def __call__(self, track, slice=None):
exp_statement = """
SELECT TPM, gene_id, sample_name
FROM sailfish_genes AS A
JOIN samples AS B
ON A.sample_id = B.id"""
exp_df = self.getDataFrame(exp_statement)
factors_statement = '''
SELECT factor, factor_value, sample_name
FROM samples AS A
JOIN factors AS B
ON A.id = B.sample_id
WHERE factor != 'genome'
'''
factors_df = self.getDataFrame(factors_statement)
merged_df = pd.merge(exp_df,
factors_df,
left_on="sample_name",
right_on="sample_name")
genes = Pipeline.asList(
Pipeline.peekParameters(
".", "pipeline_rnaseqqc.py")['genes_of_interest'])
interest_df = merged_df[merged_df['gene_id'].isin(genes)]
interest_df['TPM'] = interest_df['TPM'].astype(float)
return interest_df.reset_index().set_index("factor")
# load options from the config file
PARAMS = P.getParameters(
["%s/pipeline.ini" % os.path.splitext(__file__)[0],
"../pipeline.ini",
"pipeline.ini"])
# add configuration values from associated pipelines
#
# 1. pipeline_annotations: any parameters will be added with the
# prefix "annotations_". The interface will be updated with
# "annotations_dir" to point to the absolute path names.
PARAMS.update(P.peekParameters(
PARAMS["annotations_dir"],
"pipeline_annotations.py",
on_error_raise=__name__ == "__main__",
prefix="annotations_",
update_interface=True))
GENESETS = PipelineTracks.Tracks(PipelineTracks.Sample).loadFromDirectory(
glob.glob("*.gtf.gz"),
"(\S+).gtf.gz")
TRACKS = PipelineTracks.Tracks(PipelineTracks.Sample3)
TRACKS = TRACKS.loadFromDirectory(glob.glob("*.bam"),
"(\S+).bam")
REPLICATE = PipelineTracks.Aggregate(TRACKS, labels=("replicate"))
CONDITION = PipelineTracks.Aggregate(TRACKS, labels=("condition", "tissue"))
# if necessary, update the PARAMS dictionary in any modules file.
import pysam
# load options from the config file
PARAMS = P.getParameters(
["%s/pipeline.ini" % os.path.splitext(__file__)[0],
"../pipeline.ini",
"pipeline.ini"])
# add configuration values from associated pipelines
#
# 1. pipeline_annotations: any parameters will be added with the
# prefix "annotations_". The interface will be updated with
# "annotations_dir" to point to the absolute path names.
PARAMS.update(P.peekParameters(
PARAMS["annotations_dir"],
"pipeline_annotations.py",
on_error_raise=__name__ == "__main__",
prefix="annotations_",
update_interface=True))
# define some tracks if needed
TRACKS = PipelineTracks.Tracks(PipelineTracks.Sample).loadFromDirectory(
glob.glob("*.ini"), "(\S+).ini")
# --------------------------< utility functions >---------------------------- #
def connect():
'''Connect to database.
Use this method to connect to additional databases.
Returns an sqlite3 database handle.
import CGAT.Stats as Stats
import pysam
import CGATPipelines.PipelineTracks as PipelineTracks
###################################################################
###################################################################
###################################################################
# Pipeline configuration
import CGATPipelines.Pipeline as P
P.getParameters(["%s/pipeline.ini" %
os.path.splitext(__file__)[0], "../pipeline.ini",
"pipeline.ini"])
PARAMS = P.PARAMS
PARAMS_ANNOTATIONS = P.peekParameters(
PARAMS["annotations_dir"],
"pipeline_annotations.py")
SEPARATOR = "|"
###################################################################
###################################################################
###################################################################
# Helper functions mapping tracks to conditions, etc
class TracksVCF (PipelineTracks.Tracks):
def load(self, filename, exclude=None):
'''load tracks from a vcf file.'''
tracks = []
import CGAT.Stats as Stats
import pysam
import CGATPipelines.PipelineTracks as PipelineTracks
###################################################################
###################################################################
###################################################################
# Pipeline configuration
import CGATPipelines.Pipeline as P
P.getParameters(["%s/pipeline.ini" %
os.path.splitext(__file__)[0], "../pipeline.ini",
"pipeline.ini"])
PARAMS = P.PARAMS
PARAMS_ANNOTATIONS = P.peekParameters(
PARAMS["annotations_dir"],
"pipeline_annotations.py")
SEPARATOR = "|"
###################################################################
###################################################################
###################################################################
# Helper functions mapping tracks to conditions, etc
class TracksVCF (PipelineTracks.Tracks):
def load(self, filename, exclude=None):
'''load tracks from a vcf file.'''
