Carboxylator: incorporating solvent-accessible surface area for identifying protein carboxylation sites

SuccSite: Incorporating Amino Acid Composition and Informative k-spaced Amino Acid Pairs to Identify Protein Succinylation Sites

Protein succinylation is a biochemical reaction in which a succinyl group (-CO-CH2-CH2-CO-) is attached to the lysine residue of a protein molecule. Lysine succinylation plays important regulatory roles in living cells. However, studies in this field are limited by the difficulty in experimentally identifying the substrate site specificity of lysine succinylation. To facilitate this process, several tools have been proposed for the computational identification of succinylated lysine sites. In this study, we developed an approach to investigate the substrate specificity of lysine succinylated sites based on amino acid composition. Using experimentally verified lysine succinylated sites collected from public resources, the significant differences in position-specific amino acid composition between succinylated and non-succinylated sites were represented using the Two Sample Logo program. These findings enabled the adoption of an effective machine learning method, support vector machine, to train a predictive model with not only the amino acid composition, but also the composition of k-spaced amino acid pairs. After the selection of the best model using a ten-fold cross-validation approach, the selected model significantly outperformed existing tools based on an independent dataset manually extracted from published research articles. Finally, the selected model was used to develop a web-based tool, SuccSite, to aid the study of protein succinylation. Two proteins were used as case studies on the website to demonstrate the effective prediction of succinylation sites. We will regularly update SuccSite by integrating more experimental datasets. SuccSite is freely accessible at http://csb.cse.yzu.edu.tw/SuccSite/.

Rapid classification of group B Streptococcus serotypes based on matrix-assisted laser desorption ionization-time of flight mass spectrometry and machine learning techniques

Background

Group B streptococcus (GBS) is an important pathogen that is responsible for invasive infections, including sepsis and meningitis. GBS serotyping is an essential means for the investigation of possible infection outbreaks and can identify possible sources of infection. Although it is possible to determine GBS serotypes by either immuno-serotyping or geno-serotyping, both traditional methods are time-consuming and labor-intensive. In recent years, the matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) has been reported as an effective tool for the determination of GBS serotypes in a more rapid and accurate manner. Thus, this work aims to investigate GBS serotypes by incorporating machine learning techniques with MALDI-TOF MS to carry out the identification.

Results

In this study, a total of 787 GBS isolates, obtained from three research and teaching hospitals, were analyzed by MALDI-TOF MS, and the serotype of the GBS was determined by a geno-serotyping experiment. The peaks of mass-to-charge ratios were regarded as the attributes to characterize the various serotypes of GBS. Machine learning algorithms, such as support vector machine (SVM) and random forest (RF), were then used to construct predictive models for the five different serotypes (Types Ia, Ib, III, V, and VI). After optimization of feature selection and model generation based on training datasets, the accuracies of the selected models attained 54.9–87.1% for various serotypes based on independent testing data. Specifically, for the major serotypes, namely type III and type VI, the accuracies were 73.9 and 70.4%, respectively.

Conclusion

The proposed models have been adopted to implement a web-based tool (GBSTyper), which is now freely accessible at http://csb.cse.yzu.edu.tw/GBSTyper/, for providing efficient and effective detection of GBS serotypes based on a MALDI-TOF MS spectrum. Overall, this work has demonstrated that the combination of MALDI-TOF MS and machine intelligence could provide a practical means of clinical pathogen testing.

Characterization and Identification of Lysine Succinylation Sites based on Deep Learning Method

