def parse_genomes_fa(fastas, mappings, mask_edges=True):
"""
genomes[genome name] = {order: [contig order], samples: {}}
samples[sample name] = {cov: [coverage by position], contigs: {}}
contigs[contig name] = [coverage by position]
"""
id2g = {} # contig ID to genome lookup
genomes = {} # dictionary for saving genome info
for genome in fastas:
name = genome.name
samples = {s[0]: {'contigs': {}, 'cov': []} for s in mappings}
g = genomes[name] = {'order': [], 'samples': samples}
g['len'] = 0
genome_seq = []
for seq in parse_fasta(genome):
ID = seq[0].split('>', 1)[1].split()[0]
g['order'].append(ID)
id2g[ID] = name
length = len(seq[1])
g['len'] += length
if mask_edges is True:
contig_seq = seq[1][100:len(seq[1]) - 100]
else:
contig_seq = seq[1]
genome_seq.extend(contig_seq)
for sample in list(samples.keys()):
g['samples'][sample]['contigs'][ID] = \
[0 for i in range(0, length)]
g['gc'] = gc_content(genome_seq)
return genomes, id2g
def parse_genomes(fastas, single):
"""
generator for parsing fastas
if single is True, combine sequences in multifasta file
"""
if single is True:
for genome in fastas:
sequence = []
for seq in parse_fasta(genome):
sequence.extend(list(seq[1].upper()))
yield (genome.name.rsplit('.', 1)[0], len(sequence), sequence)
else:
for genome in fastas:
for seq in parse_fasta(genome):
ID = seq[0].split('>', 1)[1].split()[0]
yield (ID, len(seq[1]), list(seq[1].upper()))
def parse_file(file_name):
name, ext = file_name.split('.')
if ext == 'fasta':
data_generator = parse_fasta(file_name)
return data_generator
else:
raise(IOError("sorry no support for %s files yet" % ext))
def parse_genomes(fastas, mappings):
"""
read fastas into dictionary:
genomes[genome] = {order: [contig order], {samples}}
samples[sample] = {cov, {contigs}, window_sum, sliding_sum, sliding_average}
contigs[contig][sample] = [cov]
"""
id2g = {} # contig ID to genome lookup
genomes = {} # dictionary for saving genome info
for genome in fastas:
sequence = []
name = genome.name
samples = {s[0]:{'contigs':{}, 'cov':[]} for s in mappings}
g = genomes[name] = {'order':[], 'samples':samples, 'ORI':[], 'TER':[]}
g['len'] = 0
for seq in parse_fasta(genome):
sequence.extend(list(seq[1].upper()))
ID = seq[0].split('>', 1)[1].split()[0]
g['order'].append(ID)
id2g[ID] = name
length = len(seq[1])
g['len'] += length
cov = [0 for i in range(0, length)]
for sample in samples.keys():
g['samples'][sample]['contigs'][ID] = \
[0 for i in range(0, length)]
g['seq'] = sequence
return genomes, id2g
def parse_genomes(fastas, single):
"""
generator for parsing fastas
if single is True, combine sequences in multifasta file
"""
if single is True:
for genome in fastas:
sequence = []
for seq in parse_fasta(genome):
sequence.extend(list(seq[1].upper()))
yield (genome.name.rsplit('.', 1)[0], len(sequence), sequence)
else:
for genome in fastas:
for seq in parse_fasta(genome):
ID = seq[0].split('>', 1)[1].split()[0]
yield (ID, len(seq[1]), list(seq[1].upper()))
def split_fasta(f, id2f):
"""
split fasta file into separate fasta files based on list of scaffolds
that belong to each separate file
"""
opened = {}
for seq in parse_fasta(f):
id = seq[0].split('>')[1].split()[0].rsplit('_', 1)[0]
if id not in id2f:
continue
fasta = id2f[id]
if fasta not in opened:
opened[fasta] = '%s.faa' % fasta
seq[1] += '\n'
with open(opened[fasta], 'a+') as f_out:
f_out.write('\n'.join(seq))
def main():
parser = argparse.ArgumentParser()
parser.add_argument('databases', nargs='+', type=str, help='databases')
args = parser.parse_args()
out = {
database: fasta.parse_fasta(database)
for database in args.databases
}
organisms = [
