def setup(self):
self._testfile = utils.get_temp_filename('test.fa')
shutil.copy(utils.get_test_data('test.fa'), self._testfile)
screed.read_fasta_sequences(self._testfile)
self._db = screed.ScreedDB(self._testfile)
self._ns = nostring()
def setup(self):
self._fileName = os.path.join(os.path.dirname(__file__), 'fastaRecovery')
self._testfa = os.path.join(os.path.dirname(__file__), 'test.fa')
screed.read_fasta_sequences(self._testfa)
screed.ToFasta(self._testfa, self._fileName)
screed.read_fasta_sequences(self._fileName)
self.db = screed.ScreedDB(self._fileName)
def setup(self):
self._testfile = utils.get_temp_filename('test.fa')
shutil.copy(utils.get_test_data('test.fa'), self._testfile)
screed.read_fasta_sequences(self._testfile)
self._db = screed.ScreedDB(self._testfile)
self._ns = nostring()
def setup(self):
self._fileName = os.path.join(os.path.dirname(__file__),
'fastaRecovery')
self._testfa = os.path.join(os.path.dirname(__file__), 'test.fa')
screed.read_fasta_sequences(self._testfa)
screed.ToFasta(self._testfa, self._fileName)
screed.read_fasta_sequences(self._fileName)
self.db = screed.ScreedDB(self._fileName)
def setup(self):
self._fileName = utils.get_temp_filename('fastaRecovery')
self._testfa = utils.get_temp_filename('test.fa')
shutil.copy(utils.get_test_data('test.fa'), self._testfa)
screed.read_fasta_sequences(self._testfa)
screed.ToFasta(self._testfa, self._fileName)
screed.read_fasta_sequences(self._fileName)
self.db = screed.ScreedDB(self._fileName)
def add_reads( self, reads, read_name_sep='_' ) :
if exists( reads + '_screed' ) :
print 'reads previously indexed.'
else :
print 'indexing records...'
read_fasta_sequences(reads)
print 'building database...'
db = ScreedDB(reads)
self.db = db
self.read_name_sep = read_name_sep
self.reads_path = reads
def setup(self):
thisdir = os.path.dirname(__file__)
self._fqName = os.path.join(thisdir, 'fa_to_fq')
self._faName = os.path.join(thisdir, 'fq_to_fa')
self._testfa = os.path.join(thisdir, 'test.fa')
screed.read_fasta_sequences(self._testfa)
screed.ToFastq(self._testfa, self._fqName) # Fasta db -> fasta text
screed.read_fastq_sequences(self._fqName) # Fastq file -> fastq db
screed.ToFasta(self._fqName, self._faName) # Fastq db -> fasta text
screed.read_fasta_sequences(self._faName) # Fasta file -> fasta db
self.db = screed.ScreedDB(self._faName)
def setup(self):
self._fqName = utils.get_temp_filename('fa_to_fq')
self._faName = utils.get_temp_filename('fq_to_fa')
self._testfa = utils.get_temp_filename('test.fa')
shutil.copy(utils.get_test_data('test.fa'), self._testfa)
screed.read_fasta_sequences(self._testfa)
screed.ToFastq(self._testfa, self._fqName) # Fasta db -> fasta text
screed.read_fastq_sequences(self._fqName) # Fastq file -> fastq db
screed.ToFasta(self._fqName, self._faName) # Fastq db -> fasta text
screed.read_fasta_sequences(self._faName) # Fasta file -> fasta db
self.db = screed.ScreedDB(self._faName)
def setup(self):
self._fqName = utils.get_temp_filename('fa_to_fq')
self._faName = utils.get_temp_filename('fq_to_fa')
self._testfa = utils.get_temp_filename('test.fa')
shutil.copy(utils.get_test_data('test.fa'), self._testfa)
screed.read_fasta_sequences(self._testfa)
screed.ToFastq(self._testfa, self._fqName) # Fasta db -> fasta text
screed.read_fastq_sequences(self._fqName) # Fastq file -> fastq db
screed.ToFasta(self._fqName, self._faName) # Fastq db -> fasta text
screed.read_fasta_sequences(self._faName) # Fasta file -> fasta db
self.db = screed.ScreedDB(self._faName)
def split(fasta_fp, ref_fp, output_dp):
if not os.path.exists(output_dp): os.makedirs(output_dp)
fasta_db = screed.read_fasta_sequences(fasta_fp)
n = 0
first_entry = True
for i, seq in enumerate(fasta_db):
if i == 0:
output = open(os.path.join(output_dp, 'tmp_{}.fasta'.format(n)),
'w+')
elif i % split_num == 0:
n += 1
output.close()
first_entry = True
output = open(os.path.join(output_dp, 'tmp_{}.fasta'.format(n)),
'w+')
