Email updates

Keep up to date with the latest news and content from BMC Bioinformatics and BioMed Central.

Open Access Research article

Efficacy of different protein descriptors in predicting protein functional families

Serene AK Ong1, Hong Huang Lin1, Yu Zong Chen1, Ze Rong Li2 and Zhiwei Cao3*

Author Affiliations

1 Department of Pharmacy, National University of Singapore, Blk S16, Level 8, 08-14, 3 Science Drive 2, Singapore 117543, Singapore

2 College of Chemistry, Sichuan University, Chengdu, 610064, P.R. China

3 Shanghai Center for Bioinformatics Technology, 100, Qinzhou Road, Shanghai 200235 P.R. China

For all author emails, please log on.

BMC Bioinformatics 2007, 8:300  doi:10.1186/1471-2105-8-300


The electronic version of this article is the complete one and can be found online at: http://www.biomedcentral.com/1471-2105/8/300


Received:1 November 2006
Accepted:17 August 2007
Published:17 August 2007

© 2007 Ong et al; licensee BioMed Central Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

Sequence-derived structural and physicochemical descriptors have frequently been used in machine learning prediction of protein functional families, thus there is a need to comparatively evaluate the effectiveness of these descriptor-sets by using the same method and parameter optimization algorithm, and to examine whether the combined use of these descriptor-sets help to improve predictive performance. Six individual descriptor-sets and four combination-sets were evaluated in support vector machines (SVM) prediction of six protein functional families.

Results

The performance of these descriptor-sets were ranked by Matthews correlation coefficient (MCC), and categorized into two groups based on their performance. While there is no overwhelmingly favourable choice of descriptor-sets, certain trends were found. The combination-sets tend to give slightly but consistently higher MCC values and thus overall best performance such that three out of four combination-sets show slightly better performance compared to one out of six individual descriptor-sets.

Conclusion

Our study suggests that currently used descriptor-sets are generally useful for classifying proteins and the prediction performance may be enhanced by exploring combinations of descriptors.

Background

Sequence-derived structural and physicochemical descriptors have frequently been used in machine learning prediction of protein structural and functional classes [1-5], protein-protein interactions [6-9], subcellular locations [10-16], peptides containing specific properties[17,18], microarray data [19] and protein secondary structure prediction [20]. These descriptors serve to represent and distinguish proteins or peptides of different structural, functional and interaction profiles by exploring their distinguished features in compositions, correlations, and distributions of the constituent amino acids and their structural and physicochemical properties [2,8,21,22]. There is thus a need to comparatively evaluate the effectiveness of these descriptor-sets for predicting different functional problems by using the same machine learning method and parameter optimization algorithm. Moreover, it is of interest to examine whether combined use of these descriptor-sets help to improve predictive performance.

This work is intended to evaluate the effectiveness of a total of six individual descriptor-sets and four combination-sets (Table 1) in the prediction of several protein functional families by using support vector machine (SVM). Six sets of individual descriptors and three combination-sets have been separately utilized in machine learning prediction of different protein functional and structural properties, all of which have shown impressive predictive performances [22-24]. The six individual sets are amino acid compositions [23] (Set D1), dipeptide compositions [24] (Set D2), normalized Moreau-Broto autocorrelation [25,26] (Set D3), Moran autocorrelation [27] (Set D4), Geary autocorrelation [28] (Set D5), and the composition, transition and distribution of structural and physicochemical properties [2-6,8,17,29,30] (Set D6). The three combination-sets are quasi sequence order formed by weighted sums of amino acid compositions and physicochemical coupling correlations [10,11,18,31] (Set D7), pseudo amino acid composition (PseAA) formed by weighted sums of amino acid compositions and physicochemical square correlations [23,32] (Set D8), and combination of amino acid compositions and dipeptide compositions (Set D9) [24,33]. In this work, we also considered a fourth combination-set that combines descriptor-sets D1 through D8 (Set D10).

Table 1. Protein descriptors commonly used for predicting protein functional families.

The protein functional families studied here include enzyme EC2.4 [34-37], G protein-coupled receptors [38-40], transporter TC8.A [41], chlorophyll [42], lipid synthesis proteins involved in lipid synthesis [43], and rRNA-binding proteins. These six protein families were selected for testing the descriptor-sets based on their functional diversity, sample size and the range of reported family member prediction accuracies [2]. The reported prediction accuracies for these families are generally lower than those of other families [3], which are ideal for critically evaluating the effectiveness of these descriptor-sets; having a lower accuracy should enable a better differentiation of the performance of the various classes. SVM was used as the machine learning method for predicting these functional families because it is a popular method that has consistently been shown better performances than other machine learning methods [44,45]. As this work is intended as a benchmarking study of the performance of various classes of descriptors, other than automatic optimization of results that is an integral part of the SVM programs, such as sigma value scanning, no further attempt was made to optimize the prediction performance of any descriptor class or of any dataset by manually tuning the parameters. Hence, prediction results reported in this paper might differ from those of reported studies.

EC2.4 includes glycosyltransferases that catalyze the synthesis of glycoconjugates and are involved in post-translational modification of proteins (glycosylation). Increased levels of glycosyltransferases have been found in disease states and inflammation [46,47]. TC8.A consists of auxiliary transport proteins that facilitate transport across membranes, which play regulatory and structural roles [48]. GPCR represents G-protein coupled receptors that transduct signals for inducing cellular responses, and members of GPCR are of great pharmacological importance, as 50–60% of approved drugs elicit their therapeutic effect by selectively addressing members of the GPCR family [49-52]. Chlorophyll proteins are essential for harvesting solar energy in photosynthetic antenna systems [53]. Lipid synthesis proteins play central roles in such processes as metabolism, and deficiencies or altered functioning of lipid binding proteins are associated with disease states such as obesity, diabetes, atherosclerosis, hyperlipidemia and insulin resistance [54]. rRNA-binding proteins play central roles in the post-transcriptional regulation of gene expression [55,56], and their binding capabilities are mediated by certain RNA binding domains and motifs [57-60].

