J. Biomedical Science and Engineering, 2009, 2, 51-56
Published Online February 2009 in SciRes. http://www.scirp.org/journal/jbise JBiSE
51
Using position specific scoring matrix and auto
covariance to predict protein subnuclear localization
Rong-Quan Xiao1, Yan-Zhi Guo2, Yu-Hong Zeng2, Hai-Feng Tan1, Xue-Mei Pu2, Meng-Long Li1,2*
1College of Life Sciences, Sichuan University, Chengdu 610064. 2College of Chemistry, Sichuan University, Chengdu 610064, P.R. China. Correspondence
should be addressed to Meng-Long Li([e-mail]). Tel: [phone]; Fax: [phone].
Received September 8th, 2008; revised November 13th, 2008; accepted November 20th, 2008
ABSTRACT
The knowledge of subnuclear localization in
eukaryotic cells is indispensable for under-
standing the biological function of nucleus,
genome regulation and drug discovery. In this
study, a new feature representation was pro-
posed by combining position specific scoring
matrix (PSSM) and auto covariance (AC). The
AC variables describe the neighboring effect
between two amino acids, so that they incorpo-
rate the sequence-order information; PSSM de-
scribes the information of biological evolution
of proteins. Based on this new descriptor, a
support vector machine (SVM) classifier was
built to predict subnuclear localization. To
evaluate the power of our predictor, the
benchmark dataset that contains 714 proteins
localized in nine subnuclear compartments was
utilized. The total jackknife cross validation ac-
curacy of our method is 76.5%, that is higher
than those of the Nuc-PLoc (67.4%), the OET-
KNN (55.6%), AAC based SVM (48.9%) and ProtLoc
(36.6%). The prediction software used in this
article and the details of the SVM parameters are
freely available at http://chemlab.scu.edu.cn/
predict_SubNL/index.htm and the dataset used
in our study is from Shen and Chou’s work by
downloading at http://chou.med.harvard.edu/
bioinf/Nuc-PLoc/Data.htm.
Keywords: Position Specific Scoring Matrix;
Auto Covariance; Support Vector Machine; Pro-
tein Subnuclear Localization Prediction
1. INTRODUCTION
The cell nucleus is complex, important subcellular or-
ganelle in eukaryotes cell. It organizes the comprehen-
sive assembly of our genes and their corresponding
regulatory factors [1]. Meanwhile, it also reflects various
intricate biological activities, and controls various kinds
of biologic processes [2]. Many proteins, from outside a
nuclear, trend to be localized into specific subnuclear
locations of the nucleus [3]. If proteins can not be cor-
rectly localized into its specific subnuclear locations in
human, it will lead to genetic disease [4], cancer [5] or
virally infected cells [6]. Thus, it’s desirable to get the
knowledge of protein subnuclear localization for in-
depth understanding cell biological processes and ge-
nomic regulation. However, it is costly and time-con-
suming to assay the subnuclear localization of proteins
by biology experiments [7]. The number of protein se-
quences is increasing more rapidly than that of identified
proteins [7]. So it is of great practical significance to
develop computational approaches for identifying the
protein subnuclear localizations in cell nucleus. At the
same time, many lines of evidences have indicated that
computational approaches, such as structural bioinfor-
matics [8], molecular docking [9], pharmacophore mod-
elling [10], QSAR [11,12,13], protein subcellular loca-
tion prediction [7,14], identification of membrane pro-
teins and their types [15], identification of enzymes and
their functional classes [16], identification of proteases
and their types [17], protein cleavage site prediction
[18,19], and signal peptide prediction [20,21] can pro-
vide very useful information for both basic research and
drug discovery in a timely manner. The present study is
devoted to develop a new method for predicting protein
subnuclear localization in hope to stimulate the devel-
opment of the relevant areas.
Recently, many algorithms have already been devel-
oped for predicting protein subcellular localizations [22,
23,24,25,26,27,28,29,30,31,32,33], as reviewed by Chou
[7]. Even several web severs have been constructed for
predicting subcellular localization of various organisms
[14,34,35,36,37]. However, there are only a few compu-
tational methods for predicting protein subnuclear local-
ization [38,39,40,41], such as OET-KNN [42], ProLoc
[43], Nuc-PLoc [44], and AdaBoost classifiers [45].
Compared to the conventional amino acid composition
(AAC), pseudo amino acid (PseAA) composition [46],
originally introduced by Chou [47,48], can include the
sequence-order information of sequences. Similarly, the
PsePSSM was also proposed by Shen and Chou in order
to incorporate the evolution information of proteins [44].