Succinylation is a type of protein post-translational modification (PTM), which can play important roles in a variety of cellular processes. Due to an increasing number of site-specific succinylated peptides obtained from high-throughput mass spectrometry (MS), various tools have been developed for computationally identifying succinylated sites on proteins. However, most of these tools predict succinylation sites based on traditional machine learning methods. Hence, this work aimed to carry out the succinylation site prediction based on a deep learning model. The abundance of MS-verified succinylated peptides enabled the investigation of substrate site specificity of succinylation sites through sequence-based attributes, such as position-specific amino acid composition, the composition of k-spaced amino acid pairs (CKSAAP), and position-specific scoring matrix (PSSM). Additionally, the maximal dependence decomposition (MDD) was adopted to detect the substrate signatures of lysine succinylation sites by dividing all succinylated sequences into several groups with conserved substrate motifs. According to the results of ten-fold cross-validation, the deep learning model trained using PSSM and informative CKSAAP attributes can reach the best predictive performance and also perform better than traditional machine-learning methods. Moreover, an independent testing dataset that truly did not exist in the training dataset was used to compare the proposed method with six existing prediction tools. The testing dataset comprised of 218 positive and 2621 negative instances, and the proposed model could yield a promising performance with 84.40% sensitivity, 86.99% specificity, 86.79% accuracy, and an MCC value of 0.489. Finally, the proposed method has been implemented as a web-based prediction tool (CNN-SuccSite), which is now freely accessible at http://csb.cse.yzu.edu.tw/CNN-SuccSite/.

Characterization and identification of lysine glutarylation based on intrinsic interdependence between positions in the substrate sites

Background

Glutarylation, the addition of a glutaryl group (five carbons) to a lysine residue of a protein molecule, is an important post-translational modification and plays a regulatory role in a variety of physiological and biological processes. As the number of experimentally identified glutarylated peptides increases, it becomes imperative to investigate substrate motifs to enhance the study of protein glutarylation. We carried out a bioinformatics investigation of glutarylation sites based on amino acid composition using a public database containing information on 430 non-homologous glutarylation sites.

Results

The TwoSampleLogo analysis indicates that positively charged and polar amino acids surrounding glutarylated sites may be associated with the specificity in substrate site of protein glutarylation. Additionally, the chi-squared test was utilized to explore the intrinsic interdependence between two positions around glutarylation sites. Further, maximal dependence decomposition (MDD), which consists of partitioning a large-scale dataset into subgroups with statistically significant amino acid conservation, was used to capture motif signatures of glutarylation sites. We considered single features, such as amino acid composition (AAC), amino acid pair composition (AAPC), and composition of k-spaced amino acid pairs (CKSAAP), as well as the effectiveness of incorporating MDD-identified substrate motifs into an integrated prediction model. Evaluation by five-fold cross-validation showed that AAC was most effective in discriminating between glutarylation and non-glutarylation sites, according to support vector machine (SVM).

Conclusions

The SVM model integrating MDD-identified substrate motifs performed well, with a sensitivity of 0.677, a specificity of 0.619, an accuracy of 0.638, and a Matthews Correlation Coefficient (MCC) value of 0.28. Using an independent testing dataset (46 glutarylated and 92 non-glutarylated sites) obtained from the literature, we demonstrated that the integrated SVM model could improve the predictive performance effectively, yielding a balanced sensitivity and specificity of 0.652 and 0.739, respectively. This integrated SVM model has been implemented as a web-based system (MDDGlutar), which is now freely available at http://csb.cse.yzu.edu.tw/MDDGlutar/.

Electronic supplementary material

The online version of this article (10.1186/s12859-018-2394-9) contains supplementary material, which is available to authorized users.

A new scheme for strain typing of methicillin-resistant Staphylococcus aureus on the basis of matrix-assisted laser desorption ionization time-of-flight mass spectrometry by using machine learning approach

Methicillin-resistant Staphylococcus aureus (MRSA), one of the most important clinical pathogens, conducts an increasing number of morbidity and mortality in the world. Rapid and accurate strain typing of bacteria would facilitate epidemiological investigation and infection control in near real time. Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry is a rapid and cost-effective tool for presumptive strain typing. To develop robust method for strain typing based on MALDI-TOF spectrum, machine learning (ML) is a promising algorithm for the construction of predictive model. In this study, a strategy of building templates of specific types was used to facilitate generating predictive models of methicillin-resistant Staphylococcus aureus (MRSA) strain typing through various ML methods. The strain types of the isolates were determined through multilocus sequence typing (MLST). The area under the receiver operating characteristic curve (AUC) and the predictive accuracy of the models were compared. ST5, ST59, and ST239 were the major MLST types, and ST45 was the minor type. For binary classification, the AUC values of various ML methods ranged from 0.76 to 0.99 for ST5, ST59, and ST239 types. In multiclass classification, the predictive accuracy of all generated models was more than 0.83. This study has demonstrated that ML methods can serve as a cost-effective and promising tool that provides preliminary strain typing information about major MRSA lineages on the basis of MALDI-TOF spectra.