tracks = []
P.getParameters(
["%s/pipeline.ini" % os.path.splitext(__file__)[0],
"../pipeline.ini",
"pipeline.ini"],
defaults={"annotations_dir": "",
"genesets_abinitio_coding": "pruned.gtf.gz",
"genesets_abinitio_lncrna": "pruned.gtf.gz",
"genesets_reference": "reference.gtf.gz",
"genesets_refcoding": "refcoding.gtf.gz",
"genesets_previous": ""})
PARAMS = P.PARAMS
PARAMS.update(P.peekParameters(
PARAMS["annotations_dir"],
"pipeline_annotations.py",
prefix="annotations_",
update_interface=True))
PREVIOUS = P.asList(PARAMS["genesets_previous"])
def connect():
'''connect to database.
This method also attaches to helper databases.
'''
dbh = sqlite3.connect(PARAMS["database_name"])
statement = '''ATTACH DATABASE '%s' as annotations''' % (
PARAMS["annotations_database"])
###################################################
###################################################
###################################################
# Pipeline configuration
###################################################
P.getParameters([
"%s/pipeline.ini" % os.path.splitext(__file__)[0], "../pipeline.ini",
"pipeline.ini"
],
defaults={'paired_end': False})
PARAMS = P.PARAMS
PARAMS.update(
P.peekParameters(PARAMS["annotations_dir"],
"pipeline_annotations.py",
prefix="annotations_",
update_interface=True))
PipelinePeakcalling.PARAMS = PARAMS
PipelineMotifs.PARAMS = PARAMS
###################################################################
###################################################################
###################################################################
# Helper functions mapping tracks to conditions, etc
###################################################################
# load all tracks - exclude input/control tracks
# determine the location of the input files (reads).
DATADIR = PARAMS.get('input', '.')
if not os.path.exists(DATADIR):
raise OSError('data directory %s does not exists')
import CGATPipelines.Pipeline as Pipeline
import CGATPipelines.PipelineTracks as PipelineTracks
###################################################################
###################################################################
# parameterization
EXPORTDIR = P.get('calling_exportdir', P.get('exportdir', 'export'))
DATADIR = P.get('calling_datadir', P.get('datadir', '.'))
DATABASE = P.get('calling_backend',
P.get('report_sql_backend', 'sqlite:///./csvdb'))
###################################################################
# cf. pipeline_chipseq.py
# This should be automatically gleaned from pipeline_chipseq.py
###################################################################
PARAMS_PIPELINE = Pipeline.peekParameters(".", "pipeline_chipseq.py")
Sample = PipelineTracks.Sample3
suffixes = ["export.txt.gz", "sra", "fastq.gz", "fastq.1.gz", "csfasta.gz"]
TRACKS = sum(
itertools.chain([
PipelineTracks.Tracks(Sample).loadFromDirectory([
x for x in glob.glob("%s/*.%s" % (DATADIR, s)) if "input" not in x
], "%s/(\S+).%s" % (DATADIR, s)) for s in suffixes
]), PipelineTracks.Tracks(Sample))
Sample.setDefault("asTable")
ALL = PipelineTracks.Aggregate(TRACKS)
from CGATReport.Utils import PARAMS as P
import CGATPipelines.Pipeline as Pipeline
import CGATPipelines.PipelineTracks as PipelineTracks
###################################################################
###################################################################
# parameterization
EXPORTDIR = P.get('calling_exportdir', P.get('exportdir', 'export'))
DATADIR = P.get('calling_datadir', P.get('datadir', '.'))