Results and Discussion

The statistics of the six datasets are given in Table 2. Training and prediction statistics for each of the studied descriptor-sets are given in Table 3. Independent validation datasets were used to test the prediction accuracies. Among the 5-fold cross-validation test, independent dataset test and jackknife test, the jackknife is deemed the most rigorous [61]; however, it would have taken a lot of time to use SVM to conduct the jackknife test, thus as a compromise, here we adopted the independent dataset test. The program CDHIT [62-64] was used to remove redundancy at both 90% and 70% sequence identity so to avoid bias, subsequently, the datasets are tested again with the independent evaluation sets and the statistics are given in Table 4. It should be emphasized that the performance evaluation for the studied descriptor-sets are based only on the datasets studied in this work and the conclusions from this study might not be readily extended to other datasets.

Table 2. Summary of datasets statistics, including size of training, testing and independent evaluation sets, and average sequence length.

Table 3. Dataset training statistics and prediction accuracies of six protein functional families. DS refers to descriptor set, where D1 = amino acid composition; D2 = dipeptide composition; D3 = Moreau-Broto autocorrelation; D4 = Moran autocorrelation; D5 = Geary autocorrelation; D6 = composition, transition and distribution descriptors; D7 = quasi sequence order; D8 = pseudo amino acid composition; D9 = combination of D1+D2; and D10 = combination of D1-D8. Predicted results given as TP (true positive), FN (false negative), TN (true negative), FP (false positive), Sen (sensitivity), Spec (specificity), Q (overall accuracy) and MCC (Matthews correlation coefficient).

Table 4. Dataset statistics and prediction accuracies after homologous sequences removal (HSR) at 90% and 70% identity. DS refers to descriptor set, where D1 = amino acid composition; D2 = dipeptide composition; D3 = Moreau-Broto autocorrelation; D4 = Moran autocorrelation; D5 = Geary autocorrelation; D6 = composition, transition and distribution descriptors; D7 = quasi sequence order; D8 = pseudo amino acid composition; D9 = combination of D1+D2; and D10 = combination of D1-D8. Predicted results given as TP (true positive), FN (false negative), TN (true negative), FP (false positive), Sen (sensitivity), Spec (specificity), Q (overall accuracy) and MCC (Matthews correlation coefficient).

The performance of the ten descriptor-sets were ranked by the Matthews correlation coefficient (MCC) values of the respective SVM prediction of the six functional families, which are given in Table 5. The computed MCC scores for these descriptor-sets are in the range of 0.64~0.97 for all protein families studied. Accordingly, the performance of these descriptor-sets is categorized into two groups based on their MCC values: 'Exceptional' (>0.85) and 'Good' (≤0.85). Moreover, these descriptor-sets are aligned in the order of their MCC values with "=" being of equal values and ">" indicating that one is better than the other. It is noted that, as the differences of many of these MCC values are rather small, such alignment is likely superficial to some extent and may not best reflect the real ranking of performance. Overall, the performances of these descriptor-sets are not significantly different, there is no overwhelmingly preferred descriptor-set, and SVM prediction performance appears to be highly dependent on the dataset.

Table 5. Descriptor sets ranked and grouped by MCC (Matthews correlation coefficient), before and after removal of homologous sequences at 90% and 70% identity, respectively.

As shown in Table 3 and Table 4, for many of the studied datasets, the differences in prediction accuracies and MCC values between different descriptor-sets are small. In particular, for GPCR and rRNA binding proteins, the results of almost all descriptor-sets are in the 'Exceptional' category. Examining the range of MCC values of the descriptor-sets for each of the studied protein families (after removal of 70% homologous sequences), the differences between the largest and smallest MCC values are, in order of increasing magnitude: 0.10, 0.12, 0.14, 0.16, 0.21 and 0.21 for rRNA binding proteins, GPCR, TC8.A, lipid synthesis proteins, chlorophyll proteins and EC.2.4 families respectively. Given that a difference of 0.10 and 0.20 in MCC values translates to an approximate 4% and 7% difference in overall prediction accuracy, this separation is not large indeed.

Though the dataset is a more important determinant of prediction performance than the choice of descriptor class, a few general trends could be observed. Three out of four of the combination-sets tend to exhibit slightly but consistently higher MCC values for the protein families studied in this work. These sets are Sets D8, D9 and D10. In contrast, only one out of six individual sets, Set D6, tend to exhibit slightly but consistently higher MCC values for the protein families studied in this work. Therefore, statistically speaking, it appears that the use of combination-sets tend to give slightly better prediction performance than the use of individual-sets.

When each class was examined individually in this study, we find that the combination of amino acid composition and dipeptide composition (Set D9) tends to give consistently better results than that of the individual descriptor-sets (Set D1 and Set D2). It has been reported that one drawback of amino acid composition descriptors is that the same amino acid composition may correspond to diverse sequences as sequence order is lost [24,33]. This sequence order information can be partially covered by considering dipeptide composition (Set D2). On the other hand, dipeptide composition lacks information concerning the fraction of the individual residue in the sequence, thus, a combination-set is expected to give better prediction results [24,33,65,66].

Using all descriptor-sets (Set D10) generally, but not always, gives the best result, which is consistent with the findings on the use of molecular descriptors for predicting compounds of specific properties. [67,68] For instance, Xue et al. found that feature selection methods are capable of reducing the noise generated by the use of overlapping and redundant molecular descriptors, and in some cases, improving the accuracy of SVM classification of pharmacokinetic behaviour of chemical agents [69]. In our study, for example, the three autocorrelation descriptor-sets (Sets D3, D4 and D5) all utilize the same physicochemical properties, only differing in the correlation algorithm. The use of all available descriptors likely results in the inclusion of partially redundant information, some of which may to some extent become noise that interferes with the prediction results or obscures relevant information. Based on the results of previous studies [69], it is possible that feature selection methods may be applied for selecting the optimal set of descriptors to improve prediction accuracy as well as computing efficiency for predicting protein functional families.

