def test_get_various_datatypes(tmpdir):
collection = GenomeCollection(data_dir=str(tmpdir))
for data_type in ["protein_fasta", "genomic_fasta", "genomic_genbank"]:
path = collection.get_taxid_genome_data_path(
taxid=PHAGE_TAXID, data_type=data_type
)
assert os.path.exists(path)
def test_blast_against_taxid(tmpdir):
collection = GenomeCollection(data_dir=str(tmpdir))
blast_results_file = os.path.join(str(tmpdir), "results.txt")
queries_file = os.path.join("tests", "queries.fa")
assert not os.path.exists(blast_results_file)
collection.blast_against_taxid(
PHAGE_TAXID,
"nucl",
["blastn", "-query", queries_file, "-out", blast_results_file],
)
file_size = os.stat(blast_results_file).st_size
assert 1200 > file_size > 800
def test_avoid_matches_with_phage():
PHAGE_TAXID = "697289"
collection = GenomeCollection()
index = collection.get_taxid_bowtie_index_path(PHAGE_TAXID, version="1")
problem = DnaOptimizationProblem(
sequence=random_dna_sequence(30, seed=123),
constraints=[AvoidMatches(bowtie_index=index, match_length=10)],
logger=None,
)
all_breaches = problem.constraints_evaluations().all_locations()
assert len(all_breaches) == 5
problem.resolve_constraints()
assert problem.all_constraints_pass()
def test_avoid_phage_blast_matches():
PHAGE_TAXID = "697289"
collection = GenomeCollection()
blastdb = collection.get_taxid_blastdb_path(PHAGE_TAXID, db_type="nucl")
problem = DnaOptimizationProblem(sequence=random_dna_sequence(30,
seed=123),
constraints=[
AvoidBlastMatches(blast_db=blastdb,
min_align_length=10,
word_size=7)
],
logger=None)
assert not problem.all_constraints_pass()
problem.resolve_constraints()
assert problem.all_constraints_pass()
def test_get_blast_database(tmpdir):
collection = GenomeCollection(data_dir=str(tmpdir))
taxid = PHAGE_TAXID
path = collection.datafile_path(taxid, data_type="blast_nucl")
assert not os.path.exists(path + ".nsq")
# Test nucleotide database
blast_db_path = collection.get_taxid_blastdb_path(taxid, db_type="nucl")
assert path == blast_db_path
file_size = os.stat(blast_db_path + ".nsq").st_size
assert 50_000 > file_size > 30_000
# Test protein database
blast_db_path = collection.get_taxid_blastdb_path(taxid, db_type="prot")
file_size = os.stat(blast_db_path + ".psq").st_size
assert 60_000 > file_size > 40_000
def test_delete_all_data_files(tmpdir):
collection = GenomeCollection(data_dir=str(tmpdir))
taxids = ["224308", "511145", "559292"]
for taxid in taxids:
collection.get_taxid_infos(taxid)
found_taxids = collection.list_locally_available_taxids("infos")
assert len(found_taxids) == 3
collection.remove_all_local_data_files()
found_taxids = collection.list_locally_available_taxids("infos")
assert len(found_taxids) == 0
def test_autodownload_false(tmpdir):
collection = GenomeCollection(data_dir=str(tmpdir))
collection.autodownload = False
with pytest.raises(FileNotFoundError) as excinfo:
collection.get_taxid_infos("224308")
assert "No infos" in str(excinfo.value)
with pytest.raises(FileNotFoundError) as excinfo:
collection.get_taxid_genome_data_path("224308")
assert "No genome" in str(excinfo.value)
def test_get_biopython_records(tmpdir):
collection = GenomeCollection(data_dir=str(tmpdir))
records = collection.get_taxid_biopython_records(PHAGE_TAXID)
assert len(records) == 1
assert 168000 < len(records[0]) < 170000
"""Example of use for AvoidBlastMatches.
In this example we create a 1000bp random sequence, then edit out every match
with E. coli that is 14bp or longer.
"""
from dnachisel import DnaOptimizationProblem, random_dna_sequence, AvoidMatches
from genome_collector import GenomeCollection
# THIS CREATES THE ECOLI BLAST DATABASE ON YOUR MACHINE IF NOT ALREADY HERE
collection = GenomeCollection()
ecoli_index = collection.get_taxid_bowtie_index_path(511145, version="1")
# DEFINE AND SOLVE THE PROBLEM
problem = DnaOptimizationProblem(
sequence=random_dna_sequence(500, seed=123),
constraints=[
AvoidMatches(bowtie_index=ecoli_index, match_length=15, mismatches=1)
],
)
print(
"Constraints validity before optimization\n",
problem.constraints_text_summary(),
)
print("\nNow resolving the problems\n")
problem.resolve_constraints(final_check=True)
In this example we create a 1000bp random sequence, then edit out every match
with E. coli that is 14bp or longer.
"""
import os
from genome_collector import GenomeCollection
from dnachisel import (
DnaOptimizationProblem,
random_dna_sequence,
AvoidBlastMatches,
)
# THIS CREATES THE ECOLI BLAST DATABASE ON YOUR MACHINE IF NOT ALREADY HERE
collection = GenomeCollection()
ecoli_blastdb = collection.get_taxid_blastdb_path(511145, db_type="nucl")
# DEFINE AND SOLVE THE PROBLEM
problem = DnaOptimizationProblem(
sequence=random_dna_sequence(500, seed=123),
constraints=[
AvoidBlastMatches(
blast_db=ecoli_blastdb,
min_align_length=13,
perc_identity=100,
word_size=5, # The bigger the word size, the faster
e_value=1e20,
# ungapped=False
)
# This is the basic example of the README to get you started.
from genome_collector import GenomeCollection
import subprocess
# GET A BLAST PATH
collection = GenomeCollection()
db_path = collection.get_taxid_blastdb_path(taxid=511145, db_type="nucl")
process = subprocess.run([
"blastn",
"-db",
db_path,
"-query",
"basic_example_queries.fa",
"-out",
"basic_example_results.txt",
],
stderr=subprocess.PIPE)
if process.returncode:
raise OSError("BLAST failed: %s" % process.stderr)
print("All good! see basic_example_results.txt for results.")
