view Contra/scripts/average_count.py @ 4:9a76d500b049

Uploaded

# ----------------------------------------------------------------------#
# Copyright (c) 2011, Richard Lupat & Jason Li.
#
# > Source License <
# This file is part of CONTRA.
#
#    CONTRA is free software: you can redistribute it and/or modify
#    it under the terms of the GNU General Public License as published by
#    the Free Software Foundation, either version 3 of the License, or
#    (at your option) any later version.
#
#    CONTRA is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#    GNU General Public License for more details.
#
#    You should have received a copy of the GNU General Public License
#    along with CONTRA.  If not, see <http://www.gnu.org/licenses/>.
#
# 
#-----------------------------------------------------------------------#
# Last Updated : 28 Sep 2011 11:00AM

import sys
import math

def getAverage(list1):
	if len(list1) > 0:
		return float(sum(list1))/len(list1)
	
	return 0.0

def getStdDev(list1, avg):
	var = 0.0
	for x in list1:
		var += (avg - x) ** 2

	if (len(list1)-1) > 0:
		var /= (len(list1)-1)

	return math.sqrt(var)

def getMinMax(list1):
	length = len(list1)
	if length != 0:
       		min = list1[0]
        	max = list1[length-1]
	else:
        	min = 0
        	max = 0

	return min, max

def getMedian(list1):
	length = len(list1)
	if length == 0:
        	median = 0
	elif length % 2 == 0:
        	median = (list1[length/2]+list1[(length/2) - 1])/2
	else:
        	median = list1[length/2]
	return median

def createDataDict(count, list1, r, offset, id_check, exon_check):
	tDict 	 = {}
	tDictOri = {}

	while count < len(list1):
		t 	= list1[count].split()
		tId     = t[5]
		tExon   = t[6]

		if (tId != id_check) or (tExon != exon_check):
       	 		return count, tDict, tDictOri

        	tStart  = int(t[2])
       		tEnd    = int(t[3])
	        tCov    = float(t[4]) / r + offset       #GeoMean Normalisation
        	tCovOri = float(t[4]) + offset     #without scaling

        	#filling dict
        	while tStart < tEnd:
                	tDict[tStart] = tCov
                	tDictOri[tStart] = tCovOri #without scaling
                	tStart += 1

        	count += 1

	return count, tDict, tDictOri

def getFactor (val1, val2):
	r = math.sqrt(val1 * val2)
	r1 = val1/r
	r2 = val2/r
	return r1, r2

def averageCount(tFile, nFile, averageOut, tReadCount, nReadCount, rd_threshold, minNBases):
	tList = file.readlines(open(tFile))
	nList = file.readlines(open(nFile))
	# constant & counter
	OFF    = 1
	tCount = 0
	nCount = 0

	# create and open files
	output = open(averageOut, "w")
	
	# Offset and Ratio for Geometric Mean Normalisation
	r1, r2 = getFactor(tReadCount, nReadCount)	
	if rd_threshold > 0:
		#OFF = 0 
		OFF = 0.5

	#big loop
	while (nCount < len(nList)):
		# initialisation, get the chr, geneID, geneName
		init	= tList[tCount].split()
		initial = init[5] 
		_exon 	= init[6]
		chr 	= init[1]
		gene 	= init[0]
		_start 	= int(init[2])

		# check if t-gene and n-gene refer to the same gene
		check_init = nList[nCount].split()
		if check_init[5] != initial or check_init[6] != _exon:
			print "Initial: ", initial
			print "Check_Init.id: ", check_init[5]
			print "_Exon: ", _exon
			print "Check_Init.exon: ", check_init[6]
			print "Error. Comparing different Gene"
			sys.exit(1)
		
		# create data dictionary for tumour and normal data (per each regions/ exon)
		tCount, tDict, tDictOri = createDataDict(tCount, tList, r1, OFF, initial, _exon)
		nCount, nDict, nDictOri = createDataDict(nCount, nList, r2, OFF, initial, _exon)		
		# check number of bases in the both gene dict
		if len(nDict) != len(tDict):
			print "N:", len(nDict)
			print "T:", len(tDict)
			print "Error. Different length of dict"
			sys.exit(1)

		# compare coverage
		count = _start
		_max = max(nDict.keys())
		ratioList = []
		tumourList = []
		normalList = []
		tumourOriList = []
		normalOriList = []
		while count <= _max:
			# get ratio
			if (nDict[count] < rd_threshold) and (tDict[count] < rd_threshold):
				ratio = 0.0
			else:
				if tDict[count] == 0:
					tDict[count] = 0.5

				ratio = math.log((float(tDict[count]) / nDict[count]),2)
				tumourList.append(tDict[count])
				tumourOriList.append(tDictOri[count])
				normalList.append(nDict[count])
				normalOriList.append(nDictOri[count])
				ratioList.append(ratio)
		
			count += 1

		ratioLen = len(ratioList)

		# get average
		avg = getAverage(ratioList)
		sd = getStdDev(ratioList, avg)
		tumourAvg= str(round(getAverage(tumourList),3))
		normalAvg= str(round(getAverage(normalList),3))
		tumourOriAvg = str(round(getAverage(tumourOriList),3))
		normalOriAvg = str(round(getAverage(normalOriList),3))

		# get median
		ratioList.sort()
		min_logratio, max_logratio = getMinMax(ratioList)
		median	= getMedian(ratioList)

		# write output
		if ratioLen >= minNBases:
			output.write(initial + "\t" + gene + "\t" + str(ratioLen) + "\t")
			output.write(str(round(avg,3))+ "\t"+ str(count)+ "\t" + _exon + "\t")
			output.write(str(round(sd ,3))+ "\t"+ tumourAvg + "\t" + normalAvg +"\t") 
			output.write(tumourOriAvg + "\t" + normalOriAvg + "\t")
			output.write(str(round(median,3)) + "\t" + str(round(min_logratio,3)) + "\t")
			output.write(str(round(max_logratio,3)) + "\n")

	output.close()

	#print "End of averageCount.py with the last target = '%s'" %(initial)