def prepare_fasta(id, _run_id,load_ids, species, seq_type,
molecule_type, genome_type, outdir):
""" Write fasta file adapted for OrthoFinder """
fasta_dir = utils.safe_mkdir('fastas_%s' % (_run_id))
nloads = 0
nseqs = 0
for load_id in load_ids:
nloads +=1
taxon = species[load_id].name
fasta = os.path.join(fasta_dir, taxon + '.fa')
with open(fasta, 'w') as ffasta:
for record in database.load_seqs(
load_id, taxon, seq_type, molecule_type, genome_type):
newid = taxon + "@" + str(record.id)
nseqs += 1
utils.write_fasta(ffasta, record.seq, newid)
if not nseqs:
utils.die("no sequences were written to the FASTA file")
diagnostics.log('nloads', nloads)
diagnostics.log('nseqs', nseqs)
ingest('fasta_dir', 'fasta')
def write_fasta(id, _run_id, load_ids, species, seq_type, molecule_type, genome_type, outdir):
"""Write sequences from the Agalma database to a FASTA file"""
blast_dir = utils.safe_mkdir("allvall_blast_%s_%s" % (id, _run_id))
fasta = os.path.join(blast_dir, "all.fa")
# The nodes file contains a header, which describes the attributes, as well
# as a line for every node (transcript) in the analysis.
nodes = os.path.join(outdir, "nodes.txt")
nloads = 0
nseqs = 0
nbases = 0
with open(nodes, "w") as fnodes, open(fasta, "w") as ffasta:
print >> fnodes, "label\tid\tassembly\tassembly_number"
for load_id in load_ids:
nloads += 1
taxon = species[load_id].name
for record in database.load_seqs(load_id, taxon, seq_type, molecule_type, genome_type):
nseqs += 1
nbases += len(record.seq)
# The id of the node (the second column) is a unique identifier
# and because we already have one in the table, use that here.
print >> fnodes, "%s\t%d\t%s\t%d" % (record.header, record.id, load_id, nloads)
utils.write_fasta(ffasta, record.seq, record.id)
if not nseqs:
utils.die("no sequences were written to the FASTA file")
diagnostics.log_path(nodes, "nodes")
diagnostics.log("nloads", nloads)
diagnostics.log("nseqs", nseqs)
diagnostics.log("nbases", nbases)
ingest("blast_dir", "fasta", "nodes")