def process(dirIn, es, ee):
roi_w = 30
roi_h = 30
iStart = 6-1
#srcDir = DirectoryChooser("Choose directory").getDirectory()
srcDir=dirIn
if not srcDir:
return
destDir = srcDir+"--analysis"
print "creating: "+destDir
# destDir = srcDir
if os.path.exists(destDir):
shutil.rmtree(destDir)
time.sleep(1)
os.mkdir(destDir)
# fileId = IJ.getString("filenames MUST contain:", "FRAP.lsm");
fileId = "FRAP.lsm"
iFile = 0
sOut = []
print "starting analysis in folder: "+srcDir
for root, directories, filenames in os.walk(srcDir):
print "sadsdfs"
print directories
for filename in filenames:
print filename
if not (fileId in filename):
continue
iFile = iFile + 1
if iFile < es:
continue
if iFile > ee:
f.close()
return()
# extract information from file and foldernames
print "root = %s" % root
(tmp,folder2) = splitLastFolder(root)
tmp = folder2.split("--")[4]
replicate = folder2
print replicate
wells = tmp.split("-")
print wells
tmp = filename.split("--")[1]
tmp = tmp.split("W")[1]
wellNum = int(tmp)
print wellNum
well = wells[wellNum-1]
print well
tmp = filename.split("--")[0:4]
tmp = '--'.join(tmp)
filenameDocu = tmp+".lsm--docu.png"
filenameLSM = tmp+".lsm"
pathFRAP = os.path.join(root,filename)
pathDocu = os.path.join(root,filenameDocu)
pathLSM = os.path.join(root,filenameLSM)
pathFRAPandSEG = os.path.join(destDir,filename+"--raw+seg.tif")
print pathFRAP
print pathFRAPandSEG
print pathDocu
print pathLSM
f = open(os.path.join(destDir, filename+"--metadata.csv"), 'w')
f.write("well,replicate,pathFRAP,pathFRAPandSEG,pathDocu,pathLSM\n")
f.write(well+","+replicate+","+pathFRAP+","+pathFRAPandSEG+","+pathDocu+","+pathLSM+"\n")
f.close()
#IJ.run("Close all forced", "")
# load data (possible headless)
path = os.path.join(root, filename)
#imp = IJ.openImage(path)
impA = getImps(path)
imp = impA[0]
imp.show()
#IJ.run("Bio-Formats Importer","open="+path+" color_mode=Default split_channels view=Hyperstack stack_order=XYCZT")
# remove transmisson channel
IJ.run("Slice Remover", "first=2 last=100000 increment=2");
imp = IJ.getImage()
#imp.show()
imp.setTitle("raw")
IJ.run("Properties...", "unit=pixels pixel_width=1 pixel_height=1");
dt = imp.getFileInfo().frameInterval
print "dt = %f" % dt
if dt==0:
dt=0.28
print "dt = %f" % dt
# todo: can i get the real time-stamps?
im_w = imp.getWidth()
im_h = imp.getHeight()
print "im_w ="+str(im_w)
print "im_h ="+str(im_h)
nt = max(imp.getNSlices(),imp.getNFrames())
print "nt = %f" % nt
roi_x = im_w/2 - roi_w/2
roi_y = im_h/2 - roi_h/2
roi_x2 = roi_x + roi_w
roi_y2 = roi_y + roi_h
# preprocessing
IJ.run("Duplicate...","title=gb duplicate stack")
# smooth
IJ.run("Gaussian Blur...", "sigma=2 stack");
# tophat
IJ.run("3D Fast Filters","filter=TopHat radius_x_pix=10 radius_y_pix=10 radius_z_pix=0 Nb_cpus=4");
IJ.getImage().setTitle("gb_th")
# threshold
IJ.run("Duplicate...","title=bw duplicate stack")