Conclusion

The effectiveness of ten protein descriptor-sets in six protein functional family prediction using SVM was evaluated. Corroborating with previous work done on chemical descriptors [67,68,70-76] and protein descriptors [4,21,30,32,35,43,77,78], we found that the descriptor-sets evaluated in this paper, which comprise some of the commonly used descriptors, generally return good results and do not differ significantly. In particular, the use of combination descriptor-sets tends to give slightly better prediction performance than the use of individual descriptor-sets. While there seems to be no preferred descriptor-set that could be utilized for all datasets as prediction results is highly dependent on datasets, the performance of protein classification may be enhanced by selection of optimal combinations of descriptors using established feature-selection methods [79,80]. Incorporation of appropriate sets of physicochemical properties not covered by some of the existing descriptor-sets may also help improving the prediction performance.

Methods

Datasets

The datasets were obtained from SwissProt [81], except for TC8.A, which was downloaded from Transport Classification Database (TCDB) [41]. These datasets were chosen for their functional diversity, sample size and the range of reported family member prediction accuracies. As SVM is essentially a statistical method, the datasets cannot be too small; yet it would also be convenient for the purposes of this study if they were not too large as to be unwieldy computationally. These downloaded datasets were used to construct the positive dataset for the corresponding SVM classification system. A negative dataset, representing non-class members, was generated by a well-established procedure [2,3,21,30] such that all proteins was grouped into domain families [82] in the PFAM database, and the representative proteins of these families unrelated to the protein family being studied were chosen as negative samples.

These proteins, positive and negative, were further divided into separate training, testing and independent evaluation sets by the following procedure: First, proteins were converted into descriptor vectors and then clustered using hierarchical clustering into groups in the structural and physicochemical feature space [83], where more homologous sequences will have shorter distances between them, and the largest separation between clusters was set to a ceiling of 20. One representative protein was randomly selected from each group to form a training set that is sufficiently diverse and broadly distributed in the feature space. Another protein within the group was randomly selected to form the testing set. The selected proteins from each group were further checked to ensure that they are distinguished from the proteins in other groups. The remaining proteins were then designated as the independent evaluation set, also checked to be at a reasonable level of diversity. Fragments, defined as smaller than 60 residues, were discarded. This selection process ensures that the training, testing and evaluation sets constructed are sufficiently diverse and broadly distributed in the feature space. Though an analysis of the 'similar' proteins in each cluster showed that the majority of the proteins in a cluster are quite non-homologous, the program CDHIT (Cluster Database at High Identity with Tolerance) [62-64] was further used after the SVM model was trained to remove redundancy at both 90% and 70% sequence identity, so as to avoid bias as far as possible. CDHIT removes homologous sequences by clustering the protein dataset at some user-defined sequence identity threshold, for example 90%, and then generating a database of only the cluster representatives, thus eliminating sequences with greater than 90% identity. The statistical details are given in Tables 2 and 3.

Algorithms for generating protein descriptors

Ten sets of commonly used composition and physicochemical descriptors were generated from the protein sequence (see Table 1). These descriptors can be computed via the PROFEAT server [22].

Amino acid composition (Set D1) is defined as the fraction of each amino acid type in a sequence

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M1','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M1">View MathML</a>

(1)

where r = 1, 2, ..., 20, Nr is the number of amino acid of type r, and N is the length of the sequence. Dipeptide composition (Set D2) is defined as

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M2','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M2">View MathML</a>

(2)

where r, s = 1, 2, ..., 20, Nij is the number of dipeptides composed of amino acid types r and s.

Autocorrelation descriptors are a class of topological descriptors, also known as molecular connectivity indices, describe the level of correlation between two objects (protein or peptide sequences) in terms of their specific structural or physicochemical property [84], which are defined based on the distribution of amino acid properties along the sequence [85]. Eight amino acid properties are used for deriving the autocorrelation descriptors: hydrophobicity scale [86]; average flexibility index [87]; polarizability parameter [88]; free energy of amino acid solution in water [88]; residue accessible surface areas [89]; amino acid residue volumes [90]; steric parameters [91]; and relative mutability [92].

These autocorrelation properties are normalized and standardized such that

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M3','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M3">View MathML</a>

(3)

where <a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M4','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M4">View MathML</a> is the average value of a particular property of the 20 amino acids. <a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M4','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M4">View MathML</a> and σ are given by

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M5','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M5">View MathML</a>

(4)

and

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M6','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M6">View MathML</a>

(5)

Moreau-Broto autocorrelation descriptors (Set D3) [84,93] are defined as

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M7','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M7">View MathML</a>

(6)

where d = 1, 2, ..., 30 is the lag of the autocorrelation, and Pi and Pi+d are the properties of the amino acid at positions i and i+d respectively. After applying normalization, we get

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M8','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M8">View MathML</a>

(7)

Moran autocorrelation descriptors (Set D4) [94] are calculated as

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M9','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M9">View MathML</a>

(8)

where d, Pi and Pi+d are defined in the same way as that for Moreau-Broto autocorrelation and <a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M4','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M4">View MathML</a> is the average of the considered property P along the sequence:

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M10','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M10">View MathML</a>

(9)

Geary autocorrelation descriptors (Set D5) [95] are written as

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M11','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M11">View MathML</a>

(10)

where d, <a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M4','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M4">View MathML</a>, Pi and Pi+d are defined as above. Comparing the three autocorrelation descriptors: while Moreau-Broto autocorrelation uses the property values as the basis for measurement, Moran autocorrelation utilizes property deviations from the average values, and Geary utilizes the square-difference of property values instead of vector-products (of property values or deviations). The Moran and Geary autocorrelation descriptors measure spatial autocorrelation, which is the correlation of a variable with itself through space.