organism_name[:-len('.fasta')]
if organism_name.endswith('.fasta') else organism_name
for organism_name in out.keys()
]
averages = [info['average'] for info in out.values()]
stdevs = [info['stdev'] / 2 for info in out.values()]
# Make a dataset:
height = averages
bars = organisms
y_pos = np.arange(len(bars))
# Create bars
plt.bar(y_pos, height)
# Create names on the x-axis
plt.xticks(y_pos, bars, rotation=45)
# Add label on y-axis
plt.ylabel('Average protein sequence length')
# Add title
plt.title(f'Average protein sequence lengths for selected organisms')
# Create error bars based on standard deviation
plt.errorbar(organisms,
averages,
yerr=stdevs,
linestyle='None',
ecolor='red')
# Save and show graphic
plt.savefig(f'avg_length.png')
plt.show()
def n50(fasta):
length_list = []
sequences = []
for sequence in parse_fasta(fasta):
length_list.append(float(len(sequence[1])))
sequences.append(sequence[1])
length_list.sort()
length_list.reverse()
total = float(sum(length_list))
n = total * float(0.50)
n50_value = running = length_list[0]
for length in length_list:
if running >= n:
return n50_value, total, \
len(length_list), gc(''.join(sequences))
else:
n50_value = length
running += n50_value
def setup_genomes(fastas,pileups):
genomes = {}
for genome in fastas:
genomes[genome] = g = {'samples':{}}
g['length'] = 0
g['contig_order'] = []
g['contig_length'] = []
g['masked_bases'] = set()
for seq in parse_fasta(genome):
ID = seq[0].split('>', 1)[1].split()[0]
length = len(seq[1])
g['contig_order'].append(ID)
g['contig_length'].append(length)
g['length'] += length
for sample in pileups:
g['samples'][sample] = {'SNPs':[]}
g['samples'][sample]['cov'] = [0 for i in range(0, g['length'])]
g['samples'][sample]['var'] = [0 for i in range(0, g['length'])]
g['samples'][sample]['unmasked_length'] = 0
return genomes
def setup_genomes(fastas, pileups):
genomes = {}
for genome in fastas:
genomes[genome] = g = {'samples': {}}
g['length'] = 0
g['contig_order'] = []
g['contig_length'] = []
g['masked_bases'] = set()
for seq in parse_fasta(genome):
ID = seq[0].split('>', 1)[1].split()[0]
length = len(seq[1])
g['contig_order'].append(ID)
g['contig_length'].append(length)
g['length'] += length
for sample in pileups:
g['samples'][sample] = {'SNPs': []}
g['samples'][sample]['cov'] = [0 for i in range(0, g['length'])]
g['samples'][sample]['var'] = [0 for i in range(0, g['length'])]
g['samples'][sample]['unmasked_length'] = 0
return genomes
def parse_genomes_fa(fastas, mappings):
"""
genomes[genome name] = {order: [contig order], samples: {}}
samples[sample name] = {cov: [coverage by position], contigs: {}}
contigs[contig name] = [coverage by position]
"""
id2g = {} # contig ID to genome lookup
genomes = {} # dictionary for saving genome info
for genome in fastas:
name = genome.name
samples = {s[0]:{'contigs':{}, 'cov':[]} for s in mappings}
g = genomes[name] = {'order':[], 'samples':samples}
g['len'] = 0
for seq in parse_fasta(genome):
ID = seq[0].split('>', 1)[1].split()[0]
g['order'].append(ID)
id2g[ID] = name
length = len(seq[1])
g['len'] += length
for sample in list(samples.keys()):
g['samples'][sample]['contigs'][ID] = \
[0 for i in range(0, length)]
return genomes, id2g
import sys
from fasta import parse_fasta
filename = sys.argv[1] if len(sys.argv) > 1 else "multispecies_aligned.fasta"
with open(filename,'r') as f:
d = parse_fasta(f.read())
seqs = d.values()
seqlen = len(seqs[0])
n_seqs = len(seqs)
assert all(len(v)==seqlen for v in seqs), \
"Alleles with length != %d: %s" % (
seqlen,
[
(k, len(v))
for (k,v) in d.iteritems()
if len(v) != seqlen
]
conserved = set([])
for i in xrange(seqlen):
