def __init__(self, index="", threads=None):
super().__init__()
self.programName = "STAR"
self.dep_list = [self.programName]
#check if star exists
if not pe.check_dependencies(self.dep_list):
raise Exception("ERROR: " + self.programName + " not found.")
"""
self.valid_args=['--help','--parametersFiles','--sysShell','--runMode','--runThreadN','--runDirPerm','--runRNGseed','--quantMode','--quantTranscriptomeBAMcompression','--quantTranscriptomeBan','--twopassMode','--twopass1readsN',
'--genomeDir','--genomeLoad','--genomeFastaFiles','--genomeChrBinNbits','--genomeSAindexNbases','--genomeSAsparseD','--genomeSuffixLengthMax','--genomeChainFiles','--genomeFileSizes',
'--sjdbFileChrStartEnd','--sjdbGTFfile','--sjdbGTFchrPrefix','--sjdbGTFfeatureExon','--sjdbGTFtagExonParentTranscript','--sjdbGTFtagExonParentGene','--sjdbOverhang','--sjdbScore','--sjdbInsertSave',
'--inputBAMfile','--readFilesIn','--readFilesCommand','--readMapNumber','--readMatesLengthsIn','--readNameSeparator','--clip3pNbases','--clip5pNbases','--clip3pAdapterSeq','--clip3pAdapterMMp','--clip3pAfterAdapterNbases',
'--limitGenomeGenerateRAM','--limitIObufferSize','--limitOutSAMoneReadBytes','--limitOutSJoneRead','--limitOutSJcollapsed','--limitBAMsortRAM ','--limitSjdbInsertNsj','--outFileNamePrefix','--outTmpDir','--outTmpKeep',
'--outStd','--outReadsUnmapped','--outQSconversionAdd','--outMultimapperOrder','--outSAMtype','--outSAMmode','--outSAMstrandField','--outSAMattributes','--outSAMattrIHstart','--outSAMunmapped','--outSAMorder',
'--outSAMprimaryFlag','--outSAMreadID','--outSAMmapqUnique','--outSAMflagOR','--outSAMflagAND','--outSAMattrRGline','--outSAMheaderHD','--outSAMheaderPG','--outSAMheaderCommentFile','--outSAMfilter','--outSAMmultNmax',
'--outBAMcompression','--outBAMsortingThreadN','--bamRemoveDuplicatesType','--bamRemoveDuplicatesMate2basesN','--outWigType','--outWigStrand','--outWigReferencesPrefix','--outWigNorm','--outFilterType',
'--outFilterMultimapScoreRange','--outFilterMultimapNmax','--outFilterMismatchNmax','--outFilterMismatchNoverLmax','--outFilterMismatchNoverReadLmax','--outFilterScoreMin','--outFilterScoreMinOverLread',
'--outFilterMatchNmin','--outFilterMatchNminOverLread','--outFilterIntronMotifs','--outSJfilterReads','--outSJfilterOverhangMin','--outSJfilterCountUniqueMin','--outSJfilterCountTotalMin','--outSJfilterDistToOtherSJmin',
'--outSJfilterIntronMaxVsReadN','--scoreGap','--scoreGapNoncan','--scoreGapGCAG ','--scoreGapATAC','--scoreGenomicLengthLog2scale','--scoreDelOpen','--scoreDelBase','--scoreInsOpen','--scoreInsBase','--scoreStitchSJshift',
'--seedSearchStartLmax','--seedSearchStartLmaxOverLread','--seedSearchLmax','--seedMultimapNmax','--seedPerReadNmax','--seedPerWindowNmax','--seedNoneLociPerWindow','--alignIntronMin','--alignIntronMax','--alignMatesGapMax',
'--alignSJoverhangMin','--alignSJstitchMismatchNmax','--alignSJDBoverhangMin','--alignSplicedMateMapLmin','--alignSplicedMateMapLminOverLmate','--alignWindowsPerReadNmax','--alignTranscriptsPerWindowNmax','--alignTranscriptsPerReadNmax',
'--alignEndsType','--alignEndsProtrude','--alignSoftClipAtReferenceEnds','--winAnchorMultimapNmax','--winBinNbits','--winAnchorDistNbins','--winFlankNbins','--winReadCoverageRelativeMin','--winReadCoverageBasesMin',
'--chimOutType','--chimSegmentMin','--chimScoreMin','--chimScoreDropMax','--chimScoreSeparation','--chimScoreJunctionNonGTAG','--chimJunctionOverhangMin','--chimSegmentReadGapMax','--chimFilter','--chimMainSegmentMultNmax']
"""
if not threads:
threads = os.cpu_count()
self.threads = threads
#if index is passed, update the passed arguments
if index and pu.check_starindex(index):
print("STAR index is: " + index)
self.star_index = index
else:
print(
"No STAR index provided. Please build index now to generate an index using build_index()...."