The descriptors in Set D6 comprise of the composition (C), transition (T) and distribution (D) features of seven structural or physicochemical properties along a protein or peptide sequence [5,29]. The seven physicochemical properties [2,5,29] are hydrophobicity; normalized Van der Waals volume; polarity; polarizibility; charge; secondary structures; and solvent accessibility. For each of these properties, the amino acids are divided into three groups such that those in a particular group are regarded to have approximately the same property. For instance, residues can be divided into hydrophobic (CVLIMFW), neutral (GASTPHY), and polar (RKEDQN) groups. C is defined as the number of residues with that particular property divided by the total number of residues in a protein sequence. T characterizes the percent frequency with which residues with a particular property is followed by residues of a different property. D measures the chain length within which the first, 25%, 50%, 75% and 100% of the amino acids with a particular property are located respectively. There are 21 elements representing these three descriptors: 3 for C, 3 for T and 15 for D, and the protein feature vector is constructed by sequentially combining the 21 elements for all of these properties and the 20 residues, resulting in a total of 188 dimensions.

The quasi-sequence order descriptors (Set D7) [96] are derived from both the Schneider-Wrede physicochemical distance matrix [10,18,97] and the Grantham chemical distance matrix [31], between each pair of the 20 amino acids. The physicochemical properties computed include hydrophobicity, hydrophilicity, polarity, and side-chain volume. Similar to the descriptors in Set D6, sequence order descriptors can also be used for representing amino acid distribution patterns of a specific physicochemical property along a protein or peptide sequence [18,31]. For a protein chain of N amino acid residues R1R2...RN, the sequence order effect can be approximately reflected through a set of sequence order coupling numbers

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M12','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M12">View MathML</a>

(11)

where τd is the dth rank sequence order coupling number (d = 1, 2, ..., 30) that reflects the coupling mode between all of the most contiguous residues along a protein sequence, and di,i+d is the distance between the two amino acids at position i and i+d. For each amino acid type, the type 1 quasi sequence order descriptor can be defined as

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M13','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M13">View MathML</a>

(12)

where r = 1, 2, ..., 20, fr is the normalized occurrence of amino acid type i and w is a weighting factor (w = 0.1). The type 2 quasi sequence order is defined as

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M14','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M14">View MathML</a>

(13)

where d = 21, 22, ..., 50. The combination of these two equations gives us a vector that describes a protein: the first 20 components reflect the effect of the amino acid composition, while the components from 21 to 50 reflect the effect of sequence order.

Similar to the quasi-sequence order descriptor, the pseudo amino acid descriptor (Set D8) is made up of a 50-dimensional vector in which the first 20 components reflect the effect of the amino acid composition and the remaining 30 components reflect the effect of sequence order, only now, the coupling number τd is now replaced by the sequence order correlation factor θλ [32]. The set of sequence order correlated factors is defined as follows:

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M15','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M15">View MathML</a>

(14)

where θλ is the first-tier correlation factor that reflects the sequence order correlation between all of the λ-most contiguous resides along a protein chain (λ = 1,...30) and N is the number of amino acid residues. Θ(Ri, Rj) is the correlation factor and is given by

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M16','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M16">View MathML</a>

(15)

where H1(Ri), H2(Ri) and M(Ri) are the hydrophobicity [98], hydrophilicity [99] and side-chain mass of amino acid Ri, respectively. Before being substituted in the above equation, the various physicochemical properties P(i) are subjected to a standard conversion,

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M17','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M17">View MathML</a>

(16)

This sequence order correlation definition [Eqs. (14), (15)] introduce more correlation factors of physicochemical effects as compared to the coupling number [Eq. (11)], and has shown to be an improvement on the way sequence order effect information is represented [32,35,100]. Thus, for each amino acid type, the first part of the vector is defined as

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M18','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M18">View MathML</a>

(17)

where r = 1, 2, ..., 20, fr is the normalized occurrence of amino acid type i and w is a weighting factor (w = 0.1), and the second part is defined as

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M19','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M19">View MathML</a>

(18)

Support Vector Machines (SVM)

As the SVM algorithms have been extensively described in the literature [2,3,101], only a brief description is given here. In the case of a linear SVM, a hyperplane that separates two different classes of feature vectors with a maximum margin is constructed. One class represents positive samples, for example EC2.4 proteins, and the other the negative samples. This hyperplane is constructed by finding a vector w and a parameter b that minimizes ||w||2 that satisfies the following conditions: w·xi + b ≥ 1, for yi = 1 (positive class) and w·xi + b ≤ -1, for yi = -1 (negative class). Here xi is a feature vector, yi is the group index, w is a vector normal to the hyperplane, <a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M20','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M20">View MathML</a> is the perpendicular distance from the hyperplane to the origin, and ||w||2 is the Euclidean norm of w. In the case of a nonlinear SVM, feature vectors are projected into a high dimensional feature space by using a kernel function such as <a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M21','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M21">View MathML</a>. The linear SVM procedure is then applied to the feature vectors in this feature space. After the determination of w and b, a given vector x can be classified by using sign [(w.x) + b], a positive or negative value indicating that the vector x belongs to the positive or negative class respectively.

As a discriminative method, the performance of SVM classification can be accessed by measuring the true positive TP (correctly predicted positive samples), false negative FN (positive samples incorrectly predicted as negative), true negative TN (correctly predicted negative samples), and false positive FP (negative samples incorrectly predicted as positive) [4,102,103]. As the numbers of positive and negative samples are imbalanced, the positive prediction accuracy or sensitivity Qp = TP/(TP+FN) and negative prediction accuracy or specificity Qn = TN/(TN+FP) [101] are also introduced. The overall accuracy is defined as Q = (TP+TN)/(TP+FN+TN+FP). However, in some cases, Q, Qp, and Qnare insufficient to provide a complete assessment of the performance of a discriminative method [102,104]. Thus the Matthews correlation coefficient (MCC) was used in this work to evaluate the randomness of the prediction:

<a onClick="popup('http://www.biomedcentral.com/1471-2105/8/300/mathml/M22','MathML',630,470);return false;" target="_blank" href="http://www.biomedcentral.com/1471-2105/8/300/mathml/M22">View MathML</a>

(19)

where MCC ∈ [-1,1], with a negative value indicating disagreement of the prediction and a positive value indicating agreement. A zero value means the prediction is completely random. The MCC utilizes all four basic elements of the accuracy and it provides a better summary of the prediction performance than the overall accuracy.