MDD-carb: a combinatorial model for the identification of protein carbonylation sites with substrate motifs

Background

Carbonylation, which takes place through oxidation of reactive oxygen species (ROS) on specific residues, is an irreversibly oxidative modification of proteins. It has been reported that the carbonylation is related to a number of metabolic or aging diseases including diabetes, chronic lung disease, Parkinson’s disease, and Alzheimer’s disease. Due to the lack of computational methods dedicated to exploring motif signatures of protein carbonylation sites, we were motivated to exploit an iterative statistical method to characterize and identify carbonylated sites with motif signatures.

Results

By manually curating experimental data from research articles, we obtained 332, 144, 135, and 140 verified substrate sites for K (lysine), R (arginine), T (threonine), and P (proline) residues, respectively, from 241 carbonylated proteins. In order to examine the informative attributes for classifying between carbonylated and non-carbonylated sites, multifarious features including composition of twenty amino acids (AAC), composition of amino acid pairs (AAPC), position-specific scoring matrix (PSSM), and positional weighted matrix (PWM) were investigated in this study. Additionally, in an attempt to explore the motif signatures of carbonylation sites, an iterative statistical method was adopted to detect statistically significant dependencies of amino acid compositions between specific positions around substrate sites. Profile hidden Markov model (HMM) was then utilized to train a predictive model from each motif signature. Moreover, based on the method of support vector machine (SVM), we adopted it to construct an integrative model by combining the values of bit scores obtained from profile HMMs. The combinatorial model could provide an enhanced performance with evenly predictive sensitivity and specificity in the evaluation of cross-validation and independent testing.

Conclusion

This study provides a new scheme for exploring potential motif signatures at substrate sites of protein carbonylation. The usefulness of the revealed motifs in the identification of carbonylated sites is demonstrated by their effective performance in cross-validation and independent testing. Finally, these substrate motifs were adopted to build an available online resource (MDD-Carb, http://csb.cse.yzu.edu.tw/MDDCarb/) and are also anticipated to facilitate the study of large-scale carbonylated proteomes.

Electronic supplementary material

The online version of this article (10.1186/s12918-017-0511-4) contains supplementary material, which is available to authorized users.

MDD-Palm: Identification of protein S-palmitoylation sites with substrate motifs based on maximal dependence decomposition

S-palmitoylation, the covalent attachment of 16-carbon palmitic acids to a cysteine residue via a thioester linkage, is an important reversible lipid modification that plays a regulatory role in a variety of physiological and biological processes. As the number of experimentally identified S-palmitoylated peptides increases, it is imperative to investigate substrate motifs to facilitate the study of protein S-palmitoylation. Based on 710 non-homologous S-palmitoylation sites obtained from published databases and the literature, we carried out a bioinformatics investigation of S-palmitoylation sites based on amino acid composition. Two Sample Logo indicates that positively charged and polar amino acids surrounding S-palmitoylated sites may be associated with the substrate site specificity of protein S-palmitoylation. Additionally, maximal dependence decomposition (MDD) was applied to explore the motif signatures of S-palmitoylation sites by categorizing a large-scale dataset into subgroups with statistically significant conservation of amino acids. Single features such as amino acid composition (AAC), amino acid pair composition (AAPC), position specific scoring matrix (PSSM), position weight matrix (PWM), amino acid substitution matrix (BLOSUM62), and accessible surface area (ASA) were considered, along with the effectiveness of incorporating MDD-identified substrate motifs into a two-layered prediction model. Evaluation by five-fold cross-validation showed that a hybrid of AAC and PSSM performs best at discriminating between S-palmitoylation and non-S-palmitoylation sites, according to the support vector machine (SVM). The two-layered SVM model integrating MDD-identified substrate motifs performed well, with a sensitivity of 0.79, specificity of 0.80, accuracy of 0.80, and Matthews Correlation Coefficient (MCC) value of 0.45. Using an independent testing dataset (613 S-palmitoylated and 5412 non-S-palmitoylated sites) obtained from the literature, we demonstrated that the two-layered SVM model could outperform other prediction tools, yielding a balanced sensitivity and specificity of 0.690 and 0.694, respectively. This two-layered SVM model has been implemented as a web-based system (MDD-Palm), which is now freely available at http://csb.cse.yzu.edu.tw/MDDPalm/.