# maybe global TH now, because there is already a tophat?
#IJ.run("Auto Threshold", "method=Default white stack use_stack_histogram");
IJ.run("Auto Local Threshold", "method=Niblack radius=40 parameter_1=3 parameter_2=0 white stack");
# segment particles
#IJ.getImage().setRoi(roi_x, roi_y, roi_w, roi_h)
IJ.run("Set Measurements...", " mean min integrated center stack redirect=gb_th decimal=2");
IJ.run("Analyze Particles...", "size=10-10000 pixel circularity=0.00-1.00 show=Masks display exclude clear stack");
IJ.getImage().setTitle("particles")
# measure particles
rt = Analyzer.getResultsTable()
# todo: add the particles that are actually analyzed (size filter, see above)
# combine for documentation
IJ.run("Combine...", "stack1=raw stack2=bw");
IJ.getImage().setTitle("combine_raw_bw")
IJ.run("Combine...", "stack1=combine_raw_bw stack2=particles");
IJ.getImage().setTitle("combine_raw_bw_particles")
impFRAPandSEG = IJ.getImage()
IJ.saveAs(impFRAPandSEG, "Tiff", pathFRAPandSEG)
# extract intensity informations
nb = [0 for i in range(nt)]
nc = [0 for i in range(nt)]
ib = [0 for i in range(nt)]
ic = [0 for i in range(nt)]
fb = [0 for i in range(nt)]
fc = [0 for i in range(nt)]
t = [i*dt for i in range(nt)]
if(rt.getColumnIndex("Slice")==-1):
state = "no particles at all"
else:
Slice = rt.getColumn(rt.getColumnIndex("Slice"))
Mean = rt.getColumn(rt.getColumnIndex("Mean"))
Max = rt.getColumn(rt.getColumnIndex("Max"))
X = rt.getColumn(rt.getColumnIndex("XM"))
Y = rt.getColumn(rt.getColumnIndex("YM"))
IntDen = rt.getColumn(rt.getColumnIndex("IntDen"))
for i in range(len(Slice)):
# inside bleach roi?!
it = int(Slice[i])-1
# todo: maybe use a dictionary instead (one can remove items and have differnt t
if ( (X[i]>roi_x) & (X[i]<roi_x2) & (Y[i]>roi_y) & (Y[i]<roi_y2) ):
nb[it] = nb[it]+1
ib[it] = ib[it]+IntDen[i] # the IntDen copes best with in and out of focus motions as well as shape changes
fb[it] = fb[it]+Max[i]
else:
nc[it] = nc[it]+1
ic[it] = ic[it]+IntDen[i]
fc[it] = fc[it]+Max[i]
# compute mean values per particle
for i in range(len(t)):
if nb[i]>0:
fb[i] = fb[i]/ib[i]
ib[i] = ib[i]/nb[i]
if nc[i]>0:
fc[i] = fc[i]/ic[i]
ic[i] = ic[i]/nc[i]
nb1 = [(0,1)[i==1] for i in nb]
iShortlyAfter = iStart+10
if( sum(nb1[0:iStart]) < 0.5*iStart ):
state = "not enough pre-bleach measurements"
elif( sum(nb1[iStart:iShortlyAfter]) < 0.5*10 ):
state = "not enough short term measurements"
elif( sum(nb1[iShortlyAfter+1:len(nb1)-1]) < 0.5*(len(nb1)-(iShortlyAfter+1)) ):
state = "not enough long term measurements"
plot_size_x = 500
plot_size_y = 500
# plot number of particles
plotParticles = Plot( "Particles", "time", "number of particles", t, nb )
plotParticles.setFrameSize(plot_size_x,plot_size_y)
plotParticles.setSize(plot_size_x,plot_size_y)
plotParticles.setLimits(min(t),max(t),0,1.2*max(max(nb),max(nc)))
plotParticles.addPoints( t, nb, 3 )
plotParticles.addPoints( t, nc, 4 )
plotParticles.show()
# plot raw intensities
plotIntensities = Plot( "Intensities", "time", "gray values in particles", t, ib )