Authors' contributions

SAK generated the datasets, carried out the calculations and drafted the manuscript, HH generated the datasets and participated in the design of the study, ZR updated the descriptor generation program to calculate PseAA descriptors (ZR wrote the original descriptor generation program, introduced in previous works), YZ conceived of the study and corrected the manuscript, and YZ and ZW oversaw the design and coordination of this work and provided invaluable advice. All authors read and approved the final manuscript.

References

  1. Karchin R, Karplus K, Haussler D: Classifying G-protein coupled receptors with support vector machines.

    Bioinformatics 2002, 18:147-159. PubMed Abstract | Publisher Full Text OpenURL

  2. Cai CZ, Han LY, Ji ZL, Chen X, Chen YZ: SVM-Prot: web-based support vector machine software for functional classification of a protein from its primary sequence.

    Nuclei Acid Res 2003, 31:3692-3697. Publisher Full Text OpenURL

  3. Cai CZ, Han LY, Ji ZL, Chen YZ: Enzyme family classification by support vector machines.

    Proteins 2004, 55:66-76. PubMed Abstract | Publisher Full Text OpenURL

  4. Han LY, Cai CZ, Lo SL, Chung MC, Chen YZ: Prediction of RNA-binding proteins from primary sequence by a support vector machine approach .

    RNA 2004, 10:355-368. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  5. Dubchak I, Muchnick I, Mayor C, Dralyuk I, Kim SH: Recognition of a protein fold in the context of the Structural Classification of Proteins (SCOP) classification.

    Proteins 1999, 35:401-407. PubMed Abstract | Publisher Full Text OpenURL

  6. Bock JR, Gough DA: Predicting protein--protein interactions from primary structure.

    Bioinformatics 2001, 17:455-460. PubMed Abstract | Publisher Full Text OpenURL

  7. Bock JR, Gough DA: Whole-proteome interaction mining .

    Bioinformatics 2003, 19:125-134. PubMed Abstract | Publisher Full Text OpenURL

  8. Lo SL, Cai CZ, Chen YZ, Chung MC: Effect of training datasets on support vector machine prediction of protein-protein interactions.

    Proteomics 2005, 5:876-884. PubMed Abstract | Publisher Full Text OpenURL

  9. Chou KC, Cai YD: Predicting protein-protein interactions from sequences in a hybridization space.

    J Proteome Res 2006, 5:316-322. PubMed Abstract | Publisher Full Text OpenURL

  10. Chou KC: Prediction of protein subcellular locations by incorporating quasi-sequence-order effect.

    Biochem Biophys Res Commun 2000, 278:477-483. PubMed Abstract | Publisher Full Text OpenURL

  11. Chou KC, Cai YD: Prediction of protein subcellular locations by GO-FunD-PseAA predictor.

    Biochem Biophys Res Commun 2004, 320:1236-1239. PubMed Abstract | Publisher Full Text OpenURL

  12. Chou KC, Shen HB: Hum-PLoc: A novel ensemble classifier for predicting human protein subcellular localization.

    Biochem Biophys Res Commun 2006, 347:150-157. PubMed Abstract | Publisher Full Text OpenURL

  13. Chou KC, Shen HB: Large-scale plant protein subcellular location prediction.

    J Cell Biochem 2006, 100(3):665-678. Publisher Full Text OpenURL

  14. Bhasin M, Garg A, Raghava GP: PSLpred: prediction of subcellular localization of bacterial proteins.

    Bioinformatics 2005, 21(10):2522-2524. PubMed Abstract | Publisher Full Text OpenURL

  15. Guo J, Lin Y, Liu XJ: GNBSL: a new integrative system to predict the subcellular location for Gram-negative bacteria proteins.

    Proteomics 2006, 6(19):5099-5105. PubMed Abstract | Publisher Full Text OpenURL

  16. Guo J, Lin Y: TSSub: eukaryotic protein subcellular localization by extracting features from profiles.

    Bioinformatics 2006, 22(14):1784-1785. PubMed Abstract | Publisher Full Text OpenURL

  17. Cui J, Han LY, Lin HH, Zhang HL, Tang ZQ, Zheng CJ, Cao ZW, Chen YZ: Prediction of MHC-binding peptides of flexible lengths from sequence-derived structural and physicochemical properties.

    Mol Immunol 2007, 44:866-877. PubMed Abstract | Publisher Full Text OpenURL

  18. Schneider G, Wrede P: The rational design of amino acid sequences by artificial neural networks and simulated molecular evolution: de novo design of an idealized leader peptidase cleavage site.

    Biophys J 1994, 66:355-344. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  19. Brown MP, Grundy WN, Lin D, Cristianini N, Sugnet CW, Furey TS, Ares MJ Jr, Haussler D: Knowledge-based analysis of microarray gene expression data by using support vector machines.

    Proc Natl Acad Sci USA 2000, 97(1):262-267. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  20. Ward JJ, McGuffin LJ, Buxton BF, Jones DT: Secondary structure prediction with support vector machines .

    Bioinformatics 2003, 19(13):1650–1655. Publisher Full Text OpenURL

  21. Han LY, Cai CZ, Ji ZL, Cao ZW, Cui J, Chen YZ: Predicting functional family of novel enzymes irrespective of sequence similarity: a statistical learning approach.

    Nuclei Acid Res 2004, 32:6437-6444. Publisher Full Text OpenURL

  22. Li ZR, Lin HH, Han LY, Jiang L, Chen X, Chen YZ: PROFEAT: A web server for computing structural and physicochemical features of proteins and peptides from amino acid sequence.