Investigation and identification of protein carbonylation sites based on position-specific amino acid composition and physicochemical features

BACKGROUND

Protein carbonylation, an irreversible and non-enzymatic post-translational modification (PTM), is often used as a marker of oxidative stress. When reactive oxygen species (ROS) oxidized the amino acid side chains, carbonyl (CO) groups are produced especially on Lysine (K), Arginine (R), Threonine (T), and Proline (P). Nevertheless, due to the lack of information about the carbonylated substrate specificity, we were encouraged to develop a systematic method for a comprehensive investigation of protein carbonylation sites.

RESULTS

After the removal of redundant data from multipe carbonylation-related articles, totally 226 carbonylated proteins in human are regarded as training dataset, which consisted of 307, 126, 128, and 129 carbonylation sites for K, R, T and P residues, respectively. To identify the useful features in predicting carbonylation sites, the linear amino acid sequence was adopted not only to build up the predictive model from training dataset, but also to compare the effectiveness of prediction with other types of features including amino acid composition (AAC), amino acid pair composition (AAPC), position-specific scoring matrix (PSSM), positional weighted matrix (PWM), solvent-accessible surface area (ASA), and physicochemical properties. The investigation of position-specific amino acid composition revealed that the positively charged amino acids (K and R) are remarkably enriched surrounding the carbonylated sites, which may play a functional role in discriminating between carbonylation and non-carbonylation sites. A variety of predictive models were built using various features and three different machine learning methods. Based on the evaluation by five-fold cross-validation, the models trained with PWM feature could provide better sensitivity in the positive training dataset, while the models trained with AAindex feature achieved higher specificity in the negative training dataset. Additionally, the model trained using hybrid features, including PWM, AAC and AAindex, obtained best MCC values of 0.432, 0.472, 0.443 and 0.467 on K, R, T and P residues, respectively.

CONCLUSION

When comparing to an existing prediction tool, the selected models trained with hybrid features provided a promising accuracy on an independent testing dataset. In short, this work not only characterized the carbonylated substrate preference, but also demonstrated that the proposed method could provide a feasible means for accelerating preliminary discovery of protein carbonylation.

A New Scheme to Characterize and Identify Protein Ubiquitination Sites

Protein ubiquitination, involving the conjugation of ubiquitin on lysine residue, serves as an important modulator of many cellular functions in eukaryotes. Recent advancements in proteomic technology have stimulated increasing interest in identifying ubiquitination sites. However, most computational tools for predicting ubiquitination sites are focused on small-scale data. With an increasing number of experimentally verified ubiquitination sites, we were motivated to design a predictive model for identifying lysine ubiquitination sites for large-scale proteome dataset. This work assessed not only single features, such as amino acid composition (AAC), amino acid pair composition (AAPC) and evolutionary information, but also the effectiveness of incorporating two or more features into a hybrid approach to model construction. The support vector machine (SVM) was applied to generate the prediction models for ubiquitination site identification. Evaluation by five-fold cross-validation showed that the SVM models learned from the combination of hybrid features delivered a better prediction performance. Additionally, a motif discovery tool, MDDLogo, was adopted to characterize the potential substrate motifs of ubiquitination sites. The SVM models integrating the MDDLogo-identified substrate motifs could yield an average accuracy of 68.70 percent. Furthermore, the independent testing result showed that the MDDLogo-clustered SVM models could provide a promising accuracy (78.50 percent) and perform better than other prediction tools. Two cases have demonstrated the effective prediction of ubiquitination sites with corresponding substrate motifs.