plotIntensities.setFrameSize(plot_size_x,plot_size_y)
plotIntensities.setSize(plot_size_x,plot_size_y)
plotIntensities.setLimits(min(t),max(t),0,1.2*max(max(ic),max(ib)))
plotIntensities.addPoints( t, ib, 3 )
plotIntensities.addPoints( t, ic, 4 )
#plotIntensities.show()
# plot sharpness
plotSharpness = Plot( "Sharpness", "time", "Max/IntDen", t, fb )
plotSharpness.setFrameSize(plot_size_x,plot_size_y)
plotSharpness.setSize(plot_size_x,plot_size_y)
plotSharpness.setLimits(min(t),max(t),0,1.2*max(max(fc),max(fb)))
plotSharpness.addPoints( t, fb, 3 )
plotSharpness.addPoints( t, fc, 4 )
#plotSharpness.show()
# FITTING
state = "ok"
# extract all time points relevant to the fitting
xTmp = []
yTmp = []
#print len(t)
#print range(iStart,len(t))
ipb = ib[0:iStart-1]
yTmpNorm = max(1,sum(ipb)/len(ipb)) # in order to avoid division by zero
#print ib[iStart-1]
for i in range(iStart,len(t)):
if nb1[i]==1: # only time-point with 1 particle
xTmp.append(float(t[i]-t[iStart]))
yTmp.append(float(ib[i]/yTmpNorm))
# do the fitting
# todo: how to add initial guesses??
imFrac = 0
tau = 0
xFit = []
yFit = []
if(len(yTmp)>20):
cf = CurveFitter( xTmp, yTmp )
cf.doFit(cf.EXP_RECOVERY)
print cf.getFormula()
p = cf.getParams()
for i in p:
print i
imFrac = (1-p[2]-p[0])
tau = (1/p[1])
for i in range(len(xTmp)):
xFit.append(float(xTmp[i]))
yFit.append(float(cf.f(cf.getParams(),xTmp[i])))
else:
state = "not enough data points for fitting"
# shift the fitting curves back to the original bleaching time point
#for i in range(len(xTmp)):
# xFit[i]=xFit[i]+t[iStart]
# xTmp[i]=xTmp[i]+t[iStart]
# plot fitting
plotFit = Plot( "Fitting", "time after bleach", "normalised intensity of bleached particle", xFit, yFit )
plotFit.setFrameSize(plot_size_x,plot_size_y)
plotFit.setSize(plot_size_x,plot_size_y)
if(len(xTmp)>0):
plotFit.setLimits(min(xTmp),max(xTmp),0,1.2*max(yTmp))
plotFit.addPoints( xTmp, yTmp, 3 )
#plotFit.addPoints( xFit, yFit, 4 )
plotFit.addLabel(0.1, 0.95, "imm_frac=%.2f tau[s]=%.2f" % (imFrac,tau))
plotFit.addLabel(0.1, 0.9, state)
# show the plots
IJ.run("Close all forced", "")
plotIntensities.show()
plotParticles.show()
plotSharpness.show()
plotFit.show()
# make one figure from the plots
IJ.run("Images to Stack", "name=Stack title=[] use");
IJ.run("Make Montage...", "columns=4 rows=1 scale=1 first=1 last=4 increment=1 border=0 font=12");
imp = IJ.getImage()
print "saving image: "+os.path.join(destDir, filename+"--IJ_graphs.png")
dest = os.path.join(destDir, filename+"--IJ_graphs.png")
IJ.saveAs(imp,"PNG", dest)
# write text files
dest = os.path.join(destDir, filename+"--intensBleach.csv")
writeXYfile(t,ib,dest,",")
dest = os.path.join(destDir, filename+"--intensCtrl.csv")
writeXYfile(t,ic,dest,",")
dest = os.path.join(destDir, filename+"--numParticlesBleach.csv")
writeXYfile(t,nb,dest,",")
dest = os.path.join(destDir, filename+"--numParticlesCtrl.csv")
writeXYfile(t,nc,dest,",")
IJ.run("Close all forced", "")
return()