if first_entry:
output.write('>{}\n{}'.format(fasta_db[seq].name,
str(fasta_db[seq].sequence)))
first_entry = False
else:
output.write('\n>{}\n{}'.format(fasta_db[seq].name,
str(fasta_db[seq].sequence)))
output.close()
def main():
if os.path.isfile(fa + "_screed"):
from screed import ScreedDB
fadb = ScreedDB(fa)
else:
fadb = screed.read_fasta_sequences(fa)
makeKmerArray(fadb, int(args.ksize), norm)
def main():
if os.path.isfile(fa + "_screed"):
from screed import ScreedDB
fadb = ScreedDB(fa)
else:
fadb = screed.read_fasta_sequences(fa)
makeSlices(fadb,maxSlice)
def main(): if os.path.isfile(fa + "_screed"): from screed import ScreedDB fadb = ScreedDB(fa) else: fadb = screed.read_fasta_sequences(fa) makeKmerArray(fadb,int(args.ksize),norm)
def openDB(fileName):
"""Opening screed DB; making if not already existing
Args:
fileName -- Name of sequence file or screedDB file
"""
logging.info('{}: Making/opening screed database for: "{}"'.format(my_time(), fileName))
# making db if needed
if not fileName.endswith('_screed'):
try:
screed.read_fastq_sequences(fileName)
fileName = fileName + '_screed'
except KeyError:
try:
screed.read_fasta_sequences(fileName)
fileName = fileName + '_screed'
except IOError:
msg = 'Cannot open {}'.format(fileName)
raise IOError(msg)
# init screed db
return screed.ScreedDB(fileName)
def build_get_hit_length_function(referenceLengths):
"""
Given the referenceLengths parameter return a lambda function that will
map a reference sequence id to its sequence length
The referenceLenths parameter may be either a python dict or a str name
of a fasta file. In the latter case, the file is parsed to get lengths
"""
if isinstance(referenceLengths, str):
import screed
# assume we have the path to a fasta file
# has it been parsed by screed?
if not os.path.exists("%s_screed" % (referenceLengths)):
# TODO: just use Bio.SeqIO to get lengths if
# screed module or screed index is missing.
# screed is overkill here.
screed.read_fasta_sequences(referenceLengths)
refScreed = screed.ScreedDB(referenceLengths)
return lambda h: len(refScreed[h]['sequence'])
else:
return lambda h: referenceLengths[h]
def save_reads_only_in_k2g(index):
k2g = k2g_fastas[index]
no_k2g = os.path.join(NO_K2G_FASTA_PATH, os.path.basename(k2g))
record_names_no_k2g = get_record_names(no_k2g)
filename = os.path.join(
K2G_INTERSECT,
os.path.basename(k2g).replace(".fasta", "_intersect.fasta"))
result_fasta = open(filename, "a")
print(record_names_no_k2g[:5])
print(k2g)
db_k2g = screed.read_fasta_sequences(k2g)
for name in db_k2g:
if name not in record_names_no_k2g:
continue
result_fasta.write(">{}\n{}\n".format(name, db_k2g[name].sequence))
def psl_filter(fasta_fp, psl_fp, output_fp):
hits = pd.read_csv(psl_fp, sep='\t', names=['matches', 'misMatches', 'repMatches', 'nCount',
'qNumInsert', 'qBaseInsert', 'tNumInsert', 'tBaseInsert',
'strand', 'qName', 'qSize', 'qStart', 'qEnd',
'tName', 'tSize', 'tStart', 'tEnd',
'blockCount', 'blockSizes', 'qStarts', 'tStarts'])
fasta_db = screed.read_fasta_sequences(fasta_fp)
output = open(output_fp, 'w+')
first_entry = True
for seq in fasta_db:
if seq in hits['qName'].unique():
if first_entry:
output.write('>{}\n{}'.format(fasta_db[seq].name, str(fasta_db[seq].sequence)))
first_entry = False
else:
output.write('\n>{}\n{}'.format(fasta_db[seq].name, str(fasta_db[seq].sequence)))
output.close()
"""
import time as t
import os, os.path
import glob
import platform
global array
import screed
import sys
from screed import ScreedDB
import string
#corriendo = "N"
bioinfo_path = "/Users/ivanjimenez/Desktop/CLASES/INTERNSHIPS/BIOINFO INTERNSHIP FILES/RESULTS/newresults/"
viralgenome_path = bioinfo_path + "copy_birna_x_virus.fa"
screed.read_fasta_sequences("/Users/ivanjimenez/Desktop/CLASES/INTERNSHIPS/BIOINFO INTERNSHIP FILES/RESULTS/newresults/copy_birna_x_virus.fa")
birna_x_virusdb = ScreedDB(viralgenome_path + "_screed")
#Setting the number of mismatches that are allowed...
