def __init__(self, inputFnameLs=None, **keywords):
"""
"""
AbstractMatrixFileWalker.__init__(
self, inputFnameLs=inputFnameLs,
**keywords) #self.connectDB() called within its __init__()
#if user wants to preserve data in a data structure that is visisble throughout reading different files.
# then use this self.invariantPData.
#AbstractMatrixFileWalker has initialized a structure like below.
#self.invariantPData = PassingData()
#2012.11.22
self.xMin = None
self.xMax = None
self.yMin = None
self.yMax = None
#2012.11.25 for legend
self.plotObjectLs = []
self.plotObjectLegendLs = []
#2013.07.01 set these attributes if the class or its descendants have valid values
if getattr(self, 'defaultFontLabelSize', None):
yh_matplotlib.setFontAndLabelSize(self.defaultFontLabelSize)
if getattr(self, "defaultFigureWidth", None) and getattr(
self, "defaultFigureHeight", None):
yh_matplotlib.setDefaultFigureSize(
(self.defaultFigureWidth, self.defaultFigureHeight))
if getattr(self, "plotLeft", None) and getattr(self, "plotBottom",
None):
yh_matplotlib.setPlotDimension(left=self.plotLeft,
right=self.plotRight,
bottom=self.plotBottom,
top=self.plotTop)
def __init__(self, inputFnameLs=None, **keywords):
"""
"""
AbstractVervetMapper.__init__(self, inputFnameLs=inputFnameLs, **keywords)
if self.monkey1ID and self.monkey2ID:
self.monkeyPairDesignated = [self.monkey1ID, self.monkey2ID]
self.monkeyPairDesignated.sort()
self.monkeyPairDesignated = tuple(self.monkeyPairDesignated)
else:
self.monkeyPairDesignated = None
yh_matplotlib.setFontAndLabelSize(self.defaultFontLabelSize)
yh_matplotlib.setDefaultFigureSize((self.defaultFigureWidth, self.defaultFigureHeight))
def run(self,):
if self.debug: # 2010-4-18 enter debug mode "~/.../variation/misc.py -b"
import pdb
pdb.set_trace()
debug = True
else:
debug =False
sampleId2index, samplePair2data = self.readInput(self.inputFnameLs)
sys.stderr.write("Calculating distance matrix for aggregated data ...")
distanceMatrix = numpy.zeros([len(sampleId2index), len(sampleId2index)])
for samplePair, data in samplePair2data.iteritems():
no_of_mismatches, no_of_total_non_NA = data[:2]
distance = no_of_mismatches/no_of_total_non_NA
sample1Id, sample2Id = samplePair[:2]
sample1Index = sampleId2index[sample1Id]
sample2Index = sampleId2index[sample2Id]
distanceMatrix[sample1Index, sample2Index] = distance
distanceMatrix[sample2Index, sample1Index] = distance
sys.stderr.write("Done.\n")
sampleIdLs = sampleId2index.keys()
for sampleId, list_index in sampleId2index.iteritems():
sampleIdLs[list_index] = sampleId
if self.outputFname:
self.outputMismatchData(self.outputFname, samplePair2data, distanceMatrix, sampleId2index, sampleIdLs)
massagedSampleIDLs = self.massageSampleId(sampleIdLs)
#2012.9-6 stop massaging sample IDs for PCA output. mapper/AppendInfo2SmartPCAOutput.py could be applied to this.
self.runPCAOnDistanceMatrix(distanceMatrix, col_id_ls=sampleIdLs, outputFname='%s_PCA.tsv'%(self.figureFnamePrefix))
import pylab
from hcluster import pdist, linkage, dendrogram
pylab.clf()
Z=linkage(distanceMatrix, 'single')
yh_matplotlib.setFontAndLabelSize(base_size=3)
dendrogram(Z, color_threshold=0.001, labels=massagedSampleIDLs, orientation='right', leaf_font_size=None) #leaf_font_size=1 or 5 has no effect
pylab.savefig('%s.svg'%self.figureFnamePrefix, dpi=200)
pylab.savefig('%s.png'%self.figureFnamePrefix, dpi=300)
sys.exit(0)