They built a new web server called Nuc-PLoc for pre-
dicting protein subnuclear localization by fusing PseAA
composition and PsePSSM with a promising prediction
result. In this study, we developed a new method by fus-
SciRes Copyright © 2009

52 R. Q. Xiao et al. / J. Biomedical Science and Engineering 2 (2009) 51-56
SciRes Copyright © 2009 JBiSE
ing position specific scoring matrix (PSSM) and auto
covariance (AC), so that this method can incorporate
sequence-order information by AC and the evolutionary
information by PSSM. A classifier based on SVM was
constructed to predict protein subnuclear localization
using jackknife test. The result indicates that our method
has successfully enhanced accuracies of the existing
methods for predicting protein subnuclear localization.
2. MATERIALS AND METHODS
2.1. Data Sets
In this paper, our dataset is obtained from article by Shen
and Chou [44]. And anyone can freely download it at this
page (http://chou.med.harvard.edu/bioinf/Nuc-PLoc/
Data.htm). This dataset consists of nine classes and 714
proteins in total. Details of this benchmark dataset are
shown in Table 1. S
i (i=1, 2… 9) is used to represent
each of nine subsets and S represents the total dataset.
2.2. Feature Representations
2.2.1. Auto Covariance (AC)
We selected three common physicochemical properties,
hydrophobicity [49], volumes of side chains of amino
acids [50], and polarity [51], to represent the structure
and function [52], the stereospecific blockade [53] and
the electronic property [54] of residues in a protein re-
spectively. These original values were taken from Guo et
al. [55] and were first normalized to zero mean value and
unit standard deviation (SD) by Equation (1):
,
'
,
ij
j
ij
j
PP
PS
−
= (1)
(i=1, 2, 3; j=1, 2, 3…, 20.)
Where Pi,j is the i-th descriptor value for j-th amino
acid, Pj is the mean of the j-t h descriptor of the 20
amino acids and Sj is the value of SD. So each protein
sequence was translated into three vectors with each
amino acid represented by the normalized values.
There are many approaches to convert the protein se-
quences into numerical order sequences, including auto-
correlations and auto covariance (AC). Autocorrelations,
quite similar to AC, has been used in the prediction of
secondary structure content [56,57,58] and structural
class [59,60,61,62]; however, AC as a statistical tool for
Table 1. The benchmark dataset consists of 714 nuclear proteins
classified into nine subnuclear localizations
Subnuclear localization Subset No. of proteins
Chromatin S1 99
Heterochromatin S2 22
Nuclear envelope S3 61
Nuclear matrix S4 29
Nuclear pore complex S5 79
Nuclear speckle S6 67
Nucleolus S7 307
Nucleoplasm S8 37
Nuclear PML body S9 13
Total S 714
analyzing sequences of vectors has also been success-
fully adopted by our research group for protein classifi-
cations [55,63] from primary sequence. So in our study,
AC was selected to transform these numerical vectors
into uniform matrices in order to take the neighboring
effect of the sequences into account. Here, lag is the dis-
tance between one residue and its neighbour, a certain
number of residues away. The AC variables are calcu-
lated by the Equation (2) [55].
,, ,
,(),
11 1
11
1
()( )
−
+
== = −
=−×−
∑∑
∑
Llag L
L
ij ijij
lag jilagj
ii i Llag
A
CPPPP
L L
(2)
Where i is the position in the sequence P, j is one de-
scriptor, L is the length of the sequence P and lag is the
value of the lag.
In this way, the number of AC variables, D, can be
calculated according to Equation (3) [55].
Dlgp
=
× (3)
Where lg is the maximum lag (lag=1, 2, 3…, lg) and p
represents the number of descriptors.
2.2.2. Position Specific Scoring Matrix (PSSM)
A PSSM is a Position Specific Scoring Matrix and is a
commonly used representation of motifs (patterns) in
biological sequences [64]. So far, this method has been
used for predicting protein subcellular localization [65]
and subnuclear localization [40,44].
For a protein sequence P with L amino acid residues,
PSSM is obtained according to the following Equation [44].
111 211 20
212 22220
12 20
12 20
j
j
PSSM
iiij i
LLLj L
PPPP
PP P P
PPP P P
PPPP
→→ → →
→→→→
→→ → →
→→ → →
⎡
⎤
⎢
⎥
⎢
⎥
⎢
⎥
⎢
⎥
⎢
⎥
=
⎢
⎥
⎢
⎥
⎢
⎥
⎢
⎥
⎢
⎥
⎢
⎥
⎣
⎦
LL
LL
MMMMM M
LL
MMMMM M
LL
(4)
In Equation (4), where i→j describes i-th amino acid
residue of the protein sequence P being mutated to
amino acid type j in the biology evolution process, Pi→j
is the score of this mutation and L is the length of the
sequence P. Here we used the numerical codes 1, 2, 3…
20 to represent the single character of ordered 20 native
amino acid types in Equation (4). To get the 20L
×
scores of the PPSSM in the Equation (4), we used three
iterations of PSI-BLAST [66] with default threshold (the
default E-value is 0.001) to search the Swiss-Prot database
(version 54.4, released on 25 Oct. 2007) for multiple
sequence alignment against the protein P. Then, the
value of Pi→j is standardized by Equation (5), as given
below.