UbiNet: an online resource for exploring the functional associations and regulatory networks of protein ubiquitylation

Protein ubiquitylation catalyzed by E3 ubiquitin ligases are crucial in the regulation of many cellular processes. Owing to the high throughput of mass spectrometry-based proteomics, a number of methods have been developed for the experimental determination of ubiquitylation sites, leading to a large collection of ubiquitylation data. However, there exist no resources for the exploration of E3-ligase-associated regulatory networks of for ubiquitylated proteins in humans. Therefore, the UbiNet database was developed to provide a full investigation of protein ubiquitylation networks by incorporating experimentally verified E3 ligases, ubiquitylated substrates and protein-protein interactions (PPIs). To date, UbiNet has accumulated 43 948 experimentally verified ubiquitylation sites from 14 692 ubiquitylated proteins of humans. Additionally, we have manually curated 499 E3 ligases as well as two E1 activating and 46 E2 conjugating enzymes. To delineate the regulatory networks among E3 ligases and ubiquitylated proteins, a total of 430 530 PPIs were integrated into UbiNet for the exploration of ubiquitylation networks with an interactive network viewer. A case study demonstrated that UbiNet was able to decipher a scheme for the ubiquitylation of tumor proteins p63 and p73 that is consistent with their functions. Although the essential role of Mdm2 in p53 regulation is well studied, UbiNet revealed that Mdm2 and additional E3 ligases might be implicated in the regulation of other tumor proteins by protein ubiquitylation. Moreover, UbiNet could identify potential substrates for a specific E3 ligase based on PPIs and substrate motifs. With limited knowledge about the mechanisms through which ubiquitylated proteins are regulated by E3 ligases, UbiNet offers users an effective means for conducting preliminary analyses of protein ubiquitylation. The UbiNet database is now freely accessible via http://csb.cse.yzu.edu.tw/UbiNet/ The content is regularly updated with the literature and newly released data.Database URL: http://csb.cse.yzu.edu.tw/UbiNet/.

SOHSite: incorporating evolutionary information and physicochemical properties to identify protein S-sulfenylation sites

Background

Protein S-sulfenylation is a type of post-translational modification (PTM) involving the covalent binding of a hydroxyl group to the thiol of a cysteine amino acid. Recent evidence has shown the importance of S-sulfenylation in various biological processes, including transcriptional regulation, apoptosis and cytokine signaling. Determining the specific sites of S-sulfenylation is fundamental to understanding the structures and functions of S-sulfenylated proteins. However, the current lack of reliable tools often limits researchers to use expensive and time-consuming laboratory techniques for the identification of S-sulfenylation sites. Thus, we were motivated to develop a bioinformatics method for investigating S-sulfenylation sites based on amino acid compositions and physicochemical properties.

Results

In this work, physicochemical properties were utilized not only to identify S-sulfenylation sites from 1,096 experimentally verified S-sulfenylated proteins, but also to compare the effectiveness of prediction with other characteristics such as amino acid composition (AAC), amino acid pair composition (AAPC), solvent-accessible surface area (ASA), amino acid substitution matrix (BLOSUM62), position-specific scoring matrix (PSSM), and positional weighted matrix (PWM). Various prediction models were built using support vector machine (SVM) and evaluated by five-fold cross-validation. The model constructed from hybrid features, including PSSM and physicochemical properties, yielded the best performance with sensitivity, specificity, accuracy and MCC measurements of 0.746, 0.737, 0.738 and 0.337, respectively. The selected model also provided a promising accuracy (0.693) on an independent testing dataset. Additionally, we employed TwoSampleLogo to help discover the difference of amino acid composition among S-sulfenylation, S-glutathionylation and S-nitrosylation sites.

Conclusion

This work proposed a computational method to explore informative features and functions for protein S-sulfenylation. Evaluation by five-fold cross validation indicated that the selected features were effective in the identification of S-sulfenylation sites. Moreover, the independent testing results demonstrated that the proposed method could provide a feasible means for conducting preliminary analyses of protein S-sulfenylation. We also anticipate that the uncovered differences in amino acid composition may facilitate future studies of the extensive crosstalk among S-sulfenylation, S-glutathionylation and S-nitrosylation.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-015-2299-1) contains supplementary material, which is available to authorized users.

