plotExpBox.py

#!/usr/bin/env python

'''

newer version plotExpBox2.py

Copyright 2010 Wu Albert Cheng <albertwcheng@gmail.com>

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.

'''

from pylab import *
from sys import stderr,stdout,argv
from getopt import getopt
import sys
from albertcommon import *
from welchttest import welchs_approximate_ttest_arr
from scipy.stats.stats import ttest_ind,ttest_1samp,mannwhitneyu
from scipy.stats import wilcoxon #this is paired!!


def plotExpBox(data,xtickLabels,showIndPoints,mark,markMean,showMean,notch,whisker,outliers,xlegendrotation,xlabe,ylabe,titl,showSampleSizes):
	prevHoldState=ishold()	
	hold(True)
	
	if outliers:	
		fliers="b+"
	else:
		fliers=""

	whisValue=0.0

	if whisker:
		whisValue=1.5


	boxplot(data,notch,widths=0.5,sym=fliers,whis=whisValue)
	#print >> stderr,resultD
	
	maxMax=-10000000.0
	minMin=10000000.0
	
	for i in range(0,len(data)):
		curCol=data[i]
#		datasorted=data[:]
#		datasorted.sort()
#		numData=len(datasorted)
#		HQn=numData*3/4		
#		LQn=numData*1/4
#		maxMax=max(maxMax,datasorted[HQn]*1.5)
#		minMin=min(minMax,datasorted[LQn]*1.5)
			
		maxMax=max(maxMax,max(curCol))
		minMin=min(minMin,min(curCol))
		if showMean:
			plot([i+0.75,i+1.25],[mean(curCol)]*2,markMean)
		if showIndPoints:
			plot([i+1]*len(curCol),curCol,mark)
	
	if showSampleSizes:
		for i in range(0,len(data)):
			curCol=data[i]
			text(i+1,maxMax*1.05,str(len(curCol)),horizontalalignment='center',verticalalignment='center',color='red')

			
	xticks( range(1,len(data)+1), xtickLabels , rotation=xlegendrotation)
	xlabel(xlabe)
	ylabel(ylabe)
	title(titl)
	ylim([minMin-maxMax*0.1,maxMax*1.1])
	hold(prevHoldState)

def findIndices(needles,haystack):
	indices=[]
	for needle in needles:
		indices.append(haystack.index(needle))

	return indices

def rearrangeColAndRowSqMatrix(M,from_indices):
	newM=[]
	lenM=len(M)
	for r in range(0,lenM):
		newRow=[]
		newM.append(newRow)
		for c in range(0,lenM):
			newRow.append(M[from_indices[r]][from_indices[c]])

	return newM

def printMatrix(stream,M,prefixes):
	for row,prefix in zip(M,prefixes):
		for cell in row:
			print >> stream,"%g\t" % (cell),

		print >> stream,prefix
#use pvalue as a distance metric
#dist=1-pvalues
#fake the record for PyCluster
def makePValueClusterPlot(jobname,sampleNames,pvaluesM,methodCluster):


	#fake a record
	record=Record()
	#fake M
	M=[]
	Mr=[]
	M.append(Mr)
	for sample in sampleNames:
		Mr.append(0)

	record.data=M
	record.mask=None
	record.geneid=["Dummy"]
	record.genename=["Dummy"]
	record.expid=sampleNames
	record.uniqid="GENENAME"

	#now do something serious
	distM=[]
	for pvalueRows in pvaluesM:
		distRow=[]
		distM.append(distRow)
		for pvalue in pvalueRows:
			distRow.append(1.0-pvalue)

	
	#now cluster
	Tree=treecluster(distancematrix=distM,method=methodCluster)	

	record.save(jobname,expclusters=Tree)

