Python ConiferPipeline示例

示例#1

文件： QC_01_sample_sd.py 项目： oklasoft/conifer-tools

from conifertools import ConiferPipeline
import argparse
import numpy as np

if __name__ == "__main__":

parser = argparse.ArgumentParser()
parser.add_argument("--infile", "-i", action="store", required=True)
parser.add_argument("--outfile", "-o", action="store", required=True)
parser.add_argument("--plotfile", action="store", required=False, default=None)
parser.add_argument("--verbose", action="store_true", required=False, default=False)
args = parser.parse_args(["--infile","/net/eichler/vol8/home/nkrumm/EXOMES/ESP2000/SVD/ESP.BROAD.SVD70.QC.SampleQC.hdf5", "--outfile","test.SVD70sd.txt", "--plotfile","test.SVD70.png", "--verbose"])

p = ConiferPipeline(args.infile)
#n_probes = sum([len(p.getProbesByChrom(x)) for x in range(1,24)])
out_sd_values = []
with open(args.outfile, 'w') as out_file:
    for sampleID in p.samples:
        sample_data = []
        for c in xrange(1,24):
            sample_data.extend(p.getConiferData(sampleID, c).rpkm)
        sample_sd = np.std(np.array(sample_data))
        out_sd_values.append(sample_sd)
        if args.verbose:
            print sampleID, sample_sd
        out_file.write("%s\t%f\n" % (sampleID, sample_sd))

    if args.plotfile:
        from conifertools.plotting import QC_SampleSD_Plot
        kwargs = {"bins": 20, "color":'r'}
        QC_SampleSD_Plot(out_sd_values, title='SD values', outfile=args.plotfile, logscale=True, **kwargs)

示例#2

文件： QC_01_sample_sd.py 项目： tianyunwang/conifer-tools

from conifertools import ConiferPipeline
import argparse
import numpy as np

if __name__ == "__main__":

parser = argparse.ArgumentParser()
parser.add_argument("--infile", "-i", action="store", required=True)
parser.add_argument("--outfile", "-o", action="store", required=True)
parser.add_argument("--plotfile", action="store", required=False, default=None)
parser.add_argument("--verbose", action="store_true", required=False, default=False)

p = ConiferPipeline(args.infile)

out_sd_values = []
with open(args.outfile, 'w') as out_file:
    for sampleID in p.samples:
        sample_data = []
        for c in xrange(1,24):
            sample_data.extend(p.getConiferData(sampleID, c).rpkm)
        sample_sd = np.std(np.array(sample_data))
        out_sd_values.append(sample_sd)
        if args.verbose:
            print sampleID, sample_sd
        out_file.write("%s\t%f\n" % (sampleID, sample_sd))

    if args.plotfile:
        from conifertools.plotting import QC_SampleSD_Plot
        kwargs = {"bins": 20, "color":'r'}
        QC_SampleSD_Plot(out_sd_values, title='SD values', outfile=args.plotfile, logscale=True, **kwargs)

示例#3

文件： 02_filter_calls.py 项目： tianyunwang/conifer-tools

from conifertools import ConiferPipeline, CallTable, CallFilterTemplate
import argparse
import numpy as np

if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("--conifer_file", action="store", required=True)
    parser.add_argument("--call_file", action="store", required=True)
    parser.add_argument("--outfile", "-o", action="store", required=True)
    args = parser.parse_args()
    p = ConiferPipeline(args.conifer_file)

    SDFilter = CallFilterTemplate(
        p,
        "/net/eichler/vol8/home/nkrumm/REFERENCE_INFO/hg19genomicSuperDups.bed",
        name="SegDupProbeOverlap",
        filter_type="overlap",
        func=lambda x: x < 0.5)

    SDCount = CallFilterTemplate(
        p,
        "/net/eichler/vol8/home/nkrumm/REFERENCE_INFO/hg19genomicSuperDups.bed",
        name="SegDupProbeCount",
        filter_type="count")

    PPGFilter = CallFilterTemplate(
        p,
        "/net/eichler/vol8/home/nkrumm/REFERENCE_INFO/pp_genes.hg19.spans.bed",
        name="PPG_overlap",
        filter_type="overlap",
        func=lambda x: x < 0.9)

示例#4

文件： 01_create_calls.py 项目： oklasoft/conifer-tools

import argparse


if __name__ == "__main__":

    parser = argparse.ArgumentParser()
    parser.add_argument("--infile", "-i", action="store", required=True)
    parser.add_argument("--outfile", "-o", action="store", required=True)
    parser.add_argument("--chr", "-c", action="store", nargs="+", required=False, default="All")
    parser.add_argument("--samples", "-s", action="store", nargs="+", required=False, default="All")
    parser.add_argument("--ncpu", "--ncpus", action="store", type=int, required=False, default=10)
    parser.add_argument("--nretry", "--nretries", action="store", type=int, required=False, default=3)
    parser.add_argument("--verbose", "-v", action="store_true", required=False, default=False)
    args = parser.parse_args()

    p = ConiferPipeline(args.infile)
    if args.chr == "All":
        chromosomes = None  # process all
        print "All Chromosomes"
    else:
        chromosomes = args.chr
        print "Chromosomes: %s" % ", ".join(chromosomes)
    if args.samples == "All":
        samples = None  # process all
        print "All samples"
    else:
        samples = args.samples
        print "Samples: ", ", ".join(samples)

    calls = p.makeCallsMPI(chromosomes=chromosomes,
                           samples=samples,

示例#5

import argparse


if __name__ == "__main__":

    parser = argparse.ArgumentParser()
    parser.add_argument("--infile", "-i", action="store", required=True)
    parser.add_argument("--outfile", "-o", action="store", required=True)
    parser.add_argument("--chr", "-c", action="store", nargs="+", required=False, default="All")
    parser.add_argument("--samples", "-s", action="store", nargs="+", required=False, default="All")
    parser.add_argument("--ncpu", "--ncpus", action="store", type=int, required=False, default=10)
    parser.add_argument("--nretry", "--nretries", action="store", type=int, required=False, default=3)
    parser.add_argument("--verbose", "-v", action="store_true", required=False, default=False)
    args = parser.parse_args()

    p = ConiferPipeline(args.infile)
    if args.chr == "All":
        chromosomes = None  # process all
        print "All Chromosomes"
    else:
        chromosomes = args.chr
        print "Chromosomes: %s" % ", ".join(chromosomes)
    if args.samples == "All":
        samples = None  # process all
        print "All samples"
    else:
        samples = args.samples
        print "Samples: ", ", ".join(samples)

    calls = p.makeCallsMPI(chromosomes=chromosomes,
                           samples=samples,