UbiSite: incorporating two-layered machine learning method with substrate motifs to predict ubiquitin-conjugation site on lysines

Background

The conjugation of ubiquitin to a substrate protein (protein ubiquitylation), which involves a sequential process – E1 activation, E2 conjugation and E3 ligation, is crucial to the regulation of protein function and activity in eukaryotes. This ubiquitin-conjugation process typically binds the last amino acid of ubiquitin (glycine 76) to a lysine residue of a target protein. The high-throughput of mass spectrometry-based proteomics has stimulated a large-scale identification of ubiquitin-conjugated peptides. Hence, a new web resource, UbiSite, was developed to identify ubiquitin-conjugation site on lysines based on large-scale proteome dataset.

Results

Given a total of 37,647 ubiquitin-conjugated proteins, including 128026 ubiquitylated peptides, obtained from various resources, this study carries out a large-scale investigation on ubiquitin-conjugation sites based on sequenced and structural characteristics. A TwoSampleLogo reveals that a significant depletion of histidine (H), arginine (R) and cysteine (C) residues around ubiquitylation sites may impact the conjugation of ubiquitins in closed three-dimensional environments. Based on the large-scale ubiquitylation dataset, a motif discovery tool, MDDLogo, has been adopted to characterize the potential substrate motifs for ubiquitin conjugation. Not only are single features such as amino acid composition (AAC), positional weighted matrix (PWM), position-specific scoring matrix (PSSM) and solvent-accessible surface area (SASA) considered, but also the effectiveness of incorporating MDDLogo-identified substrate motifs into a two-layered prediction model is taken into account. Evaluation by five-fold cross-validation showed that PSSM is the best feature in discriminating between ubiquitylation and non-ubiquitylation sites, based on support vector machine (SVM). Additionally, the two-layered SVM model integrating MDDLogo-identified substrate motifs could obtain a promising accuracy and the Matthews Correlation Coefficient (MCC) at 81.06 % and 0.586, respectively. Furthermore, the independent testing showed that the two-layered SVM model could outperform other prediction tools, reaching at 85.10 % sensitivity, 69.69 % specificity, 73.69 % accuracy and the 0.483 of MCC value.

Conclusion

The independent testing result indicated the effectiveness of incorporating MDDLogo-identified motifs into the prediction of ubiquitylation sites. In order to provide meaningful assistance to researchers interested in large-scale ubiquitinome data, the two-layered SVM model has been implemented onto a web-based system (UbiSite), which is freely available at http://csb.cse.yzu.edu.tw/UbiSite/. Two cases given in the UbiSite provide a demonstration of effective identification of ubiquitylation sites with reference to substrate motifs.

Electronic supplementary material

The online version of this article (doi:10.1186/s12918-015-0246-z) contains supplementary material, which is available to authorized users.

A two-layered machine learning method to identify protein O-GlcNAcylation sites with O-GlcNAc transferase substrate motifs

Protein O-GlcNAcylation, involving the β-attachment of single N-acetylglucosamine (GlcNAc) to the hydroxyl group of serine or threonine residues, is an O-linked glycosylation catalyzed by O-GlcNAc transferase (OGT). Molecular level investigation of the basis for OGT's substrate specificity should aid understanding how O-GlcNAc contributes to diverse cellular processes. Due to an increasing number of O-GlcNAcylated peptides with site-specific information identified by mass spectrometry (MS)-based proteomics, we were motivated to characterize substrate site motifs of O-GlcNAc transferases. In this investigation, a non-redundant dataset of 410 experimentally verified O-GlcNAcylation sites were manually extracted from dbOGAP, OGlycBase and UniProtKB. After detection of conserved motifs by using maximal dependence decomposition, profile hidden Markov model (profile HMM) was adopted to learn a first-layered model for each identified OGT substrate motif. Support Vector Machine (SVM) was then used to generate a second-layered model learned from the output values of profile HMMs in first layer. The two-layered predictive model was evaluated using a five-fold cross validation which yielded a sensitivity of 85.4%, a specificity of 84.1%, and an accuracy of 84.7%. Additionally, an independent testing set from PhosphoSitePlus, which was really non-homologous to the training data of predictive model, was used to demonstrate that the proposed method could provide a promising accuracy (84.05%) and outperform other O-GlcNAcylation site prediction tools. A case study indicated that the proposed method could be a feasible means of conducting preliminary analyses of protein O-GlcNAcylation and has been implemented as a web-based system, OGTSite, which is now freely available at http://csb.cse.yzu.edu.tw/OGTSite/.