k = 6
def getpath():
wd = os.path.dirname(os.path.abspath(__file__))
if platform.system() == 'Windows':
array = wd.split('\\')
destination = "\\\\".join(array)
destination += '\\\\'
else:
array = wd.split('//')
destination = "////".join(array)
destination += '////'
def run(reads_dp, mothur_dp, dependencies_dp, num_cpu):
"""
TODO: save outputs to independent directory for easier use by downstream processes
"""
if not os.path.exists(reads_dp):
sys.exit('read_dp {} DOES NOT EXIST'.format(reads_dp))
if not os.path.exists(mothur_dp):
os.makedirs(mothur_dp)
if not os.path.exists(dependencies_dp):
sys.exit('dependencies_dp {} DOES NOT EXIST'.format(dependencies_dp))
generate_stability_file(reads_dp, mothur_dp)
if not os.path.exists(
os.path.join(dependencies_dp, 'silva',
'silva.seed_v132.pcr.align')):
cmd = [
'''mothur "#pcr.seqs(fasta={}, start=11894, end=25319, keepdots=F, processors={});"'''
.format(
os.path.join(dependencies_dp, 'silva',
'silva.seed_v132.align'), num_cpu)
]
call(cmd, shell=True)
if not os.path.exists(
os.path.join(dependencies_dp, 'silva',
'silva.seed_v132.pcr.align')):
sys.exit('silva.seed_v132.pcr.align NOT CREATED.')
miniconda_bin_dp = os.path.join(dependencies_dp, 'mothur', 'bin')
# make contigs
cmd = [
'''mothur "#set.dir(input={}, tempdefault={});
make.contigs(file='stability.files',
processors={});
screen.seqs(fasta=current,
group=current,
maxambig=0,
maxlength=275,
minlength=240);
summary.seqs(count=current);
unique.seqs(fasta=current);
count.seqs(name=current,
group=current);
align.seqs(fasta=current,
reference={});
summary.seqs(fasta=current,
count=current);
screen.seqs(fasta=current,
count=current,
summary=current,
start=1968,
end=11550,
maxhomop=8);
summary.seqs(count=current,
count=current);
filter.seqs(fasta=current,
vertical=T,
trump=.);
unique.seqs(fasta=current,
count=current);
pre.cluster(fasta=current,
count=current,
diffs=2);
chimera.vsearch(fasta=current,
count=current,
dereplicate=t);
remove.seqs(fasta=current, accnos=current);
summary.seqs(fasta=current,
count=current);
classify.seqs(fasta=current,
count=current,
reference={},
taxonomy={},
cutoff=80);
remove.lineage(fasta=current,
count=current,
taxonomy=current,
taxon=Chloroplast-Mitochondria-unknown-Archaea-Eukaryota);
summary.seqs(fasta=current,
count=current);
cluster.split(fasta=current,
count=current,
taxonomy=current,
splitmethod=classify,
taxlevel=4,
cutoff=0.03);
make.shared(list=current,
count=current,
label=0.03);
classify.otu(list=current,
count=current,
taxonomy=current,
label=0.03);
tree.shared(shared=current,
calc=jest-thetayc-braycurtis);
get.oturep(column=current,
name=current,
fasta=current,
list=current);"'''.format(
mothur_dp, miniconda_bin_dp, num_cpu,
os.path.join(dependencies_dp, 'silva',
'silva.seed_v132.pcr.align'),
os.path.join(dependencies_dp, 'silva', 'silva.nr_v132.align'),
os.path.join(dependencies_dp, 'silva', 'silva.nr_v132.tax'))
]
call(cmd, shell=True)
#db = screed.read_fasta_sequences(os.path.join(mothur_dp, 'stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.opti_mcc.unique_list.0.03.rep.fasta'))
db = screed.read_fasta_sequences(
os.path.join(
mothur_dp,
'stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.opti_mcc.0.03.rep.fasta'
))
output = open(os.path.join(mothur_dp, '../', 'otus.fasta'), 'w+')
for otu in db:
output.write('>{}\n'.format(otu.split('\t')[1].split('|')[0]))
output.write('{}\n'.format(db[otu].sequence))
output.close()
return None
def setup(self):
self._testfa = os.path.join(os.path.dirname(__file__),
'test-whitespace.fa')
screed.read_fasta_sequences(self._testfa)
self.db = screed.ScreedDB(self._testfa)
def setup(self):
self._testfa = os.path.join(os.path.dirname(__file__), 'test.fa')
screed.read_fasta_sequences(self._testfa)
self.db = screed.ScreedDB(self._testfa)
def plotHitCoverageByLengthBins(ax, lengths, hits, referenceLengths, bins=20, lengthRange=None, barcolor='b',baredgecolor='k',hlog=False,hcolor='r', includeMissed=False):
"""
Given a dictionary of transcript lengths, a dictionary of hits, and a dict of reference sequence lengths...