	#now hijack the result file and change it to pvalue heatmap
	fil=open(jobname+".cdt")
	firstthreelines=[]
	lino=0
	for lin in fil:
		lino+=1
		if lino>3:
			break
		lin=lin.rstrip()
		firstthreelines.append(lin)
		if lino==1:
			fields=lin.split("\t")	
			arrayOrdered=fields[3:]
	

	fil.close()

	fil=open(jobname+".cdt","w")
	for lin in firstthreelines:
		print >> fil, lin

	rearrangedCorrMatrix=rearrangeColAndRowSqMatrix(pvaluesM,findIndices(arrayOrdered,sampleNames))

	for i in range(0,len(arrayOrdered)):
		print >> fil, arrayOrdered[i]+"\t"+arrayOrdered[i]+"\t"+"1.000000",
		for j in range(0,len(arrayOrdered)):
			print >> fil,"\t"+str(rearrangedCorrMatrix[i][j]),

		print >> fil,""

	fil.close()
	

	fil=open(jobname+".mat","w")

	print >> fil, "Correlation Matrix (Not Clustered)"
	print >> fil,"\t".join(sampleNames)
	printMatrix(fil, pvaluesM, sampleNames)

	print >> fil, "Correlation Matrix (Clustered)"
	print >> fil,"\t".join(arrayOrdered)
	printMatrix(fil, rearrangedCorrMatrix, arrayOrdered)
	fil.close()


def makePValueRawPlot(jobname,sampleNames,pvaluesM):


	#fake a record
	record=Record()
	#fake M
	M=[]
	Mr=[]
	M.append(Mr)
	for sample in sampleNames:
		Mr.append(0)

	record.data=M
	record.mask=None
	record.geneid=["Dummy"]
	record.genename=["Dummy"]
	record.expid=sampleNames
	record.uniqid="GENENAME"

	#now do something serious


	record.save(jobname)

	#now hijack the result file and change it to pvalue heatmap
	fil=open(jobname+".cdt")
	firstthreelines=[]
	lino=0
	for lin in fil:
		lino+=1
		if lino>2:
			break
		lin=lin.rstrip()
		firstthreelines.append(lin)
		if lino==1:
			fields=lin.split("\t")	
			arrayOrdered=fields[3:]
	

	fil.close()

	fil=open(jobname+".cdt","w")
	for lin in firstthreelines:
		print >> fil, lin

	rearrangedCorrMatrix=pvaluesM


	for i in range(0,len(arrayOrdered)):
		print >> fil, arrayOrdered[i]+"\t"+arrayOrdered[i]+"\t"+"1.000000",
		for j in range(0,len(arrayOrdered)):
			print >> fil,"\t"+str(rearrangedCorrMatrix[i][j]),

		print >> fil,""

	fil.close()
	

def plotExpBox_Main(inputFile,header,cols,outputFile,sep,startRow,showIndPoints,mark,markMean,showMean,notch,whisker,outliers,plotPvalueCluster,outputClusterPrefix,methodCluster,xlegendrotation,xlabe,ylabe,figsz,titl,showSampleSizes):

	#if plotPvalueCluster:
		#if pvalue cluster is needed:
	#	from Bio.Cluster.cluster import *
	#	from Bio.Cluster import *
		#endif


	#the real deal!


	fin=generic_istream(inputFile)
	
	plotData=[]	
	xtickLabels=[]
	for col in cols:
		plotData.append([])
		xtickLabels.append(header[col])

	colIndices=range(0,len(cols))

	lino=0
	for lin in fin:
		lino+=1
		if lino<startRow:
			continue		
		fields=lin.rstrip("\r\n").split(sep)
		
		for idx,col in zip(colIndices,cols):
			try:
				value=float(fields[col])			
				plotData[idx].append(value)
			except:
				pass		
	fin.close()

	print >> stdout,"student t-test (1 sample; mean=0)"
	print >> stdout,"sample","mean","p-val"


	for x in range(0,len(plotData)):
		print >> stdout, xtickLabels[x],mean(plotData[x]),ttest_1samp(plotData[x],0)[1]


	pvalueM=[]
	
	print >> stdout,""
	
	print >> stdout,"student t-test (2 samples)"
	print >> stdout,"p-val",
	for x in range(0,len(plotData)):
		print >> stdout,xtickLabels[x],
	
	print >> stdout,""