DATABASE = P.get('calling_backend', P.get('report_sql_backend', 'sqlite:///./csvdb'))
###################################################################
# cf. pipeline_chipseq.py
# This should be automatically gleaned from pipeline_chipseq.py
###################################################################
PARAMS_PIPELINE = Pipeline.peekParameters(".",
"pipeline_chipseq.py")
Sample = PipelineTracks.Sample3
suffixes = ["export.txt.gz",
"sra",
"fastq.gz",
"fastq.1.gz",
"csfasta.gz"]
TRACKS = sum(itertools.chain([PipelineTracks.Tracks(Sample).loadFromDirectory(
[x for x in glob.glob("%s/*.%s" % (DATADIR, s)) if "input" not in x],
"%s/(\S+).%s" % (DATADIR, s)) for s in suffixes]),
PipelineTracks.Tracks(Sample))
import CGAT.Database as Database
import CGAT.GTF as GTF
import CGATPipelines.PipelineTracks as PipelineTracks
# load options from the config file
P.getParameters([
"%s/pipeline.ini" % os.path.splitext(__file__)[0], "../pipeline.ini",
"pipeline.ini"
])
PARAMS = P.PARAMS
USECLUSTER = True
# link up with annotations
PARAMS_ANNOTATIONS = P.peekParameters(PARAMS["annotations_dir"],
"pipeline_annotations.py")
# link up with ancestral repeats
PARAMS_ANCESTRAL_REPEATS = P.peekParameters(PARAMS["ancestral_repeats_dir"],
"pipeline_ancestral_repeats.py")
###################################################################
###################################################################
# Helper functions mapping tracks to conditions, etc
###################################################################
# collect sra nd fastq.gz tracks
TRACKS = PipelineTracks.Tracks(PipelineTracks.Sample).loadFromDirectory(
glob.glob("*.gtf.gz"),
"(\S+).gtf.gz",
exclude=("repeats.gtf.gz", "introns.gtf.gz", "merged.gtf.gz"))
import CGAT.GTF as GTF
import CGATPipelines.PipelineTracks as PipelineTracks
# load options from the config file
P.getParameters(
["%s/pipeline.ini" % os.path.splitext(__file__)[0],
"../pipeline.ini",
"pipeline.ini"])
PARAMS = P.PARAMS
USECLUSTER = True
# link up with annotations
PARAMS_ANNOTATIONS = P.peekParameters(
PARAMS["annotations_dir"],
"pipeline_annotations.py")
# link up with ancestral repeats
PARAMS_ANCESTRAL_REPEATS = P.peekParameters(
PARAMS["ancestral_repeats_dir"],
"pipeline_ancestral_repeats.py")
###################################################################
###################################################################
# Helper functions mapping tracks to conditions, etc
###################################################################
# collect sra nd fastq.gz tracks
TRACKS = PipelineTracks.Tracks(PipelineTracks.Sample).loadFromDirectory(
glob.glob("*.gtf.gz"), "(\S+).gtf.gz", exclude=("repeats.gtf.gz", "introns.gtf.gz", "merged.gtf.gz"))
###################################################################
###################################################################
# Load options and annotations
###################################################################
# load options from the config file
PARAMS = P.getParameters(
["%s/pipeline.ini" % os.path.splitext(__file__)[0],
"../pipeline.ini",
"pipeline.ini"])
# add configuration values from associated pipelines
PARAMS = P.PARAMS
PARAMS.update(P.peekParameters(
PARAMS["annotations_dir"],
"pipeline_genesets.py",
prefix="annotations_",
update_interface=True,
restrict_interface=True)) # add config values from associated pipelines
# The DEXSeq R directory contains important python helper functions
PYTHONSCRIPTSDIR = R('''
f = function(){
pythonScriptsDir = system.file("python_scripts", package="DEXSeq")
}
f()''').tostring()
###################################################################
###################################################################
###################################################################
# Utility functions
# load options from the config file
P.getParameters([
"%s/pipeline.ini" % os.path.splitext(__file__)[0], "../pipeline.ini",
"pipeline.ini"
])
PARAMS = P.PARAMS
# Add parameters from the gtf_subset pipeline, but
# only the interface section. All PARAMS options
# will have the prefix `annotations_`
PARAMS.update(
P.peekParameters(PARAMS["gtf_dir"],
"pipeline_genesets.py",
prefix="annotations_",
update_interface=True,
restrict_interface=True))
# -----------------------------------------------
# Utility functions
def connect():
'''utility function to connect to database.
Use this method to connect to the pipeline database.
Additional databases can be attached here as well.
Returns an sqlite3 database handle.
'''
# load options from the config file
P.getParameters(
["%s/pipeline.ini" % os.path.splitext(__file__)[0],
"../pipeline.ini",
"pipeline.ini"])
PARAMS = P.PARAMS
# Add parameters from the gtf_subset pipeline, but
# only the interface section. All PARAMS options
# will have the prefix `annotations_`
PARAMS.update(P.peekParameters(
PARAMS["gtf_dir"],
"pipeline_genesets.py",
prefix="annotations_",
update_interface=True,
restrict_interface=True))
# -----------------------------------------------
# Utility functions
def connect():
'''utility function to connect to database.
Use this method to connect to the pipeline database.
Additional databases can be attached here as well.
Returns an sqlite3 database handle.
'''
