Home

submission

Help

Output format

Algorithm

Team

Algorithm

General Basis of Predictions

• Parameters such as hydrophilicity, flexibility, accessibility, turns, exposed surface, polarity and antegenic propensity of polypeptides chains have been correlated with the location of continuous epitopes in a few well-characterized proteins. This has led to a search for emperical rules that would allow the position of continuous epitopes to be predicted from certain features of the protein sequence. All prediction calculations are based on propensity scales for each of the 20 amino acids.

• Each scale consists of 20 values assigned to each of the amino acid residues on the basis of their relative propensity to possess the property described by the scale

  SCALE	X	A	R	N	D	C	Q	E	G	H	I	L	K	M	F	P	S	T	W	Y	V
  Hydrophilicity	0.00	2.10	4.20	7.00	10.00	1.40	6.00	7.80	5.70	2.10	-8.00	-9.20	5.70	-4.20	-9.20	2.10	6.50	5.20	-10.0	-1.9	-3.7
  Flexibility	0.00	-1.27	2.79	1.77	1.42	-1.09	1.18	1.60	1.86	-0.82	-2.89	-2.29	2.88	-1.84	-2.14	0.52	3.00	1.18	-3.78	-3.3	-1.75
  Accessibility	0.00	0.49	0.95	0.81	0.78	0.26	0.84	0.84	0.48	0.66	0.34	0.40	0.97	0.48	0.42	0.75	0.65	0.70	0.51	0.76	0.36
  Exposed	0.00	15.00	67.00	49.00	50.00	5.00	56.00	55.00	10.00	34.00	13.00	16.00	85.00	20.00	10.00	45.00	32.00	32.00	17.00	41.00	14.00
  Turns	0.00	-0.8	-1.15	3.00	1.65	0.83	0.11	-0.92	-0.55	2.56	-1.53	-1.01	-1.06	-1.48	0.18	0.64	1.34	0.27	-0.97	-0.29	-0.83
  Antegenic	0.00	1.064	0.873	0.776	0.866	1.412	1.015	0.851	0.874	1.105	1.152	1.250	0.930	0.826	1.091	1.064	1.012	0.909	0.893	1.161	1.383
  Polarity	5.00	0.00	52	3.38	40.7	1.48	3.53	49.91	0.00	51.6	0.15	0.45	49.50	1.43	0.35	1.58	1.67	1.66	2.1	1.61	0.13

  
  

  Brief description of each methods:

  Parker Method: In this method, hydrophilic scale based on peptide retention times during high-performance liquid chromatography (HPLC) on a reversed-phase column was constructed (Parker et al., 1986). A window of seven residues was used for analyzing epitope region. The corresponding value of the scale was introduced for each of the seven residues and the arithmetical mean of the seven residue value was assigned to the fourth, ( i+3), residue in the segment.

  Karplus Method: In this method, flexibility scale based on mobility of protein segments on the basis of the known temperature B factors of the a-carbons of 31 proteins of known structure was constructed (Karplus and Schulz, 1985). The calculation based on a flexibility scale is similar to classical calculation, except that the center is the first amino acid of the six amino acids window length, and there are three scales for describing flexibility instead of a single one. In the present study, 3Karplus scale was used for the prediction of the epitope region.

  Emini Method: The calculation was based on surface accessibility scale on a product instead of an addition within the window. The accessibility profile was obtained using the formulae Sn = ( n+4+i) (0.37) -6 Where Sn is the surface probability, dn is the fractional surface probability value, and i vary from 1 to 6. A hexapeptide sequence, with Sn equal to unity and probabilities greater than 1.0 indicates an increased probability for being found on the surface (Emini et al., 1985).

  Pellequer Method: This method is based on incidence of b turns (Pellequer et al., 1993). The calculation was based on a turn scale and there were three scales for describing turns instead of a single one. A window of seven residues is used for analyzing epitope region. The corresponding value of the scale was introduced for each of the seven residues and the arithmetical mean of the seven residue value is assigned to the fourth, (i+3), residue in the segment. Gaussian smoothing curve was used, which assigns the residue weights in a window of seven residues (the weights were 0.05/0.11/0.19/0.22/0.19/0.11/0.05).