Produce a plot of reference coverate rate by length bin. IE: What fracton of total residues in the reference sequences were matched.
The param referenceLengths can be a dictionary from hit names to lengths or a fasta file of sequences. The names in both should match the hit names in the "hits" dictionary.
"""
import screed
# Don't try to plot empty data
if len(lengths)==0:
raise Exception("Lengths cannot be empty!")
transcriptCounts,boundaries=numpy.histogram(lengths.values(), bins=bins, range=lengthRange)
if isinstance(referenceLengths, str):
# assume we have the path to a fasta file
# has it been parsed by screed?
if not os.path.exists("%s_screed" % (referenceLengths)):
screed.read_fasta_sequences(referenceLengths)
refScreed=screed.ScreedDB(referenceLengths)
getHitLength=lambda h: len(refScreed[h]['sequence'])
else:
getHitLength=lambda h: referenceLengths[h]
# count bases by bin
hitBaseCounts=numpy.zeros(transcriptCounts.shape)
referenceBaseCounts=numpy.zeros(transcriptCounts.shape)
totalBaseCounts=numpy.zeros(transcriptCounts.shape)
for transcript,hitList in hits.iteritems():
try:
index=getBin(lengths[transcript],boundaries)
except ValueError:
# length was outside range
continue
totalBaseCounts[index]+=lengths[transcript]
firstHit=hitList[0]
hitLength=getHitLength(firstHit.hit)
logger.debug("Hit of length %d goes from %d to %d" % (hitLength,
firstHit.hstart, firstHit.hend))
referenceBaseCounts[index]+=hitLength
hitBaseCounts[index]+=numpy.abs(firstHit.hend-firstHit.hstart)+1
if includeMissed:
for transcript,length in lengths.iteritems():
if transcript not in hits:
totalBaseCounts[index]+=length
# Simulate stepped histogram of total bases
ax2=ax.twinx()
x,y = getSteppedBars(totalBaseCounts, boundaries)
if hlog:
ax2.set_yscale("log",nonposy='clip')
ax2.plot(x,y,color=hcolor)
ax2.set_ylabel('total bases',color=hcolor)
for tl in ax2.get_yticklabels():
tl.set_color(hcolor)
# normalize hit counts by transcript counts
hitRate = hitBaseCounts/referenceBaseCounts
# remove infinities
hitRate[totalBaseCounts==0]=0
# Draw histogram bars
lefts=boundaries[:-1]
widths=[boundaries[i+1]-boundaries[i] for i in range(len(boundaries)-1)]
ax.bar(lefts,hitRate,width=widths,color=barcolor,edgecolor=baredgecolor)
ax.set_ylim([0,1])
ax.set_ylabel('% reference matched')
ax.set_xlabel('transcript length')
def setup(self):
self._testfa = utils.get_temp_filename('test-whitespace.fa')
shutil.copy(utils.get_test_data('test-whitespace.fa'), self._testfa)
screed.read_fasta_sequences(self._testfa)
self.db = screed.ScreedDB(self._testfa)
def setup():
screed.read_fasta_sequences(testfa)
#! /usr/bin/env python
import sys
import os
import screed
filename = sys.argv[1]
try:
os.unlink(filename + '_screed')
except OSError:
pass
db = screed.read_fasta_sequences(filename)
###
from whoosh.index import create_in
from whoosh.fields import *
schema = Schema(name=TEXT(stored=True),
description=TEXT(stored=True))
import os, shutil
indexdir = filename + '.whooshd'
try:
shutil.rmtree(indexdir)
except OSError: # doesn't exit
pass
os.mkdir(indexdir)
ix = create_in(indexdir, schema)
def setup():
screed.read_fasta_sequences(testfa)
def generate_tetra(project, scaffolds, min_len=2000):
"""
gene_file: HMMER predicted genes.