20
1
1
20
)) max( min(
o
ij
j
ij
o
ij
oo
ij ij
PP
PPP
→
=
→
→
→→
−∑
=− (5)

R. Q. Xiao et al. / J. Biomedical Science and Engineering 2 (2009) 51-56 53
SciRes Copyright © 2009 JBiSE
(i= 1, 2, 3… L; j= 1, 2, 3…20)
Where o
ij→
Pis the original scores generated by
PSI-BLAST, ij→ P is a zero mean value over the 20
native amino acids and the value is between -1 and 1.
However, because of proteins with different lengths L,
the matrices of the PSSM descriptor in Equation (4) have
different numbers of rows. To gain the uniform matrix
for protein sequences of different lengths, we converted
the PSSM of protein P to a uniform vector through the
Equation (6) [44].
12 20
(1,2, ,20)
PSSM jj
PPP P P
Τ
⎡⎤
⎢⎥
⎣⎦
==LL L (6)
Where T is the transpose operator, j
P is the average
score over j-th column in Equation (4).
Finally, the
P
SSM
P describes the evolutionary infor-
mation of a protein sample, and AC variables contain the
interaction information between two amino acid residues
of a sequence. So each protein sequence was converted
into a numerical vector by concatenating PSSM and AC. Here,
each AC variable was appended a weight factor of 0.05.
2.2.3. Accuracy and Matthew's Correlation Coef-
ficient (MCC)
To evaluate the performance of this method, two pa-
rameters, accuracy and Matthew's correlation coefficient
(MCC), were selected in this article. They are calculated
by Equation (7) and Equation (8), respectively.
TP
Accuracy TP FN =+ (7)
)()()()( FNTNFPTNFNTPFPTP
FNFPTNTP
MCC +×+×+×+
×
−
×
=
(8)
Where TP represents the true positive; TN, the true nega-
tive; FP, the false positive and FN, the false negative.
3. RESULTS AND DISCUSSION
In statistical prediction, the following three cross-vali-
dation methods are often used to examine a predictor
for its effectiveness in practical application: independ-
ent dataset test, subsampling test, and jackknife test
[67]. However, as elucidated in [14] and demonstrated
by Eq.50 of [7], among the three cross-validation
methods, the jackknife test is deemed the most objective
that can always yield a unique result for a given bench-
mark dataset, and hence has been increasingly used by
tigators to examine the accuracy of various predictors
(see, e.g., [7,33,68,69,70,71,72,73,74,75,76,77,78,79,80,
81,82]). So in this paper, the jackknife test was chosen to
validate the current algorithm. Because the benchmark
dataset used has nine subsets, the one-to-one multiclass
classification system led to 9*(9-1)/2=36 SVM models
for one single encoding methods. Meanwhile, for AC
variables, the value of lg was optimized as 13 through a
series of control experiments, and the value of p is 3. So,
the number of AC variables, D, is 39 (Dlgp
=
×
13 339
=
×= ) according to Equation (3).
Amino acid composition (AAC) has been widely used
for predicting subcellular localizations [7,14,22,23,24,
25,26,27,28,30,31,32,34,35,36,37,83,84,85], so it was
also used as a substitution model in our study. And thus,
three SVM models based on AAC, AC and PSSM, were
respectively constructed.
The results according to jackknife test are listed in
Table 2. As can be seen from Table 2, the prediction
accuracy of PSSM based model is nearly equal to that of
AAC based model. However, AC based model gives the
lower accuracy of 64.13%. Then we constructed models
by fusing the three substitution models, so four fused
classifier were built. Table 2 shows that the accuracies of
the four fused models are higher than those of the three
anterior models. Among those four fused models, the
accuracy of the model combining PSSM, AAC and AC is
lower than that of PSSM and AC based model that ob-
tains the best performance with an accuracy of 76.45%.
So the final SVM model was built based on PSSM and
AC. The kernel function of SVM is radio basis function
(rbf), and the parameters of C and γ are listed in the table
by downloading at http://chemlab.scu.edu.cn/predict_
SubNL/index.htm.
In order to further examine the prediction power of the
current classifier, the performance of this method was
also compared with those of the existing methods on the
same training dataset. The results obtained by several
algorithms with different substitution models were sum-
marized in Table 3. From Table 3, we can see that the
accuracy obtained by Nuc-PLoc [44] is much higher than
those of ProtLoc [43], AAC based SVM and OET-KNN [42].