    Nuclei Acid Res 2006, 34(Web Server issue):W32-37. Publisher Full Text OpenURL

  23. Chou KC, Cai YD: Prediction of membrane protein types by incorporating amphipathic effects.

    J Chem Inf Model 2005, 45(2 ):407-413. PubMed Abstract | Publisher Full Text OpenURL

  24. Gao QB, Wang ZZ, Yan C, Du YH: Prediction of protein subcellular location using a combined feature of sequence.

    FEBS Lett 2005, 579(16):3444-3448. PubMed Abstract | Publisher Full Text OpenURL

  25. Feng ZP, Zhang CT: Prediction of membrane protein types based on the hydrophobic index of amino acids.

    J Protein Chem 2000, 19:262-275. Publisher Full Text OpenURL

  26. Lin Z, Pan XM: Accurate prediction of protein secondary structural content.

    J Protein Chem 2001, 20:217-220. PubMed Abstract | Publisher Full Text OpenURL

  27. Horne DS: Prediction of protein helix content from an autocorrelation analysis of sequence hydrophobicities.

    Biopolymers 1988, 27:451-477. PubMed Abstract | Publisher Full Text OpenURL

  28. Sokal RR, Thomson BA: Population structure inferred by local spatial autocorrelation: an example from an Amerindian tribal population.

    Am J Phys Anthropol 2006, 129:121-131. PubMed Abstract | Publisher Full Text OpenURL

  29. Dubchak I, I M, Holbrook SR, Kim SH: Prediction of protein folding class using global description of amino acid sequence.

    Proc Natl Acad Sci USA 1995, 92:8700-8704. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  30. Lin HH, Han LY, Cai CZ, Ji ZL, Chen YZ: Prediction of transporter family from protein sequence by support vector machine approach.

    Proteins 2006, 62(1):218-231. PubMed Abstract | Publisher Full Text OpenURL

  31. Grantham R: Amino acid difference formula to help explain protein evolution.

    Science 1974, 185:862-864. PubMed Abstract | Publisher Full Text OpenURL

  32. Chou KC: Prediction of protein cellular attributes using pseudo amino acid composition.

    Proteins: Structure Function and Genetics 2001, 43:246-255. Publisher Full Text OpenURL

  33. Bhasin M, Raghava GP: Classification of nuclear receptors based on amino acid composition and dipeptide composition.

    J Biol Chem 2004, 279:23262-23266. PubMed Abstract | Publisher Full Text OpenURL

  34. NC-IUBMB: Enzyme Nomenclature. San Diego, California , Academic Press; 1992.

  35. Chou KC: Using amphiphilic pseudo amino acid composition to predict enzyme subfamily classes.

    Bioinformatics 2005, 21:10-19. PubMed Abstract | Publisher Full Text OpenURL

  36. Chou KC, Cai YD: Predicting enzyme family class in a hybridization space.

    Protein Sci 2004, 13:2857-2863. PubMed Abstract | Publisher Full Text OpenURL

  37. Chou KC, Elrod DW: Prediction of enzyme family classes.

    J Proteome Res 2003, 2:183-190. PubMed Abstract | Publisher Full Text OpenURL

  38. Chou KC: Prediction of G-protein-coupled receptor classes.

    J Proteome Res 2005, 4:1413-1418. PubMed Abstract | Publisher Full Text OpenURL

  39. Chou KC, Elrod DW: Bioinformatical analysis of G-protein-coupled receptors.

    J Proteome Res 2002, 1:429-433. PubMed Abstract | Publisher Full Text OpenURL

  40. Bhasin M, Raghava GP: GPCRpred: an SVM-based method for prediction of families and subfamilies of G-protein coupled receptors.

    Nuclei Acid Res 2004, 32(Web Server issue):W383-389. Publisher Full Text OpenURL

  41. Saier MHJ, Tran CV, Barabote RD: TCDB: the Transporter Classification Database for membrane transport protein analyses and information. In Nuclei Acid Res. Volume 34. Saier Lab Bioinformatics Group; 2006::D181-D186. Publisher Full Text OpenURL

  42. Suzuki JY, Bollivar DW, Bauer CE: Genetic analysis of chlorophyll biosynthesis.

    Annu Rev Genet 1997, 31:61-89. PubMed Abstract | Publisher Full Text OpenURL

  43. Lin HH, Han LY, Zhang HL, Zheng CJ, Xie B, Chen YZ: Prediction of the functional class of lipid binding proteins from sequence-derived properties irrespective of sequence similarity.

    J Lipid Res 2006, 47:824-831. PubMed Abstract | Publisher Full Text OpenURL

  44. Brown MP, Grundy WN, Lin D, Cristianini N, Sugnet CW, Furey TS, Ares MJ, Haussler D: Knowledge-based analysis of microarray gene expression data by using support vector machines.

    Proc Natl Acad Sci USA 2000, 97(1):262-267. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  45. Burbidge R, Trotter M, Buxton B, Holden S: Drug design by machine learning: support vector machines for pharmaceutical data analysis.

    Comput Chem 2001, 26(1):5-14. PubMed Abstract | Publisher Full Text OpenURL

  46. Baenzigner JU: Protein-specific glycosyltransferase: how and why they do it!

    FASEB J 1994, 8(13):1019-1025. PubMed Abstract | Publisher Full Text OpenURL

  47. Kapitonov D, Yu RK: Conserved domains of glycosyltransferase.

    Glycobiology 1999, 9:961-978. PubMed Abstract | Publisher Full Text OpenURL

  48. Busch W, Saier MHJ: The Transporter Classification (TC) system .

    Crit Rev Biochem Mol Biol 2002, 37(5):287-337. PubMed Abstract | Publisher Full Text OpenURL

  49. Drews J: Genomic sciences and the medicine of tomorrow.

    Nat Biotechnol 1996, 14(11):1516-1518. PubMed Abstract | Publisher Full Text OpenURL

  50. Gudermann TB, Nurnberg B, Schultz G: Receptors and G proteins as primary components of transmembrane signal transduction. Part 1. G-protein-coupled receptors: structure and function.

