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")
import sys
import os
import glob
import sqlite3
from CGATCore import Pipeline as P
import CGATPipelines.PipelineTracks as PipelineTracks
# load options from the config file
P.getParameters([
"%s/pipeline.ini" % __file__[:-len(".py")], "../pipeline.ini",
"pipeline.ini"
])
PARAMS = P.PARAMS
PARAMS_ANNOTATIONS = P.peekParameters(PARAMS["annotations_dir"],
"pipeline_annotations.py")
Sample = PipelineTracks.Sample
TRACKS = PipelineTracks.Tracks(Sample).loadFromDirectory(
glob.glob("medip_*"), "medip_(\S+)")
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.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_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"])
cc = dbh.cursor()
"pipeline.ini"
])
PARAMS["projectsrc"] = os.path.dirname(__file__)
#for key, value in PARAMS.iteritems():
# print "%s:\t%s" % (key,value)
# 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))
# if necessary, update the PARAMS dictionary in any modules file.
# e.g.:
#
# import CGATPipelines.PipelineGeneset as PipelineGeneset
# PipelineGeneset.PARAMS = PARAMS
#
# Note that this is a hack and deprecated, better pass all
# parameters that are needed by a function explicitely.
# -----------------------------------------------
# Utility functions
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"))
from CGATCore import 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)