count = sum(s[i] != "-" for s in seqs)
alleles = [allele for allele, seq in d.iteritems() if seq[i] != "-"]
species = set([allele.split("-")[0] for allele in alleles])
print i, count, "/", n_seqs, list(sorted(species))
if count < 25:
print "---", alleles
elif count == n_seqs:
conserved.add(i)
def get_prot(uniprot_id):
url = "http://www.uniprot.org/uniprot/%s.fasta" % uniprot_id
resp = requests.get(url)
return fasta.parse_fasta(resp.content)[0].values()[0]
# Add a handle to the last drawn bar, which we'll need for the legend
bars.append(bar[0])
# Draw legend if we need
if legend:
ax.legend(bars, data.keys())
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('databases', nargs='+', type=str, help='databases')
args = parser.parse_args()
out = {
database: fasta.parse_fasta(database)
for database in args.databases
}
organisms = [
organism_name[:-len('.fasta')]
if organism_name.endswith('.fasta') else organism_name
for organism_name in out.keys()
]
frequencies = [{
letter: letter_cnt /
sum([letter_cnt for letter_cnt in info['letters'].values()])
for letter, letter_cnt in info['letters'].items()
} for info in out.values()]
import newick, fasta
def find_rev(t, dnas):
r = []
for i in range(len(dnas[t.u])):
r += [(p[0], p[-1], i, dnas[p[0].u][i]) for p in t.find_rev(dnas, i)]
return r
if __name__ == '__main__':
with open('rosalind_rsub.txt') as f:
nw = f.readline()
nw.split()
tree = newick.newick_parse(nw)
fst = f.read()
dnas, _ = fasta.parse_fasta(fst)
nodes = tree.nodes()
for node in nodes:
revs = find_rev(node, dnas)
for fc, dest, pos, mid in revs:
print("%s %s %d %s->%s->%s" %
(fc.u, dest.u, pos + 1, dnas[node.u][pos], mid,
dnas[dest.u][pos]))
assert (dnas[node.u][pos] == dnas[dest.u][pos])
import fasta
import newick
from pdb import set_trace
INF = 100000000000
with open("rosalind_alph.txt") as f:
nw = f.readline()
fst = f.read()
nw = nw.strip()
tree = newick.newick_parse(nw)
dnas,key = fasta.parse_fasta(fst)
adj_list,children = tree.adj_list()
ordered = tree.level_traverse()
assert(set(ordered) == set(tree.taxa()))
internal = []
for taxon in ordered:
#test
if tree.u != taxon:
assert(len(adj_list[taxon]) == len(children[taxon]) + 1)
else:
assert(len(adj_list[taxon]) == len(children[taxon]))
if len(adj_list[taxon]) > 1:
internal.append(taxon)
for taxon in ordered:
# http://rosalind.info/problems/rsub/
import newick, fasta
def find_rev(t,dnas):
r = []
for i in range(len(dnas[t.u])):
r += [(p[0],p[-1],i,dnas[p[0].u][i]) for p in t.find_rev(dnas,i)]
return r
if __name__ == '__main__':
with open('rosalind_rsub.txt') as f:
nw = f.readline()
nw.split()
tree = newick.newick_parse(nw)
fst = f.read()
dnas,_ = fasta.parse_fasta(fst)
nodes = tree.nodes()
for node in nodes:
revs = find_rev(node,dnas)
for fc, dest, pos, mid in revs:
print("%s %s %d %s->%s->%s" % (fc.u, dest.u, pos + 1, dnas[node.u][pos], mid, dnas[dest.u][pos]))
assert(dnas[node.u][pos] == dnas[dest.u][pos])
def reverse_complement(seq):
rev_c = []
for base in seq[1][::-1]:
if base not in rc:
rev_c.append('N')
else:
rev_c.append(rc[base])
return [seq[0], ''.join(rev_c)]
if __name__ == '__main__':
if len(sys.argv) != 3:
print 'specify fasta or - if from stdin and c (for complement) or rc (for reverse complement)'
exit()
fasta, option = sys.argv[1], sys.argv[2]
if fasta == '-':
fasta = sys.stdin
else:
fasta = open(fasta)
if option == 'c':
for seq in parse_fasta(fasta):
print '\n'.join(complement(seq))
elif option == 'rc':
for seq in parse_fasta(fasta):
print '\n'.join(reverse_complement(seq))
else:
print 'specify fasta or - if from stdin \
and c (for complement) or rc (for reverse complement)'
exit()