	for x in range(0,len(plotData)):
		pvalueRow=[]
		pvalueM.append(pvalueRow)
		print >> stdout, xtickLabels[x],
		for y in range(0,len(plotData)):
			if y<=x:
				print >> stdout, "",
				if x==y:
					pvalueRow.append(1.0)
				else:
					pvalueRow.append(pvalueM[y][x])
			else:
				pvalue=ttest_ind(plotData[x],plotData[y])[1]
				print >> stdout, str(pvalue),
				pvalueRow.append(pvalue)
		print >> stdout,"";	

	
	print >> stdout,""

	makePValueRawPlot(outputClusterPrefix+"_t_raw",xtickLabels,pvalueM)


	if plotPvalueCluster:
		makePValueClusterPlot(outputClusterPrefix+"_t",xtickLabels,pvalueM,methodCluster)

	pvalueM=[]

	print >> stdout,"welch t-test"
	print >> stdout,"p-val",
	for x in range(0,len(plotData)):
		print >> stdout,xtickLabels[x],
	
	print >> stdout,""
	for x in range(0,len(plotData)):
		pvalueRow=[]
		pvalueM.append(pvalueRow)
		print >> stdout, xtickLabels[x],
		for y in range(0,len(plotData)):
			if y<=x:
				print >> stdout, "",
				if x==y:
					pvalueRow.append(1.0)
				else:
					pvalueRow.append(pvalueM[y][x])
					
			else:
				pvalue=welchs_approximate_ttest_arr(plotData[x],plotData[y])[3]
				print >> stdout, str(pvalue),
				pvalueRow.append(pvalue)
		print >> stdout,"";

	makePValueRawPlot(outputClusterPrefix+"_Welch_raw",xtickLabels,pvalueM)

	if plotPvalueCluster:
		makePValueClusterPlot(outputClusterPrefix+"_Welch",xtickLabels,pvalueM,methodCluster)

	
	print >> stdout,""
	print >> stdout,"non-parametric (Mann-Whitney U)" #"non-parametric (Mann-Whitney U if larger n<=20 else Wilcoxon)"
	print >> stdout,"p-val",
	for x in range(0,len(plotData)):
		print >> stdout,xtickLabels[x],
	

	pvalueM=[]

	print >> stdout,""
	for x in range(0,len(plotData)):
		pvalueRow=[]
		pvalueM.append(pvalueRow)
		print >> stdout, xtickLabels[x],
		for y in range(0,len(plotData)):
			if y<=x:
				print >> stdout, "",
				if x==y:
					pvalueRow.append(1.0)
				else:
					pvalueRow.append(pvalueM[y][x])
			else:
				#if max(len(plotData[x]),len(plotData[y]))<=20:
				pvalue=mannwhitneyu(plotData[x],plotData[y])[1]*2				
				print >> stdout,pvalue, #mann-whiteney need to mul by 2 (one tail to two tail)
				pvalueRow.append(pvalue)
				#else:
				#	print >>  stdout,wilcoxon(plotData[x],plotData[y])[1], # this is two-tailed already stdout, "", #
		print >> stdout,"";	
	
	makePValueRawPlot(outputClusterPrefix+"_U_raw",xtickLabels,pvalueM)

	if plotPvalueCluster:
		makePValueClusterPlot(outputClusterPrefix+"_U",xtickLabels,pvalueM,methodCluster)
	
	figure(figsize=figsz)
		
	if len(titl)==0:
		titl=outputFile

	plotExpBox(plotData,xtickLabels,showIndPoints,mark,markMean,showMean,notch,whisker,outliers,xlegendrotation,xlabe,ylabe,titl,showSampleSizes)
	#ylim([0,200])

	savefig(outputFile,bbox_inches="tight")
	

