def __init__(self, input_file, output_dir, index_name, threads):
which = Cmd("which salmon")
which.run()
self.input_file = input_file
self.output_dir = output_dir
self.index_name = index_name
self.threads = threads
def build_index(self) -> None:
"""Builds the salmon index."""
logger.debug("Build salmon index.")
# TODO: Implement check to avoid duplicate runs
indexing = Cmd(
f"salmon index -p {self.threads} -t {self.input_file} -i {self.index_name} --keepDuplicates"
)
indexing.run()
def __init__(self, input_file, index_name, threads):
which = Cmd("which hisat2")
which.run()
# Check if indexing already run
self.index_build_has_run = (True if len(
glob.glob(f"{index_name}.*.ht2")) == 8 else False)
self.input_file = input_file
self.index_name = index_name
self.threads = threads
def build_index(self):
""" Build the Hisat2 index. """
if not self.index_build_has_run:
logger.debug("Build Hisat2 index.")
indexing = Cmd(
f"hisat2-build -q -p {self.threads} {self.input_file} {self.index_name}"
)
indexing.run()
self.index_build_has_run = True
else:
logger.debug("Skipping index building.")
def update_database(self, database_dir, busco_group):
"""
Updates the dammit database.
"""
logger.info("Update dammit database.")
self.database_dir = database_dir
self.busco_group = busco_group
database = Cmd(
f"dammit databases --install --n_threads {self.threads} --database-dir {self.database_dir} --busco-group {self.busco_group}"
)
database.run()
def run(self, reads):
""" Run the Hisat2 mapping with the given reads. """
logger.debug("Perform Hisat2 mapping.")
if len(reads) == 1: # single end reads
hisat = Cmd(
f"hisat2 -q --threads {self.threads} -k 1 -x {self.index_name} -U {reads[0]} --no-unal | \
samtools view --threads {self.threads} -hS -F 4 -q 1 -O SAM"
)
elif len(reads) == 2: # paired end reads
hisat = Cmd(
f"hisat2 -q --threads {self.threads} -k 1 -x {self.index_name} -1 {reads[0]} -2 {reads[1]} --no-unal | \
samtools view --threads {self.threads} -hS -F 4 -q 1 -O SAM"
)
hisat.run()
self.mapping_has_run = True
return (entry for entry in hisat.stdout.split("\n")[:-1]
if not entry.startswith("@"))
def run(self, graph_file, output_file):
"""
MCL: The input is then a file or stream in which each line encodes an
edge in terms of two labels (the 'A' and the 'B') and a numerical value
(the 'C'), all separated by white space.
A B 20
A C 10
The output is then a file where each line is a cluster of tab-separated
labels.
"""
logger.debug("MCL clustering...")
if os.path.exists(output_file):
os.remove(output_file)
mcl = Cmd(
f"mcl {graph_file} -I {self.inflation} --abc -o {output_file} -te {self.threads} -resource 4 -V all"
)
mcl.run()
def run_pipe(self, graph_file):
"""
Runs the MCL command, but uses stdin as input and stdout as output. Is a
lot faster than writing and reading a lot of files.
MCL: The input is then a file or stream in which each line encodes an
edge in terms of two labels (the 'A' and the 'B') and a numerical value
(the 'C'), all separated by white space.
A B 20
A C 10
The output is then a file where each line is a cluster of tab-separated
labels.
"""
logger.debug("MCL clustering...")
mcl = Cmd(
f"mcl - -I {self.inflation} --abc -o - -te {self.threads} -resource 4 -V all"
)
mcl.run(in_stream=graph_file)
return mcl.stdout
def run(self, reads: list) -> None:
"""Run the salmon mapping with the given reads.
Args:
reads: List of reads. Either paired end or single end.
"""
logger.debug("Perform salmon mapping.")
if not os.path.exists(f"{self.output_dir}/aux_info/eq_classes.txt"):
if len(reads) == 1: # single end reads
salmon = Cmd(
f"salmon quant --libType A --validateMappings --dumpEq -p {self.threads} -i {self.index_name} --unmatedReads {reads[0]} -o {self.output_dir}"
)
elif len(reads) == 2: # paired end reads
salmon = Cmd(
f"salmon quant --libType A --validateMappings --dumpEq -p {self.threads} -i {self.index_name} -1 {reads[0]} -2 {reads[1]} -o {self.output_dir}"
)
salmon.run()
else:
logger.info("Skipping mapping.")
def run(self):
"""
Executes the dammit annotation for the original and reduced fasta file.
"""
logger.info("Run dammit annotation.")
for name, transcriptome in self.transcriptomes.items():
output_dir = f"{self.output_dir}/{name}"
annotation_file = (
f"{output_dir}/{os.path.basename(transcriptome)}.dammit.gff3")
self.gff_files[name] = annotation_file
namemap_file = (
f"{output_dir}/{os.path.basename(transcriptome)}.dammit.namemap.csv"
)
self.namemaps[name] = namemap_file
if not (os.path.exists(annotation_file)
and os.path.exists(namemap_file)):
dammit = Cmd(
f"dammit annotate {transcriptome} -o {output_dir} --database-dir {self.database_dir} --busco-group {self.busco_group} --n_threads {self.threads}"
)
dammit.run()
def __init__(self, threads, inflation):
self.threads = threads
self.inflation = inflation
mcl = Cmd("which mcl")
mcl.run()
""" Benjamin Weeks CS472 Project 2 November, 1st, 2015 Command Line Interface """ from cmd import Cmd cmd = Cmd() cmd.run()