When compared to Nuc-PLoc, our method obtains a better
performance with the accuracy of 76.5%. It means our
method is successful in predicting protein subnuclear
localization only using primary sequences of proteins
Table 2. Overall accuracies by jackknife tests with different substitution models on the benchmark dataset of Table 1
Substitution Model AACa ACb PSSMc AAC+ACd PSSM+AAC PSSM+ACd PSSM+AAC+ACd
Accuracy 73.82% 64.13% 73.85%74.05% 75.97% 76.45% 75.99%
a: Amino acid composition
b: Auto covariance
c: Position specific scoring matrix
d: While fused models were constructed, a weight factor added on AC is 0.05.

54 R. Q. Xiao et al. / J. Biomedical Science and Engineering 2 (2009) 51-56
SciRes Copyright © 2009 JBiSE
Table 3. Overall accuracy by jackknife tests with different algorithms on the benchmark dataset of Table 1
Algorithm Protein sample descriptor Overall accuracy
ProtLoca,d Amino acid composition 261/714=36.6%
SVMd Amino acid composition 349/714=48.9%
OET-KNNb,d PseAA Composition 397/714=55.6%
Nuc-PLocc,d Fusion of PsePSSM and PseAA Composition 481/714=67.4%
Our method Combination of PSSM and AC 546/714=76.5%
a: See Cedano et al. (1997)[86]
b: See Shen and Chou (2005)[42]
c: See Shen and Chou (2007)[44]
d: The results were from Shen and Chou (2007)[44], and the original data could been seen in that article.
Table 4. The MCC values obtained by the jackknife tests with
Nuc-PLoc and our method on the benchmark dataset of Table 1
Matthew’s correlation
coefficient
Subnuclear localization
Nuc-PLoca Our methodb
Chromatin S1 0.60 0.55
Heterochromatin S2 0.52 0.58
Nuclear envelope S3 0.53 0.65
Nuclear matrix S4 0.52 0.61
Nuclear pore complex S5 0.70 0.72
Nuclear speckle S6 0.43 0.57
Nucleolus S7 0.57 0.57
Nucleoplasm S8 0.31 0.54
Nuclear PML body S9 0.32 0.51
a: The results were from Shen and Chou (2007)[44], and the
original data could be seen in that article.
b: The classifier fused PSSM and AC.
In addition, to evaluate the stability of our method, the
values of the MCC for the nine subsets were compared
based on Nuc-PLoc and our current predictor, respec-
tively, as seen in Table 4. For nine subsets, our method
yields a higher MCC than Nuc-PLoc, except the subset
S1. So, compared to the existing methods, our classifier
combined with PSSM and AC has further improved the
prediction accuracy of protein subnuclear localization.
4. CONCLUSION
In this paper, a new classifier was developed by fusing
PSSM and AC for predicting protein subnuclear local-
ization only using the primary sequences of nuclear pro-
teins. The SVM predictor was constructed based on
PSSM and AC. AC variables represent the interactions
between amino acids in protein sequences; PSSM de-
scribes the evolutionary information. So the method in-
corporated not only the evolution information, but also
the sequence-order information. Compared with the cur-
rent methods, this method successfully raises the predic-
tion accuracy. Hence, it may be a good supplementary
tool for protein function studies.
ACKNOWLEDGEMENT
The authors gratefully thank Shen and Chou for sharing the benchmark
dataset. The work was funded by the National Natural Science Founda-
tion of China (No. 20775052).
REFERENCES
[1] G. S. Stein, J. B. Lian, W. A. van, J. L. Stein, A. Javed, G. Bar-
nes, L. Gerstenfeld, D. Vradii, S. K. Zaidi, and J. Pratap, et al.
(2006) Organization of transcriptional regulatory machinery in
nuclear microenvironments: Implications for biological control
and cancer. Cancer Treatment Reviews, 32, 13-13.
[2] A. I. Lamond, and W. C. Earnshaw, (1998) Structure and func-
tion in the nucleus. Science, 280, 547-553.
[3] J. M. Bridger, and W. A. Bickmore, (1998) Putting the genome
on the map. Trends in Genetics, 14, 403-409.
[4] H. G. Sutherland, G. K. Mumford, K. Newton, L. V. Ford, R.
Farrall, G. Dellaire, J. F. Caceres, and W. A. Bickmore, (2001)
Large-scale identification of mammalian proteins localized to
nuclear sub-compartments. Human Molecular Genetics, 10,
1995-2011.
[5] S. K. Zaidi, D. W. Young, A. Javed, J. Pratap, M. Montecino, W.
A. van, J. B. Lian, J. L. Stein, and G. S. Stein, (2007) Nuclear
microenvironments in biological control and cancer. Nature Re-
views Cancer, 7, 454-463.