def usageExit(programName):
	print >> stderr,programName,"inputFile valCol1 outputFile"
	print >> stderr,"Options:"
	print >> stderr,"-t -F -d --fs seperator"
	print >> stderr,"-r --headerRow headerRow"
	print >> stderr,"-s --startRow startRow"
	print >> stderr,"-p --showIndPoints"
	print >> stderr,"-m --showMean"
	print >> stderr,"-n --notch"
	print >> stderr,"--xtick-rotation degree"
	print >> stderr,"--offWhisker"
	print >> stderr,"--offOutliers"
	print >> stderr,"--pvalue-cluster-as prefix  make pvalue cluster heatmap using 1-pvalue as distance metric"
	print >> stderr,"--pvalue-cluster-method method   cluster using one of the following method for the pvalue cluster heatmap"
	print >> stderr,"--xlabel label"
	print >> stderr,"--ylabel label"
	print >> stderr,"--figsize w,h"
	print >> stderr,"--title title (default is filename)"
	print >> stderr,"--show-sample-sizes"
	print >> stderr, "from PyCluster (see http://www.biopython.org/DIST/docs/api/Bio.Cluster.Record-class.html#treecluster)"
	print >> stderr, "method   : specifies which linkage method is used:"
	print >> stderr, "           method=='s': Single pairwise linkage"
	print >> stderr, "           method=='m': Complete (maximum) pairwise linkage (default)"
	print >> stderr, "           method=='c': Centroid linkage"
	print >> stderr, "           method=='a': Average pairwise linkage"
	explainColumns(stderr)

	sys.exit()

if __name__=='__main__':
	programName=argv[0]
	optlist,args=getopt(argv[1:],'t:F:d:r:s:pmn',['fs=','headerRow=','startRow=','showIndPoints','showMean','notch','offWhisker','offOutliers','pvalue-cluster-as=','pvalue-cluster-method=','xtick-rotation=','xlabel=','ylabel=','figsize=','title=','show-sample-sizes'])

	headerRow=1
	startRow=2
	fs="\t"
	showIndPoints=False
	showMean=False
	whisker=True
	outliers=True
	notch=0
	xlegendrotation=0
	makePvalueClusters=False
	pvalueClusterOutputPrefix=""
	pvalueClusterMethod="a"
	xlabe="Samples"
	ylabe="Values"
	titl=""
	figsz=(8,6)
	showSampleSizes=False
	if len(args)!=3:
		usageExit(programName)
	else:
		fileName,colString,outputFile=args
	
		for a,v in optlist:
			if a in ["-F","-t","-d","--fs"]:
				fs=replaceSpecialChar(v)
			elif a in ["-s","--startRow"]:
				startRow=int(v)
			elif a in ["-r","--headerRow"]:
				headerRow=int(v)
			elif a in ["-p","--showIndPoints"]:
				showIndPoints=True
			elif a in ["-m","--showMean"]:
				showMean=True
			elif a in ["-n","--notch"]:
				notch=1
			elif a in ["--offOutliers"]:
				outliers=False
			elif a in ["--offWhisker"]:
				whisker=False
			elif a in ["--pvalue-cluster-as"]:
				makePvalueClusters=True
				pvalueClusterOutputPrefix=v
			elif a in ["--pvalue-cluster-method"]:
				pvalueClusterMethod=v
			elif a in ["--xtick-rotation"]:
				xlegendrotation=int(v)
			elif a in ["--xlabel"]:
				xlabe=v
			elif a in ["--ylabel"]:
				ylabe=v
			elif a in ["--figsize"]:
				v=v.split(",")
				figsz=(int(v[0]),int(v[1]))
			elif a in ["--title"]:
				titl=v
			elif a in ["--show-sample-sizes"]:
				showSampleSizes=True

		header,prestarts=getHeader(fileName,headerRow,startRow,fs)
		
		cols=getCol0ListFromCol1ListStringAdv(header,colString)
	
	if makePvalueClusters:
		from Bio.Cluster.cluster import *
		from Bio.Cluster import *
	
	
	plotExpBox_Main(fileName,header,cols,outputFile,fs,startRow,showIndPoints,'b,','g--',showMean,notch,whisker,outliers,makePvalueClusters,pvalueClusterOutputPrefix,pvalueClusterMethod,xlegendrotation,xlabe,ylabe,figsz,titl,showSampleSizes)	
		
		