)
def check_index(self):
if hasattr(self, 'star_index'):
return (pu.check_starindex(self.star_index))
else:
return False
def build_index(self,
index_path,
*args,
threads=None,
overwrite=False,
verbose=False,
quiet=False,
logs=True,
objectid="NA",
**kwargs):
"""Build a star index with given parameters and saves the new index to self.star_index.
Parameters
----------
index_path: string
Path where the index will be created
args: tuple
Path to reference input files
threads: int
Num threads to use
overwrite: bool
Overwrite if index already exists
verbose: bool
Print stdout and std error
quiet: bool
Print nothing
logs: bool
Log this command to pyrpipe logs
objectid: str
Provide an id to attach with this command e.g. the SRR accession. This is useful for debugging, benchmarking and reports.
kwargs: dict
Parameters for the star command
:return: Returns status of star command
:rtype: bool
"""
#if index already exists then exit
if not overwrite:
if pu.check_starindex(index_path):
pu.print_green("STAR index already exists. Using it...")
self.star_index = index_path
return True
#check input files
if len(args) < 1:
pu.print_boldred(
"Please provide input fasta file to build STAR index")
return False
if not pu.check_files_exist(*args):
raise Exception("Please check input to star index")
return False
#create path if doesnt exist
if not pu.check_paths_exist(index_path):
if not pu.mkdir(index_path):
raise Exception("Error creating STAR index. Exiting.")
return False
if not threads:
threads = self.threads
#add runMode
newOpts = {
"--runMode": "genomeGenerate",
"--genomeDir": index_path,
"--genomeFastaFiles": " ".join(args),
"--runThreadN": str(threads)
}
mergedOpts = {**newOpts, **kwargs}
starbuild_Cmd = ['STAR']
starbuild_Cmd.extend(pu.parse_unix_args(None, mergedOpts))
#execute command
status = pe.execute_command(starbuild_Cmd,
verbose=verbose,
quiet=quiet,
logs=logs,
objectid=objectid)
if status:
if pu.check_paths_exist(index_path):
#update object's index
self.star_index = index_path
if self.check_index():
return True
else:
return False
def build_index(self, index_path, genome, objectid="NA"):
"""Build a STAR index with given parameters and saves the new index to self.index.
Parameters
----------
index_path: string
Path where the index will be created
genome: string
Path to the reference genome
objectid : string
Provide an id to attach with this command e.g. the SRR accession. This is useful for debugging, benchmarking and reports.
:return: Returns the status of STAR-build index
:rtype: bool
"""
#if index already exists then exit
if not _force:
if pu.check_starindex(index_path):
pu.print_green(
"STAR index {} already exists.".format(index_path))
self.index = index_path
return True
#check input files
if not (pu.check_files_exist(genome)):
pu.print_boldred(
"Please provide a valid input fasta file to build STAR index")
raise ValueError("Please check input to build star index")
#create index path if doesnt exist
if not pu.check_paths_exist(index_path):
if not pu.mkdir(index_path):
raise OSError(
"Error creating STAR index. Failed to create index directory."
)
return False
#determine parameters and execute cmd
#internal_args=()
internal_kwargs = {
"--runMode": "genomeGenerate",
"--genomeDir": index_path,
"--genomeFastaFiles": genome,
"--runThreadN": self._threads
}
#read build parameters
yamlfile = os.path.join(_params_dir, 'star_index.yaml')
if pu.check_files_exist(yamlfile):
yaml_params = pl.YAML_loader(yamlfile)
yaml_kwargs = yaml_params.get_kwargs()
internal_kwargs = {**yaml_kwargs, **internal_kwargs}
starbuild_Cmd = ['STAR']
starbuild_Cmd.extend(
pu.parse_unix_args(valid_args._args_STAR, internal_kwargs))
#execute command
status = pe.execute_command(starbuild_Cmd, objectid=objectid)
if status:
if pu.check_paths_exist(index_path) and not _dryrun:
#update object's index
self.index = index_path
if self.check_index():
return True
else:
raise OSError("Error building STAR index")
return True