    J Mol Med 1995, 73(2):51-63. PubMed Abstract | Publisher Full Text OpenURL

  51. Muller G: Towards 3D structures of G protein-coupled receptors: a multidisciplinary approach.

    Curr Med Chem 2000, 7(9):861-888. PubMed Abstract | Publisher Full Text OpenURL

  52. Paulson JC, Colley KJ: Glycosyltransferase.

    J Biol Chem 1989, 264(30):17645-17618. PubMed Abstract | Publisher Full Text OpenURL

  53. Beale SI, Weinstein JD: Biochemistry and regulation of photosynthetic pigment formation in plants and algae. In Biosynthesis of Tetrapyrroles. Edited by Jordan PM. Amsterdam , Elsevier; 1991:155-235. OpenURL

  54. Glatz JF, Luiken JJ, van Bilsen M, van der Vusse GJ: Cellular lipid binding proteins as facilitators and regulators of lipid metabolism.

    Mol Cell Biochem 2002, 239:3-7. PubMed Abstract | Publisher Full Text OpenURL

  55. Burd CG, Dreyfuss G: Conserved structures and diversity of functions of RNA-binding proteins .

    Science 1994, 265:615-621. PubMed Abstract | Publisher Full Text OpenURL

  56. Kiledjian M, Burd CG, Portman DS, Gorlach M, Dreyfuss G: Structure and function of hnRNP proteins. In RNA-Protein Interactions: Frontiers in Molecular Biology. Edited by Nagai K, Mattaj IW. Oxford , IRL Press; 1994:127-149. OpenURL

  57. Draper DE: Themes in RNA-protein recognition.

    J Mol Biol 1999, 293:255-270. PubMed Abstract | Publisher Full Text OpenURL

  58. Fierro-Monti I, Mathews MB: Proteins binding to duplexed RNA: one motif, multiple functions.

    Trends Biochem Sci 2000, 25:241-246. PubMed Abstract | Publisher Full Text OpenURL

  59. Perculis BA: RNA-binding proteins: If it looks like a sn(o)RNA.

    Curr Biol 2000, 10:R916-R918. PubMed Abstract | Publisher Full Text OpenURL

  60. Perez-Canadillas JM, Varani G: Recent advances in RNA-protein recognition.

    Curr Opin Struct Biol 2001, 11:53-58. PubMed Abstract | Publisher Full Text OpenURL

  61. Chou KC, Zhang CT: Prediction of protein structural classes.

    Crit Rev Biochem Mol Biol 1995, 30(4):275-349. PubMed Abstract OpenURL

  62. Li WZ, Godzik A: Cd-hit: a fast program for clustering and comparing large sets of proteins or nucleotide sequences.

    Bioinformatics 2006, 22:1658-1659. PubMed Abstract | Publisher Full Text OpenURL

  63. Li WZ, Jaroszewksi L, Godzik A: Clustering of highly homologous sequences to reduce the size of large protein database.

    Bioinformatics 2001, 17:282-283. PubMed Abstract | Publisher Full Text OpenURL

  64. Li WZ, Jaroszewksi L, Godzik A: Tolerating some redundancy significantly speeds up clustering of large protein databases.

    Bioinformatics 2002, 18:77-82. PubMed Abstract | Publisher Full Text OpenURL

  65. Garg A, Bhasin M, Raghava GP: Support vector machine-based method for subcellular localization of human proteins using amino acid compositions, their order, and similarity search.

    J Biol Chem 2005, 280(15):14427014432. Publisher Full Text OpenURL

  66. Bhasin M, Raghava GP: ESLpred: SVM-based method for subcellular localization of eukaryotic proteins using dipeptide composition and PSI-BLAST.

    Nuclei Acid Res 2004, 32(Web Server issue):414-419. Publisher Full Text OpenURL

  67. Xue L, Bajorath J: Molecular descriptors in chemoinformatics, computational combinatorial chemistry, and virtual screening.

    Comb Chem High Throughput Screen 2000, 3(5):363-372. PubMed Abstract | Publisher Full Text OpenURL

  68. Xue L, Godden JW, Bajorath J: Identification of a preferred set of descriptors for compound classification based on principal component analysis.

    J Chem Inf Comput Sci 1999, 39:669-704. OpenURL

  69. Xue Y, Li ZR, Yan CW, Sun LZ, Chen X, Chen YZ: Effect of molecular descriptor feature selection in support vector machine classification of pharmacokinetic and toxicological properties of chemical agents.

    J Chem Inf Comput Sci 2004, 44(5):1630-1638. PubMed Abstract | Publisher Full Text OpenURL

  70. Brown RD, Martin YC: Use of structure-activity data to compare structure-based clustering methods and descriptors for use in compound selection.

    J Chem Inf Comput Sci 1996, 36(3):572-584. Publisher Full Text OpenURL

  71. Cramer RD, Patterson DE, Bunce JD: Comparative molecular field analysis (CoMFA): effect of shape on binding of steroids to carrier proteins.

    J Am Chem Soc 1988, 110:5959-5967. Publisher Full Text OpenURL

  72. Glen WG, Dunn WJ, Scott RD: Principal components analysis and partial least squares regression.

    Tetrahedron Comput Methodol 1989, 2:349-376. Publisher Full Text OpenURL

  73. Matter H: Selecting optimally diverse compounds from structure databases: a validation study of two-dimensional and three-dimensional molecular descriptors.

    J Med Chem 1997, 40(8):1219-1229. PubMed Abstract | Publisher Full Text OpenURL

  74. Matter H, Pötter T: Comparing 3D pharmacophore triplets and 2D fingerprints for selecting diverse compound subsets.

    J Chem Inf Comput Sci 1999, 39:1211-1225. Publisher Full Text OpenURL

  75. Patterson DEP, Cramer RD, Ferguson AM, Clark RD, Weinberger LE: Neighborhood behavior: a useful concept for validation of "molecular diversity" descriptors.

    J Med Chem 1996, 39(16):049 -3059. Publisher Full Text OpenURL

  76. Xue L, Godden JW, Bajorath J: Evaluation of descriptors and mini-fingerprints for the identification of molecules with similar activity.