[6] R. D. Phair, and T. Misteli, (2000) High mobility of proteins in
the mammalian cell nucleus. Nature, 404, 604-609.
[7] K. C. Chou, and H. B. Shen, (2007) Recent progress in protein
subcellular location prediction. Analytical Biochemistry, 370,
1-16.
[8] K. C. Chou, (2004) Structural bioinformatics and its impact to
biomedical science. Current Medicinal Chemistry, 11, 2105-
2134.
[9] K. C. Chou, D. Q. Wei, and W. Z. Zhong, (2003) Binding
mechanism of coronavirus main proteinase with ligands and its
implication to drug design against SARS. Biochemical and Bio-
physical Research Communications, 308, 148-151.
[10] Sirois, S., Wei, D. Q., Du, Q. S. and Chou, K. C. (2004) Virtual
screening for SARS-CoV protease based on KZ7088 pharma-
cophore pointst. Journal of Chemical Information and Computer
Sciences, 44, 1111-1122.
[11] Du, Q. S., Mezey, P. G. and Chou, K. C. (2005) Heuristic mo-
lecular lipophilicity potential (HMLP): A 2D-QSAR study to
LADH of molecular family pyrazole and derivatives. Journal of
Computational Chemistry, 26, 461-470.
[12] Du, Q. S., Huang, R. B., Wei, Y. T., Du, L. Q. and Chou, K. C.
(2008) Multiple field three dimensional quantitative struc-
ture-activity relationship (MF-3D-QSAR). Journal of Computa-
tional Chemistry, 29, 211-219.
[13] Prado-Prado, F. J., Gonzalez-Diaz, H., de, V. O., Ubeira, F. M.
and Chou, K. C. (2008) Unified QSAR approach to antimicrobi-
als. Part 3: First multi-tasking QSAR model for Input-Coded
prediction, structural back-projection, and complex networks
clustering of antiprotozoal compounds. Bioorganic and Medici-
nal Chemistry, 16, 5871-5880.
[14] Chou, K. C. and Shen, H. B. (2008) Cell-PLoc: a package of
Web servers for predicting subcellular localization of proteins in
various organisms. Nature Protocols, 3, 153-162.
[15] Chou, K. C. and Shen, H. B. (2007) MemType-2L: a web server

R. Q. Xiao et al. / J. Biomedical Science and Engineering 2 (2009) 51-56 55
SciRes Copyright © 2009 JBiSE
for predicting membrane proteins and their types by incorporat-
ing evolution information through Pse-PSSM. Biochemical and
Biophysical Research Communications, 360, 339-345.
[16] Shen, H. B. and Chou, K. C. (2007) EzyPred: a top-down ap-
proach for predicting enzyme functional classes and subclasses.
Biochemical and Biophysical Research Communications, 364,
53-59.
[17] Chou, K. C. and Shen, H. B. (2008) ProtIdent: A web server for
identifying proteases and their types by fusing functional domain
and sequential evolution information. Biochemical and Bio-
physical Research Communications, 376, 321-325.
[18] Chou, K. C. (1996) Prediction of human immunodeficiency
virus protease cleavage sites in proteins. Analytical Biochemis-
try, 233, 1-14.
[19] Shen, H. B. and Chou, K. C. (2008) HIVcleave: a web-server for
predicting human immunodeficiency virus protease cleavage
sites in proteins. Analytical Biochemistry, 375, 388-390.
[20] Chou, K. C. and Shen, H. B. (2007) Signal-CF: A sub-
site-coupled and window-fusing approach for predicting signal
peptides. Biochemical and Biophysical Research Communications,
357, 633-640.
[21] Shen, H. B. and Chou, K. C. (2007) Signal-3L: A 3-layer ap-
proach for predicting signal peptides. Biochemical and Bio-
physical Research Communications, 363, 297-303.
[22] Chou, K. C. (2000) Prediction of protein subcellular locations by
incorporating quasi-sequence-order effect. Biochemical and Bio-
physical Research Communications, 278, 477-483.
[23] Chou, K. C. and Cai, Y. D. (2002) Using functional domain
composition and support vector machines for prediction of pro-
tein subcellular location. Journal of Biological Chemistry, 277,
45765-45769.
[24] Cai, Y. D., Liu, X. J., Xu, X. B. and Chou, K. C. (2002) Support
vector machines for prediction of protein subcellular location by
incorporating quasi-sequence-order effect. Journal of Cellular
Biochemistry, 84, 343-348.
[25] Cai, Y. D., Liu, X. J. and Chou, K. C. (2002) Artificial neural
network model for predicting protein subcellular location.
Computers and Chemistry, 26, 179-182.
[26] Chou, K. C. and Cai, Y. D. (2004) Prediction and classification
of protein subcellular location-sequence-order effect and pseudo
amino acid composition. Journal of Cellular Biochemistry, 90,
1250-1260.
