def test_variants_to_protein_sequences_dataframe_protein_sequence_length():
expressed_variants = load_vcf("data/b16.f10/b16.expressed.vcf")
parser = make_protein_sequences_arg_parser()
parser.print_help()
for desired_length in range(9, 20, 3):
args = parser.parse_args([
"--vcf", data_path("data/b16.f10/b16.vcf"),
"--bam", data_path("data/b16.f10/b16.combined.sorted.bam"),
"--max-protein-sequences-per-variant", "1",
"--protein-sequence-length", str(desired_length),
])
df = protein_sequences_dataframe_from_args(args)
eq_(
len(df),
len(expressed_variants),
"Expected %d entries for protein_sequence_length=%d, got %d results: %s" % (
len(expressed_variants),
desired_length,
len(df),
df))
protein_sequences = df["amino_acids"]
print(protein_sequences)
protein_sequence_lengths = protein_sequences.str.len()
assert (protein_sequence_lengths == desired_length).all(), (
protein_sequence_lengths,)
def test_mutant_amino_acids_in_mm10_chrX_8125624_refC_altA_pS460I():
# there are two co-occurring variants in the RNAseq data but since
# they don't happen in the same codon then we're considering the Varcode
# annotation to be correct
# TODO: deal with phasing of variants explicitly so that both
# variant positions are considered mutated
parser = make_protein_sequences_arg_parser()
args = parser.parse_args([
"--vcf", data_path("data/b16.f10/b16.f10.Wdr13.vcf"),
"--bam", data_path("data/b16.f10/b16.combined.sorted.bam"),
"--max-protein-sequences-per-variant", "1",
"--protein-sequence-length", "15"
])
for variant, protein_sequences in protein_sequences_generator_from_args(args):
protein_sequence = protein_sequences[0]
check_mutant_amino_acids(variant, protein_sequence)
def test_mutant_amino_acids_in_mm10_chr9_82927102_refG_altT_pT441H():
# the variant chr9:82927102 G>T occurs right next to T>G so the varcode
# prediction for the protein sequence (Asparagine) will be wrong since
# the correct translation is Histidine
parser = make_protein_sequences_arg_parser()
args = parser.parse_args([
"--vcf", data_path("data/b16.f10/b16.f10.Phip.vcf"),
"--bam", data_path("data/b16.f10/b16.combined.sorted.bam"),
"--max-protein-sequences-per-variant", "1",
"--protein-sequence-length", "15"
])
for variant, protein_sequences in protein_sequences_generator_from_args(args):
protein_sequence = protein_sequences[0]
check_mutant_amino_acids(
variant,
protein_sequence,
expected_amino_acids="H")
def test_mutant_amino_acids_in_mm10_chr9_82927102_refG_altT_pT441H():
# the variant chr9:82927102 G>T occurs right next to T>G so the varcode
# prediction for the protein sequence (Asparagine) will be wrong since
# the correct translation is Histidine
parser = make_protein_sequences_arg_parser()
args = parser.parse_args([
"--vcf",
data_path("data/b16.f10/b16.f10.Phip.vcf"), "--bam",
data_path("data/b16.f10/b16.combined.sorted.bam"),
"--max-protein-sequences-per-variant", "1",
"--protein-sequence-length", "15"
])
for variant, protein_sequences in protein_sequences_generator_from_args(
args):
protein_sequence = protein_sequences[0]
check_mutant_amino_acids(variant,
protein_sequence,
expected_amino_acids="H")
def test_mutant_amino_acids_in_mm10_chrX_8125624_refC_altA_pS460I():
# there are two co-occurring variants in the RNAseq data but since
# they don't happen in the same codon then we're considering the Varcode
# annotation to be correct
# TODO: deal with phasing of variants explicitly so that both
# variant positions are considered mutated
parser = make_protein_sequences_arg_parser()
args = parser.parse_args([
"--vcf",
data_path("data/b16.f10/b16.f10.Wdr13.vcf"), "--bam",
data_path("data/b16.f10/b16.combined.sorted.bam"),
"--max-protein-sequences-per-variant", "1",
"--protein-sequence-length", "15"
])
for variant, protein_sequences in protein_sequences_generator_from_args(
args):
protein_sequence = protein_sequences[0]
check_mutant_amino_acids(variant, protein_sequence)
RNAseq BAM from the same sample. Combine synonymous translations and assign
a read count to each protein sequence.
"""
from __future__ import print_function, division, absolute_import
import logging
import logging.config
import pkg_resources
from isovar.cli.protein_sequences import (
make_protein_sequences_arg_parser,
protein_sequences_dataframe_from_args
)
logging.config.fileConfig(pkg_resources.resource_filename('isovar.cli', 'logging.conf'))
logger = logging.getLogger(__name__)
parser = make_protein_sequences_arg_parser()
parser.add_argument(
"--output",
default="isovar-translate-variants-results.csv",
help="Name of CSV file which contains predicted sequences")
if __name__ == "__main__":
args = parser.parse_args()
logger.info(args)
df = protein_sequences_dataframe_from_args(args)
logger.info(df)
df.to_csv(args.output)
"""
Translate non-synonymous coding variants into mutant protein sequences using an
RNAseq BAM from the same sample. Combine synonymous translations and assign
a read count to each protein sequence.
"""
from __future__ import print_function, division, absolute_import
import logging
import logging.config
import pkg_resources
from isovar.cli.protein_sequences import (make_protein_sequences_arg_parser,
protein_sequences_dataframe_from_args
)
logging.config.fileConfig(
pkg_resources.resource_filename('isovar.cli', 'logging.conf'))
logger = logging.getLogger(__name__)
parser = make_protein_sequences_arg_parser()
parser.add_argument("--output",
default="isovar-translate-variants-results.csv",
help="Name of CSV file which contains predicted sequences")
if __name__ == "__main__":
args = parser.parse_args()
logger.info(args)
df = protein_sequences_dataframe_from_args(args)
logger.info(df)
df.to_csv(args.output)