"""
#TODO: screed is only for python 2, so can't use this on py3
#TODO: also screed fails if two instances try to open the same file
import screed
gene_file = project + '_hmm.txt'
tetra_file = project + '_tetra.txt'
gff_file = project + '_mgm.gff'
# generate a mapping to merge reverse complement tetranucleotides
seq_map = {'': 4 * 'N'}
for s in (''.join(i) for i in itertools.product(*(4 * ['ATGC']))):
if rc(s) not in seq_map or s not in seq_map:
seq_map[s] = s
seq_map[rc(s)] = s
# write out the headers for the gene/tetranucleotide file
fout = open(tetra_file, 'w')
srted_vals = list(set(seq_map.values()))
srted_vals.sort()
fout.write(','.join(['Gene'] + srted_vals) + '\n')
f = open(gff_file, "r")
for i in range(7):
f.readline()
# make a dictionary to map genes back to their contigs
gene2contig = {}
for ln in f:
flds = ln.strip().split('\t')
gene2contig[flds[-1].replace(' ', '_')] = flds[0].split(' ')[0]
f.close()
hmm = open(gene_file, 'r')
contigs = screed.read_fasta_sequences(scaffolds)
for i in range(3):
hmm.readline()
p_gff_name = ''
for ln in hmm:
gene_name = ln[0:21].strip()
gff_name = ln[32:53].strip()
# if the same gene_id is listed multiple times in a row
# that means HMMER found multiple matches for it. We only
# want the first one (with the lowest E-value).
if gff_name != p_gff_name:
p_gff_name = gff_name
cc = contigs[gene2contig[gff_name]]
if len(cc.sequence) < min_len:
continue
frq = dict([(s, 0) for s in seq_map.values()])
for ss in slid_win(str(cc.sequence).upper(), 4):
frq[seq_map.get(ss, 'NNNN')] += 1
sum_frq = float(sum(frq.values()))
if sum_frq == 0:
sum_frq = 1
fout.write(','.join([gene_name] + [str(frq[i] / sum_frq) \
for i in srted_vals]))
fout.write('\n')
fout.flush()
hmm.close()
fout.close()
#! /usr/bin/env python
import screed
db = screed.read_fasta_sequences('galGal4.fa.masked')
keys = [k for k in db.keys() if "_" not in k and "Un" not in k]
filtered = screed.fasta.FASTA_Writer('galGal4.fa.masked.filtered')
for k in keys:
record = db[k]
filtered.write(record)
for k in keys:
record = db[k]
filtered = screed.fasta.FASTA_Writer(k)
filtered.write(record)
#! /usr/bin/env python
import sys
import os
import screed
filename = sys.argv[1]
try:
os.unlink(filename + '_screed')
except OSError:
pass
db = screed.read_fasta_sequences(filename)
###
from whoosh.index import create_in
from whoosh.fields import *
schema = Schema(name=TEXT(stored=True), description=TEXT(stored=True))
import os, shutil
indexdir = filename + '.whooshd'
try:
shutil.rmtree(indexdir)
except OSError: # doesn't exit
pass
os.mkdir(indexdir)
ix = create_in(indexdir, schema)
writer = ix.writer()
def setup(self):
self._testfa = os.path.join(os.path.dirname(__file__), "test-whitespace.fa")
screed.read_fasta_sequences(self._testfa)
self.db = screed.ScreedDB(self._testfa)
def setup(self):
self._testfa = utils.get_temp_filename('test-whitespace.fa')
shutil.copy(utils.get_test_data('test-whitespace.fa'), self._testfa)
screed.read_fasta_sequences(self._testfa)
self.db = screed.ScreedDB(self._testfa)
def get_record_names(path):
print(path)
db = screed.read_fasta_sequences(path)
names = db.keys()
db.close()
return names