[27] Chou, K. C. and Cai, Y. D. (2004) Prediction of protein subcel-
lular locations by GO-FunD-PseAA predictor. Biochemical and
Biophysical Research Communications, 320, 1236-1239.
[28] Gao, Q. B., Wang, Z. Z., Yan, C. and Du, Y. H. (2005) Predic-
tion of protein subcellular location using a combined feature of
sequence. Febs Letters, 579, 3444-3448.
[29] Zhang, T. L., Ding, Y. S. and Chou, K. C. (2006) Prediction of
protein subcellular location using hydrophobic patterns of amino
acid sequence. Computational Biology and Chemistry, 30,
367-371.
[30] Chou, K. C. and Shen, H. B. (2006) Predicting protein subcellu-
lar location by fusing multiple classifiers. Journal of Cellular
Biochemistry, 99, 517-527.
[31] Chou, K. C. and Shen, H. B. (2006) Predicting eukaryotic pro-
tein subcellular location by fusing optimized evidence-theoretic
K-Nearest Neighbor classifiers. Journal of Proteome Research, 5,
1888-1897.
[32] Zhou, X. B., Chen, C., Li, Z. C. and Zou, X. Y. (2008) Improved
prediction of subcellular location for apoptosis proteins by the
dual-layer support vector machine. Amino Acids, 35, 383-388.
[33] Shi, F., Chen, Q. J. and Li, N. N. (2008) Hilbert Huang trans-
form for predicting proteins subcellular location. Journal of
Biomedical Science and Engineering, 1, 59-63.
[34] Chou, K. C. and Shen, H. B. (2006) Hum-PLoc: A novel ensem-
ble classifier for predicting human protein subcellular localiza-
tion. Biochemical and Biophysical Research Communications,
347, 150-157.
[35] Shen, H. B., Yang, J. and Chou, K. C. (2007) Euk-PLoc: an
ensemble classifier for large-scale eukaryotic protein subcellular
location prediction. Amino Acids, 33, 57-67.
[36] Shen, H. B. and Chou, K. C. (2007) Hum-mPLoc: An ensemble
classifier for large-scale human protein subcellular location pre-
diction by incorporating samples with multiple sites. Biochemi-
cal and Biophysical Research Communications, 355, 1006-
1011.
[37] Chou, K. C. and Shen, H. B. (2007) Euk-mPLoc: A fusion clas-
sifier for large-scale eukaryotic protein subcellular location pre-
diction by incorporating multiple sites. Journal of Proteome Re-
search, 6, 1728-1734.
[38] Lei, Z. D. and Dai, Y. (2005) An SVM-based system for pre-
dicting protein subnuclear localizations. BMC Bioinformatics, 6,
291-298.
[39] Lei, Z. D. and Dai, Y. (2006) Assessing protein similarity with
Gene Ontology and its use in subnuclear localization prediction.
BMC Bioinformatics, 7, 491-500.
[40] Mundra, P., Kumar, M., Kumar, K. K., Jayaraman, V. K. and
Kulkarni, B. D. (2007) Using pseudo amino acid composition to
predict protein subnuclear localization: Approached with PSSM.
Pattern Recognition Letters, 28, 1610-1615.
[41] Li, F. M. and Li, Q. Z. (2008) Using pseudo amino acid compo-
sition to predict protein subnuclear location with improved hy-
brid approach. Amino Acids, 34, 119-125.
[42] Shen, H. B. and Chou, K. C. (2005) Predicting protein subnu-
clear location with optimized evidence-theoretic K-nearest clas-
sifier and pseudo amino acid composition. Biochemical and Bio-
physical Research Communications, 337, 752-756.
[43] Huang, W. L., Tung, C. W., Huang, H. L., Hwang, S. F. and Ho,
S. Y. (2007) ProLoc: Prediction of protein subnuclear localiza-
tion using SVM with automatic selection from physicochemical
composition features. Biosystems, 90, 573-581.
[44] Shen, H. B. and Chou, K. C. (2007) Nuc-PLoc: a new
web-server for predicting protein subnuclear localization by
fusing PseAA composition and PsePSSM. Protein Engineering
Design and Selection, 20, 561-567.
[45] Jiang, X. Y., Wei, R., Zhao, Y. J. and Zhang, T. L. (2008) Using
Chou's pseudo amino acid composition based on approximate
entropy and an ensemble of AdaBoost classifiers to predict pro-
tein subnuclear location. Amino Acids, 34, 669-675.
[46] Shen, H. B. and Chou, K. C. (2008) PseAAC: A flexible web
server for generating various kinds of protein pseudo amino acid
composition. Analytical Biochemistry, 373, 386-388.
