Classifying chromosomes with scikit-learn

A trained dataset was build from 5 metaphases corresponding to 205 classified chromosomes or nuclei falling in one of four categories: single chromosome, overlapping chromosomes, nuclei or dusts(artefact of image segmentation).  

Two features were chosen to classify the particles:

• The particle area (normalised by image size)
• The convex-hull area of the particle ( area/convex hull). The idea is that particles that are clustered chromosomes will have a smaller ratio than single chromosomes.

The trained dataset is used to recognise 49 particles segmented from a sixth metaphase and then the predictions from the classifier compared to hand classified particles reached...

                                                             59.18%

# -*- coding: utf-8 -*-
"""
Created on Fri Oct  7 13:08:57 2011

@author: Jean-Patrick Pommier
"""
import KarIO
import os
import pylab
import pandas
import numpy as np
from sklearn import svm
#make a configurator object
config=KarIO.ClassifConf()
#build the the name feature file
##list all the feature files
##répertoire courant : os.listdir(os.getcwd())
featurespath=os.path.join(os.getcwd(),"Results","features")
labelpath=os.path.join(os.getcwd(),"Results","labels")
#print featurespath
##open features csv files
featuresFilesList=os.listdir(featurespath)
labelsFilesList=os.listdir(labelpath)
##
metalist=[0,1,2,3,4]
def makeDataShape(meta):
    featfile=config.user+'-'+config.slide+'-'+config.metaphases_list[meta]+'-'+config.counterstain+'.csv'
    labelfile='shapeCateg-'+featfile
    fea=pandas.read_csv(os.path.join(featurespath,featfile),header=None,index_col=None,names=['particle','ratio','area'])
    lab=pandas.read_csv(os.path.join(labelpath,labelfile),header=None,index_col=None,names=['particle','type'])
    del lab['particle']    
    #merge columns:features and label
    data=fea.join(lab)
    data.insert(0,'meta',int(config.metaphases_list[meta]))
    #print data
    return data

bigdata=makeDataShape(0)
for meta in metalist:
    bigdata=bigdata.append(makeDataShape(meta),ignore_index=True)

single=bigdata[bigdata['type']=='single']
touching=bigdata[bigdata['type']=='touching']
nuclei=bigdata[bigdata['type']=='nuclei']
dusts=bigdata[bigdata['type']=='dusts']

##ploting with pylab
#different colors according to the category

fig=pylab.figure()
ax = fig.add_subplot(111)
ax.scatter(single['ratio'],single['area'],c='green',marker='o')
ax.scatter(touching['ratio'],touching['area'],c='red',marker='o')
ax.scatter(nuclei['ratio'],nuclei['area'],c='blue',marker='o')
ax.scatter(dusts['ratio'],dusts['area'],c='pink',marker='o')

#train a classifier
trainedData=bigdata[bigdata['meta']<15]
untrained=bigdata[bigdata['meta']>=15]
print 'trained data'
print trainedData[:5]
#extract two columns from trainedData
#convert to numpy array
features=trainedData.ix[:,['ratio','area']].as_matrix(['ratio','area'])
test_features=untrained.ix[:,['ratio','area']].as_matrix(['ratio','area'])
print 'features'
print features[:5]
print 'features shape',features.shape
print 'features type',type(features)
##label is a string:single, touching,nuclei,dust
print 'labels convertion'
lab1=trainedData['type']
print 'lab1',type(lab1)
f=pandas.Factor(lab1)
print 'factor f',type(f)
print 'labels',f.labels[:5]
print 'labels type',type(f.labels)
print 'labels shape',f.labels.shape
#
##Classify with sklearn
classifier = svm.SVC()
model = classifier.fit(features,f.labels)
predicted=classifier.predict(test_features)

#match predicted /classified
hiddenlab1=untrained['type']
hiddf=pandas.Factor(hiddenlab1)
match=(predicted==hiddf.labels)
print"prediction"
print predicted[:5]
print 'true classification'
print hiddf.labels[:5]
print 'match'
print match[:5]
##Count sucess
success=np.sum(match[:]==True)
rate=100.0*success/(1.0*len(match))
print 'rate of good classification',success,'out of',len(match),'particles'
print rate,'% success'

Highlighting Interstitial Telomeric Spots in interphasic cells (HITS-FISH)

Telomeric fusions underscore telomeric binding proteins disfunctions. Fusions occur in senescent cells between short telomeres, they are detected in metaphasic cells by dicentric chromosomes most generally with no telomeric spots at the fusion point.

In some SV40 transformed fibroblast cell line, interstitial telomeric spots (ITS) are as bright as terminal telomeric spots, this indicates some uncoupling between telomere lenght and their capacity to block telomeric fusions. 

ITS observed in a precrisis human SV40 transformed cell line (Ducray et al., oncogene 1999)

From metaphasic chromosomes observation, fusions retaining visible telomeric spots are rare events but may be more frequent in interphasic cells.

ITS from a SV40 transformed cell line

Bal31 in-situ:

Several nucleases or DNA polymerases  were used in assays requiring fixed material (chromosome banding+restriction endonuclease, in-situ nick translation, fluorescent labelling with TUNEL, in-situ PCR). The exonuclease Bal31 removes totally terminal telomeric sequences of naked DNA after some hours of digestion (R C Allshire, M Dempster, and N D Hastie, NAR 1989; de Lange et al, MCB 1990) . In-situ,  the terminal telomeric sequences, should be removed by Bal31 digestion.  If telomeric spots can be detected in nuclei after Bal31 digestion, those spots should be  ITS. Monitoring ITS may be a tool of interest  to probe telesome functions at the scale of unique cell, or to monitor the state of a tissue.

Pig cells as model cells?

On human metaphasic chromosomes, a bright ITS indicates an  abnormal cell but for cheaper DAPI stained chromosomes rearrangements too. Normal pig karyotype  shows interstitial telomeric spots:

Normal diploid pig metaphase with a chromosomes pair having ITS.

After QFISH on interphasic pig nuclei, there are two ITS of same magnitude than terminal spots, these two pig ITS remain hidden in the forest of terminal telomeric spots:

  

Pig nuclei (25x magnification)

Some conditions of Bal31 digestion should retain these two ITS and remove totally terminal telomeric spots yielding the possibility to observe telomeric fusions in nuclei.

Assay:

Tuning the conditions of Bal31 digestion is a prerequisite step. Supposing that the exonuclease Bal31 can remove terminal sequence from fixed material, when this enzyme is usually used to digest purified DNA in solution. 

If pig cells are not available, performing Bal31 digestion on control human cells such lymphocytes should tell if Bal31 can remove telomeres in-situ.

From conventionnal cytogenetic preparation of pig cells (fibroblasts,lymphocytes), digestion time, numbers of Bal31 units could be set up as follow : 

Possible protocol for HITS FISH

The protocol modified from Lansdorp et al. (Hum Mol Genet. 1996 May;5(5):685-91.) could be also adapted to liquid hybridization for flow FISH. Under optimal conditions pig cells could be used as internal control for Bal 31 digestion by mixing human and pig cells in the same cytogenetic preparation (9:1 ratio).

Human cells can be then distinguished from pig cells with an additionnal oligo probe. In the following example (25x magnification), simultaneous hybridization with Cy3-PNA-(CCCTAA)3 (displayed in green) and Cy5-DNA (displayed in red) oligo probe was performed (unpublished data):

Candidate cells to test the assay, may be SV40 precrisis cells with highly rearranged chromosomes and numerous dicentrics, or better cells with controlled mutations.