_fasta_and_genbank(options.download_genbank,
options.download_genbank,
genbank=True, fasta=False)
return
if options.genbank:
assert os.path.exists(options.genbank), \
"%s does not exists" % options.genbank
logger.info("Reading %s. This may take time depending on "
"your input file" % options.input)
# Convert BAM to BED
if options.input.endswith(".bam"):
bedfile = options.input.replace(".bam", ".bed")
logger.info("Converting BAM into BED file")
shellcmd("bedtools genomecov -d -ibam %s > %s" % (options.input, bedfile))
elif options.input.endswith(".bed"):
bedfile = options.input
else:
raise ValueError("Input file must be a BAM or BED file")
# Set the thresholds
if options.low_threshold is None:
options.low_threshold = -options.threshold
if options.high_threshold is None:
options.high_threshold = options.threshold
# and output directory
config.output_dir = options.output_directory
config.sample_name = os.path.basename(options.input).split('.')[0]
# Now we can create the instance of GenomeCoverage
if options.chromosome == -1:
chrom_list = []
else:
chrom_list = [options.chromosome]
gc = GenomeCov(bedfile, options.genbank, options.low_threshold,
options.high_threshold, options.double_threshold,
options.double_threshold, chunksize=options.chunksize,
chromosome_list=chrom_list)
# if we have the reference, let us use it
if options.reference:
logger.info('Computing GC content')
gc.compute_gc_content(options.reference, options.w_gc,
options.circular)
# Now we scan the chromosomes,
if len(gc.chrom_names) == 1:
logger.warning("There is only one chromosome. Selected automatically.")
run_analysis(gc.chr_list[0], options, gc.feature_dict)
elif options.chromosome <-1 or options.chromosome > len(gc.chrom_names):
msg = "invalid chromosome index; must be in [1;{}]".format(len(gc.chrom_names))
logger.error(msg)
sys.exit(1)
else:
if options.chromosome == -1:
chromosomes = gc.chrom_names # take all chromosomes
else:
# For user, we start at position 1 but in python, we start at zero
chromosomes = [gc.chrom_names[options.chromosome-1]]
logger.info("There are %s chromosomes/contigs." % len(gc))
for this in gc.chrom_names:
data = (this, gc.positions[this]["start"], gc.positions[this]["end"])
logger.info(" {} (starting pos: {}, ending pos: {})".format(*data))
# here we read chromosome by chromosome to save memory.
# However, if the data is small.
for i, chrom in enumerate(chromosomes):
logger.info("==================== analysing chrom/contig %s/%s (%s)"
% (i + 1, len(gc), gc.chrom_names[i]))
# since we read just one contig/chromosome, the chr_list contains
# only one contig, so we access to it with index 0
run_analysis(gc.chr_list[i], options, gc.feature_dict)
if options.skip_multiqc is False:
logger.info("=========================")
logger.info("Creating multiqc report")
pathtocfg = sequana_data("multiqc_config.yaml", "../multiqc/")
cmd = 'multiqc . -m sequana_coverage -f -c {}'.format(pathtocfg)
import subprocess
proc = subprocess.Popen(cmd.split(), cwd=options.output_directory)
proc.wait()
def __init__(self, filename_fastq, fof_databases, threads=1,
output_directory="./kraken_hierarchical/",
keep_temp_files=False, force=False):
""".. rubric:: **constructor**
:param filename_fastq: FastQ file to analyse
:param fof_databases: file that contains a list of databases paths
(one per line). The order is important. Note that you may also
provide a list of datab ase paths.
:param threads: number of threads to be used by Kraken
:param output_directory: name of the output directory
:param keep_temp_files: bool, if True, will keep intermediate files
from each Kraken analysis, and save html report at each step
:param bool force: if the output directory already exists, the
instanciation fails so that the existing data is not overrwritten.
If you wish to overwrite the existing directory, set this
parameter to True.
"""
# When running kraken in paired mode and saving the unclassified reads
# in a file, the output file (fastq) contains both R1 and R2 so there
# are concatenated in the same file. Actually, if there is R1 and R2,
# there are concatenated as R1 N R2 (with the letter N as link).
# So, in the hiearchical search, paired case, the first iteration has 2
# input files, must subsequent iterations will have only one file as
# input, that is the output of the previous run (provided by
# --unclassified-out option)
self.filename_fastq = filename_fastq
# input databases may be stored in a file
if isinstance(fof_databases, str) and os.path.exists(fof_databases):
with open(fof_databases, 'r') as fof:
self.databases = [absolute_path.split('\n')[0] for absolute_path in fof.readlines()]
# or simply provided as a list
elif isinstance(fof_databases, list):
self.databases = fof_databases[:]
else:
raise TypeError("input databases must be a list of valid kraken "
"databases or a file (see documebntation)")
self.threads = threads
self.output_directory = output_directory
self.keep_temp_files = keep_temp_files
# check if the output directory already exist
try:
os.mkdir(output_directory)
except OSError:
if os.path.isdir(output_directory) and force is False:
logger.error('Output directory %s already exists' % output_directory)
raise Exception
elif force is True:
logger.warning("Output directory %s already exists. You may "
"overwrite existing results" % output_directory)
# list of input fastq files
if isinstance(filename_fastq, list) and len(filename_fastq) in [1, 2]:
self.inputs = filename_fastq[:]
elif isinstance(filename_fastq, str):
self.inputs = [filename_fastq]
else:
msg = "input file must be a string or list of 2 filenames"
msg += "\nYou provided {}".format(filename_fastq)
raise TypeError(msg)