[47] Chou, K. C. (2001) Prediction of protein cellular attributes using
pseudo-amino acid composition. Proteins-Structure Function and
Genetics, 43, 246-255.
[48] Chou, K. C. (2005) Using amphiphilic pseudo amino acid com-
position to predict enzyme subfamily classes. Bioinformatics, 21,
10-19.
[49] Tanford, C. (1962) Contribution of Hydrophobic Interactions to
the Stability of the Globular Conformation of Proteins. Journal
of the American Chemical Society, 84, 4240-4247.
[50] Krigbaum, W. R. and Komoriya, A. (1979) Local interactions as
a structure determinant for protein molecules: II. Biochimica et
Biophysica Acta, 576, 204-248.
[51] Grantham, R. (1974) Amino acid difference formula to help
explain protein evolution. Science, 185, 862-864.
[52] Guo, Y. Z., Li, M., Lu, M., Wen, Z., Wang, K., Li, G. and Wu, J.
(2006) Classifying G protein-coupled receptors and nuclear re-
ceptors on the basis of protein power spectrum from fast Fourier
transform. AMINO ACIDS, 30, 397-402.
[53] Hackel, M., Hinz, H. J. and Hedwig, G. R. (1999) Partial molar
volumes of proteins: amino acid side-chain contributions derived
from the partial molar volumes of some tripeptides over the
temperature range 10-90 degrees C. BIOPHYSICAL CHEMISTRY,
82, 35-50.
[54] Guo, Y. Z., Li, M. L., Wang, K. L., Wen, Z. N., Lu, M. C., Liu,
L. X. and Jiang, L. (2006) Fast fourier transform-based support
vector machine for prediction of G-protein coupled receptor
subfamilies. (Vol 37, pg 759, 2005). ACTA BIOCHIMICA ET
BIOPHYSICA SINICA, 38, 456-456.
[55] Guo, Y.Z., Yu, L.Z., Wen, Z.N. and Li, M.L. (2008) Using sup-

56 R. Q. Xiao et al. / J. Biomedical Science and Engineering 2 (2009) 51-56
SciRes Copyright © 2009 JBiSE
port vector machine combined with auto covariance to predict
protein-protein interactions from protein sequences. Nucleic
Acids Research, 36, 3025-3030.
[56] Lin, Z. and Pan, X. M. (2001) Accurate prediction of protein
secondary structural content. Journal Of Protein Chemistry, 20,
217-220.
[57] Zhang, C. T., Lin, Z. S., Zhang, Z. D. and Yan, M. (1998) Pre-
diction of the helix/strand content of globular proteins based on
their primary sequences. Protein Engineering, 11, 971-979.
[58] Zhang, Z. D., Sun, Z. R. and Zhang, C. T. (2001) A new ap-
proach to predict the helix/strand content of globular proteins.
Journal Of Theoretical Biology, 208, 65-78.
[59] Kedarisetti, K. D., Kurgan, L. and Dick, S. (2006) Classifier
ensembles for protein structural class prediction with varying
homology. Biochemical And Biophysical Research Communica-
tions, 348, 981-988.
[60] Kurgan, L. A. and Homaeian, L. (2006) Prediction of structural
classes for protein sequences and domains-Impact of prediction
algorithms, sequence representation and homology, and test
procedures on accuracy. Pattern Recognition, 39, 2323-2343.
[61] Li, X. and Pan, X. M. (2001) New method for accurate predic-
tion of solvent accessibility from protein sequence. Pro-
teins-Structure Function And Genetics, 42, 1-5.
[62] Kurgan, L. and Chen, K. (2007) Prediction of protein structural
class for the twilight zone sequences. Biochemical And Bio-
physical Research Communications, 357, 453-460.
[63] Guo, Y. Z., Li, M. L., Lu, M. C., Wen, Z.N. and Huang, Z. T.
(2006) Predicting G-protein coupled receptors-G-protein cou-
pling specificity based on autocross-covariance transform. Pro-
teins, 65, 55-60.
[64] Ben-Gal, I., Shani, A., Gohr, A., Grau, J., S, A., Shmilovici, A.,
Posch, S. and Grosse, I. (2005) Identification of transcription
factor binding sites with variable-order Bayesian networks. Bio-
informatics, 21, 2657-2666.
[65] Xie, D., Li, A., Wang, M. H., Fan, Z. W. and Feng, H. Q. (2005)
LOCSVMPSI: a web server for subcellular localization of eu-
karyotic proteins using SVM and profile of PSI-BLAST. Nucleic
Acids Research, 33, 105-110.
[66] Schaffer, A. A., Aravind, L., Madden, T. L., Shavirin, S.,
Spouge, J. L., Wolf, Y. I., Koonin, E. V. and Altschul, S. F.
(2001) Improving the accuracy of PSI-BLAST protein database
searches with composition-based statistics and other refinements.
