comparison conifer_functions.py @ 28:629681116b5f draft

Uploaded

27:02091c92f4f1

class sample(IsDescription):
	sampleID = 	StringCol(100,pos=0) # 20-char string (sampleID)

def loadProbeList(CF_probe_filename):
	# Load data files
	probefile = open(CF_probe_filename)
	s = csv.Sniffer()
	header = s.has_header(probefile.read(1024))
	probefile.seek(0)
	dialect = s.sniff(probefile.read(1024))
	probefile.seek(0)

def plotGenomicCoords(plt, rpkm_data,fontsize=10,rotation=0):
	import operator
	import locale
	exon_set = rpkm_data.exons
	genomic_coords = np.array(list(map(operator.itemgetter("start"),exon_set)))
	
	ticks = list(range(0,len(exon_set),len(exon_set)/5))
	
	ticks[-1] -= 1 # the last tick is going to be off the chart, so we estimate it as the second to last genomic coord.
	labels = [locale.format("%d", genomic_coords[i], grouping=True) for i in ticks if i < len(genomic_coords)]
	if rotation != 0:
		ha = "right"

def mergeCalls(calls):
	if len(calls) == 0:
		return []
	
	out_calls = []
	calls=np.array(calls)[np.argsort(np.array(list(map(operator.itemgetter("start"),calls)),dtype=np.int))]
	pstart = calls[0]["start"]
	pstop = calls[0]["stop"]
	for d in calls:
		if d["start"] <= pstop:
			pstop = max(d["stop"],pstop)

		self.calls = []
		self.refined_calls = []
		
	def smooth(self, window = 15, padded = False): #todo, fix the padding here
		if self.isGenotype:
			print("Warning: the data in this rpkm_data container are single genotype values. Smoothing will have no effect!")
			return self.rpkm
		
		if window > 0:
			weightings=np.blackman(window)
			weightings = weightings/weightings.sum()

class rpkm_reader:
	def __init__(self, rpkm_fn=None):
		"""Initialize an rpkm_reader instance. Specify the location of the data file"""
		
		if rpkm_fn == None:
			print("Must specify RPKM HDF5 file!")
			return 0
		# set up file access
		self.h5file = openFile(rpkm_fn, mode='r')
		self.sample_table = self.h5file.root.samples.samples
		
	def __del__(self):
		self.h5file.close()
	

28:629681116b5f

class sample(IsDescription):
	sampleID = 	StringCol(100,pos=0) # 20-char string (sampleID)

def loadProbeList(CF_probe_filename):
	# Load data files
	probefile = open(CF_probe_filename, 'rb')
	s = csv.Sniffer()
	header = s.has_header(probefile.read(1024))
	probefile.seek(0)
	dialect = s.sniff(probefile.read(1024))
	probefile.seek(0)

def plotGenomicCoords(plt, rpkm_data,fontsize=10,rotation=0):
	import operator
	import locale
	exon_set = rpkm_data.exons
	genomic_coords = np.array(map(operator.itemgetter("start"),exon_set))
	
	ticks = range(0,len(exon_set),len(exon_set)/5)
	
	ticks[-1] -= 1 # the last tick is going to be off the chart, so we estimate it as the second to last genomic coord.
	labels = [locale.format("%d", genomic_coords[i], grouping=True) for i in ticks if i < len(genomic_coords)]
	if rotation != 0:
		ha = "right"

def mergeCalls(calls):
	if len(calls) == 0:
		return []
	
	out_calls = []
	calls=np.array(calls)[np.argsort(np.array(map(operator.itemgetter("start"),calls),dtype=np.int))]
	pstart = calls[0]["start"]
	pstop = calls[0]["stop"]
	for d in calls:
		if d["start"] <= pstop:
			pstop = max(d["stop"],pstop)

		self.calls = []
		self.refined_calls = []
		
	def smooth(self, window = 15, padded = False): #todo, fix the padding here
		if self.isGenotype:
			print "Warning: the data in this rpkm_data container are single genotype values. Smoothing will have no effect!"
			return self.rpkm
		
		if window > 0:
			weightings=np.blackman(window)
			weightings = weightings/weightings.sum()

class rpkm_reader:
	def __init__(self, rpkm_fn=None):
		"""Initialize an rpkm_reader instance. Specify the location of the data file"""
		
		if rpkm_fn == None:
			print "Must specify RPKM HDF5 file!"
			return 0
		# set up file access
		# self.h5file = openFile(rpkm_fn, mode='r')
		self.h5file = open_file(rpkm_fn, mode='r')
		self.sample_table = self.h5file.root.samples.samples
		
	def __del__(self):
		self.h5file.close()