    J Chem Inf Comput Sci 2000, 40(5):1227-1234. PubMed Abstract | Publisher Full Text OpenURL

  77. Lin HH, Han LY, Zhang HL, Zheng CJ, Xie B, Chen YZ: Prediction of the functional class of DNA-binding proteins from sequence derived structural and physicochemical properties.

    2006.

  78. Chen C, Zhou X, Tian Y, Zhou X, Cai P: Predicting protein structural class with pseudo-amino acid composition and support vector machine fusion network.

    Anal Biochem 2006, 357:116-121. PubMed Abstract | Publisher Full Text OpenURL

  79. Furey TS, Cristianini N, Duffy N, Bednarski DW, Schummer M, Haussler D: Support vector machines classification and validation of cancer tissue samples using microarray expression data.

    Bioinformatics 2000, 16:906-914. PubMed Abstract | Publisher Full Text OpenURL

  80. Yu H, Yang J, Wang W, Han J: Discovering compact and highly discriminative features or feature combinations of drug activities using support vector machines.

    Proc IEEE Comput Soc Bioinform Conf 2003, (2):220-228. PubMed Abstract OpenURL

  81. Boeckmann B, Bairoch A, Apweiler R, Blatter MC, Estreicher A, Gasteiger E, Martin MJ, Michoud K, O'Donovan C, Phan I, Pilbout S, Schneider M: The SWISS-PROT protein knowledgebase and its supplement TrEMBL in 2003.

    Nuclei Acid Res 2003, 31(1):365-370. Publisher Full Text OpenURL

  82. Bateman A, Birney E, Cerruti L, Durbin R, Etwiller L, Eddy SR, Griffiths–Jones S, Howe KL, Marshall M, Sonnhammer EL: The Pfam protein families database.

    Nuclei Acid Res 2002, 31(1):276-280. Publisher Full Text OpenURL

  83. Heyer LJ, Kruglyak S, Yooseph S: Exploring expression data: Identification and analysis of coexpressed genes.

    Genome Res 1999, 9(11):1106-1115. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  84. Broto P, Moreau G, Vandicke C: Molecular structures: perception, autocorrelation descriptor and SAR studies.

    Eur J Med Chem 1984, 19:71-78. OpenURL

  85. Kawashima S, Kanehisa M: AAindex: amino acid index database.

    Nuclei Acid Res 2000, 28:374. Publisher Full Text OpenURL

  86. Cid H, Bunster M, Canales M, Gazitua F: Hydrophobicity and structural classes in proteins.

    Protein Eng 1992, 5:373-375. PubMed Abstract | Publisher Full Text OpenURL

  87. Bhaskaran R, Ponnuswammy PK: Positional flexibilities of amino acid residues in globular proteins.

    Int J Pept Protein Res 1988, 32:242-255. OpenURL

  88. Charton M, Charton BI: The structural dependence of amino acid hydrophobicity parameters.

    J Theor Biol 1982, 99:629-644. PubMed Abstract | Publisher Full Text OpenURL

  89. Chothia C: The nature of the accessible and buried surfaces in proteins.

    J Mol Biol 1976, 15:1-12. Publisher Full Text OpenURL

  90. Bigelow CC: On the average hydrophobicity of proteins and the relation between it and protein structure.

    J Theor Biol 1967, 16:187-211. PubMed Abstract | Publisher Full Text OpenURL

  91. Charton M: Protein folding and the genetic code: an alternative quantitative model.

    J Theor Biol 1981, 91:115-373. PubMed Abstract | Publisher Full Text OpenURL

  92. Dayhoff H, Calderone H: Composition of proteins.

    Atlas of Protein Sequence and Structure 1978, 5:363-373. OpenURL

  93. Moreau G, Broto P: Autocorrelation of molecular structures, application to SAR studies.

    Nour J Chim 1980, 4:757-767. OpenURL

  94. Moran PAP: Notes on continuous stochastic phenomena.

    Biometrika 1950, 37:17-23. PubMed Abstract OpenURL

  95. Geary RC: The contiguity ratio and statistical mapping.

    Incorp Statist 1954, 5:115-145. Publisher Full Text OpenURL

  96. Cai YD, Liu XJ, Xu X, Chou KC: Support vector machines for prediction of protein subcellular location by incorporating quasi-sequence-order effect.

    J Cell Biochem 2002, 84(2):343-348. PubMed Abstract | Publisher Full Text OpenURL

  97. Chou KC, Cai YD: Using functional domain composition and support vector machines for prediction of protein subcellular location.

    J Biol Chem 2002, 277:45765-45769. PubMed Abstract | Publisher Full Text OpenURL

  98. Jones DD: Amino acid properties and side-chain orientation in proteins: a cross correlation approach.

    J Theor Biol 1975, 50:167-183. PubMed Abstract | Publisher Full Text OpenURL

  99. Hopp TP, Woods KR: Prediction of protein antigenic determinants from amino acid sequences.

    Proc Natl Acad Sci USA 1981, 78:3824-3828. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  100. Feng ZP: An overview on predicting the subcellular location of a protein.

    In Silico Biol 2002, 2:291-303. PubMed Abstract | Publisher Full Text OpenURL

  101. Burges CJC: A tutorial on support vector machines for pattern recognition.

    Data Min Knowl Dis 1998, 2(2):121-167. Publisher Full Text OpenURL

  102. Baldi P, Brunak S, Chauvin Y, Andersen CA, Nielsen H: Assessing the accuracy of prediction algorithms for classification: an overview.

    Bioinformatics 2000, 16(5):412-424. PubMed Abstract | Publisher Full Text OpenURL

  103. Roulston JE: Screening with tumor markers: critical issues.

    Mol Biotechnol 2002, 20(2):153-162. PubMed Abstract | Publisher Full Text OpenURL

  104. Provost F, Fawcett T, Kohavi R: The case against accuracy estimation for comparing induction algorithms. In Proc 15th International Conf on Machine Learning. San Francisco, California , Morgan Kaufmann; 1998:445-453. OpenURL