Nucleic Acids Research, 29, 2994-3005.
[67] Chou, K. C. and Zhang, C. T. (1995) PREDICTION OF PRO-
TEIN STRUCTURAL CLASSES. Critical Reviews in Bio-
chemistry and Molecular Biology, 30, 275-349.
[68] Chen, Y. L. and Li, Q. Z. (2007) Prediction of the subcellular
location of apoptosis proteins. Journal of Theoretical Biology,
245, 775-783.
[69] Chen, Y. L. and Li, Q. Z. (2007) Prediction of apoptosis protein
subcellular location using improved hybrid approach and
pseudo-amino acid composition. Journal of Theoretical Biology,
248, 377-381.
[70] Zhou, X. B., Chen, C., Li, Z. C. and Zou, X. Y. (2007) Using
Chou's amphiphilic pseudo-amino acid composition and support
vector machine for prediction of enzyme subfamily classes.
Journal of Theoretical Biology, 248, 546-551.
[71] Chen, C., Chen, L. X., Zou, X. Y. and Cai, P. X. (2008) Predict-
ing protein structural class based on multi-features fusion. Jour-
nal of Theoretical Biology, 253, 388-392.
[72] Chen, K., Kurgan, L. A. and Ruan, J. S. (2008) Prediction of
protein structural class using novel evolutionary collocation-
based sequence representation. Journal of Computational Chem-
istry, 29, 1596-1604.
[73] Du, P. F. and Li, Y. D. (2008) Prediction of C-to-U RNA editing
sites in plant mitochondria using both biochemical and evolu-
tionary information. Journal of Theoretical Biology, 253,
579-586.
[74] Jiang, X. Y., Wei, R., Zhang, T. L. and Gu, Q. (2008) Using the
concept of Chou's Pseudo Amino Acid composition to predict
apoptosis proteins subcellular location: An approach by ap-
proximate entropy. Protein and Peptide Letters, 15, 392-396.
[75] Jin, Y. H., Niu, B., Feng, K. Y., Lu, W. C., Cai, Y. D. and Li, G.
Z. (2008) Predicting subcellular localization with AdaBoost
Learner. Protein and Peptide Letters, 15, 286-289.
[76] Li, F. M. and Li, Q. Z. (2008) Protein Subcellular Location
Using Chou’s Pseudo Amino Acid Composition and Improved
Hybrid Approach. Protein and Peptide Letters, 15, 612-616.
[77] Lin, H. (2008) The modified Mahalanobis Discriminant for
predicting outer membrane proteins by using Chou's pseudo amino
acid composition. Journal of Theoretical Biology, 252, 350-356.
[78] Lin, H., Ding, H., Guo, F. B., Zhang, A. Y. and Huang, J. (2008)
Predicting Subcellular Localization of Mycobacterial Proteins by
Using Chou’s Pseudo Amino Acid Composition. Protein and
Peptide Letters, 15, 739-744.
[79] Niu, B., Jin, Y. H., Feng, K. Y., Liu, L., Lu, W. C., Cai, Y. D.
and Li, G. Z. (2008) Predicting membrane protein types with
bragging learner. Protein and Peptide Letters, 15, 590-594.
[80] Wang, T., Yang, J., Shen, H. B. and Chou, K. C. (2008) Predict-
ing membrane protein types by the LLDA algorithm. Protein and
Peptide Letters, 15, 915-921.
[81] Wu, G. and Yan, S. M. (2008) Prediction of mutations in H3N2
hemagglutinins of influenza A virus from North America based
on different datasets. Protein and Peptide Letters, 15, 144-152.
[82] Zhang, G. Y. and Fang, B. S. (2008) Predicting the cofactors of
oxidoreductases based on amino acid composition distribution
and Chou’s amphiphilic pseudo-amino acid composition. Journal
of Theoretical Biology, 253, 310-315.
[83] Cai, Y. D. and Chou, K. C. (2003) Nearest neighbour algorithm
for predicting protein subcellular location by combining func-
tional domain composition and pseudo-amino acid composition.
Biochemical and Biophysical Research Communications, 305,
407-411.
[84] Park, K. J. and Kanehisa, M. (2003) Prediction of protein sub-
cellular locations by support vector machines using composi-
tions of amino acids and amino acid pairs. Bioinformatics, 19,
1656-1663.
[85] Chou, K. C. and Cai, Y. D. (2004) Predicting subcellular local-
ization of proteins by hybridizing functional domain composi-
tion and pseudo-amino acid composition. Journal of Cellular
Biochemistry, 91, 1197-1203.
[86] Cedano, J., Aloy, P., PerezPons, J. A. and Querol, E. (1997)
Relation between amino acid composition and cellular location
of proteins. Journal of Molecular Biology, 266, 594-600.