  Kolaskar Method: In this method 156 antigenic determinants (< 20 amino acids) in 34 different proteins were analyzed (Kolaskar and Tongaonkar, 1990) to calculate the antigenic propensity (Ap ) of residues. This antigenic scale was used to predict the epitopes in sequence.

  Exposed surface scale and Polarity scale: The physico-chemical properties like exposed surface (Janin and Wodak, 1978) and polarity (Ponnuswamy et al., 1980) has also been evaluated in this study. A window of seven residues has been used for analyzing the epitope region. The corresponding value of the scale has been introduced for each of the seven residues and the arithmetical mean of the seven residue value is assigned to the fourth, (i+3), residue in the segment.
  

  
  

  Measure of prediction accuracy

  Both threshold dependent and independent measures have been used to evaluate the prediction performance.
  

  Threshold dependent measures

  Three parameters were used to measure the performance of prediction methods. Following is the brief description of these parameters:
  
  (i) Sensitivity (Qsens) of the prediction methods is defined as the percentage of epitopes that are correctly predicted as epitopes. Higher sensitivity means that almost all of the potential epitopes will be included in the predicted results.
                   Qsens=(TP/TP+FN)*100
  Where TP and FN refer to true positives and false negatives respectively.
  
  (ii) Specificity (Qspec) of the prediction methods is defined as the percentage of correctly predicted non-epitopes.
                   Qspec=(TN/TN+FP)*100
  
  Where TN and FP refer to true negatives and false positives.
  
  (iii) A commonly used parameter to measure the prediction performance is accuracy, is defined as
                  Qacc=(TOTc/Total)*100
  TOTc is the total number of correct predictions (includes both true positives, TP and true negatives, TN) and Total is the total number of prediction made.
  
  All the three parameters, sensitivity, specificity, and accuracy were calculated for each polypeptide sequence at a threshold of -1.5 to +3.

  Threshold independent measures-ROC

  One problem with the threshold dependent measures is that they measure the performance at a given threshold. They fail to use all the information provided by a method for evaluation. The Receiver Operating Characteristics (ROC) is a threshold independent measure that has been developed as a signal processing technique (Deleo, 1993). A ROC plot is obtained by plotting all sensitivity values (true positive fraction) on the y-axis against their equivalent (1 - specificity) values (false positive fraction) for all available thresholds on the x-axis. The area under the ROC curve measures discrimination, the ability of a method to correctly classify epitopes and non-epitopes.

  
  

  Normalization procedure

  Each property scale consists of 20 values assigned to each of the amino acid types on the basis of their relative propensity as described by the scale. In order to compare the profiles obtained by different methods, normalization of the various scales is done. We calculated the average of seven maximum and seven minimum values of a given physico-chemical scale and then calculated the difference between the two. The original values of the each scale are set between +3 to -3 by using the formulae
             Normalization Score =AMS/DS * 6
  Where AMS refer to Average of seven maximum/minimum values from the physico-chemical scale and DS refer to difference between the maximum and minimum score. Normalization score are set to +3 (Maximum) and -3 (Minimum) by subtracting or adding additional values.
  
  

  References

  1. Hydrophilicity scale
       J.M.D.Parker, D.Guo, and R.S.Hodges, Biochemistry, 25, 5425 (1986)
     
  2. Flexibility scale
       P.A.Karplus and G.E.Schulz, Naturwissenschaften,  72, 212 (1985)
     
  3. Accessibility scale
        E.A. Emini,J.V.Hughes,D.S.Perlow,J.Boger, J.Virol.,  55, 836 (1985)
     
  4. Turns scale
       J-L Pellequer, E.Westhof,Regenmortel MHV, Immunol.Lett.,  36, 83 (1993)
     
  5. Antegenic propensity scale
       A.S.Kolaskar and P.C.Tongaonkar, FEBS,  276, 172 (1990)
     
  6. Exposed Residues scale
       J.Janin and S.Wodak, J.M.Biol.,  125, 357 (1978)
     
  7. Polarity scale
       P.K.Ponnuswamy,M.Prabhakaran,P.Manavalan, Biochim.Biophys.Acta.,   623, 301 (1980)
     

  Home

  ---

  contact:   G.P.S.Raghava                                                                  Department of Computational Biology                                                                 IMTECH

  ---