MDD-SOH: exploiting maximal dependence decomposition to identify S-sulfenylation sites with substrate motifs

S-sulfenylation (S-sulphenylation, or sulfenic acid), the covalent attachment of S-hydroxyl (-SOH) to cysteine thiol, plays a significant role in redox regulation of protein functions. Although sulfenic acid is transient and labile, most of its physiological activities occur under control of S-hydroxylation. Therefore, discriminating the substrate site of S-sulfenylated proteins is an essential task in computational biology for the furtherance of protein structures and functions. Research into S-sulfenylated protein is currently very limited, and no dedicated tools are available for the computational identification of SOH sites. Given a total of 1096 experimentally verified S-sulfenylated proteins from humans, this study carries out a bioinformatics investigation on SOH sites based on amino acid composition and solvent-accessible surface area. A TwoSampleLogo indicates that the positively and negatively charged amino acids flanking the SOH sites may impact the formulation of S-sulfenylation in closed three-dimensional environments. In addition, the substrate motifs of SOH sites are studied using the maximal dependence decomposition (MDD). Based on the concept of binary classification between SOH and non-SOH sites, Support vector machine (SVM) is applied to learn the predictive model from MDD-identified substrate motifs. According to the evaluation results of 5-fold cross-validation, the integrated SVM model learned from substrate motifs yields an average accuracy of 0.87, significantly improving the prediction of SOH sites. Furthermore, the integrated SVM model also effectively improves the predictive performance in an independent testing set. Finally, the integrated SVM model is applied to implement an effective web resource, named MDD-SOH, to identify SOH sites with their corresponding substrate motifs.

AVAILABILITY AND IMPLEMENTATION

The MDD-SOH is now freely available to all interested users at http://csb.cse.yzu.edu.tw/MDDSOH/. All of the data set used in this work is also available for download in the website.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

CONTACT

francis@saturn.yzu.edu.tw.

GSHSite: exploiting an iteratively statistical method to identify s-glutathionylation sites with substrate specificity

S-glutathionylation, the covalent attachment of a glutathione (GSH) to the sulfur atom of cysteine, is a selective and reversible protein post-translational modification (PTM) that regulates protein activity, localization, and stability. Despite its implication in the regulation of protein functions and cell signaling, the substrate specificity of cysteine S-glutathionylation remains unknown. Based on a total of 1783 experimentally identified S-glutathionylation sites from mouse macrophages, this work presents an informatics investigation on S-glutathionylation sites including structural factors such as the flanking amino acids composition and the accessible surface area (ASA). TwoSampleLogo presents that positively charged amino acids flanking the S-glutathionylated cysteine may influence the formation of S-glutathionylation in closed three-dimensional environment. A statistical method is further applied to iteratively detect the conserved substrate motifs with statistical significance. Support vector machine (SVM) is then applied to generate predictive model considering the substrate motifs. According to five-fold cross-validation, the SVMs trained with substrate motifs could achieve an enhanced sensitivity, specificity, and accuracy, and provides a promising performance in an independent test set. The effectiveness of the proposed method is demonstrated by the correct identification of previously reported S-glutathionylation sites of mouse thioredoxin (TXN) and human protein tyrosine phosphatase 1b (PTP1B). Finally, the constructed models are adopted to implement an effective web-based tool, named GSHSite (http://csb.cse.yzu.edu.tw/GSHSite/), for identifying uncharacterized GSH substrate sites on the protein sequences.

Characterization and identification of ubiquitin conjugation sites with E3 ligase recognition specificities

Background

In eukaryotes, ubiquitin-conjugation is an important mechanism underlying proteasome-mediated degradation of proteins, and as such, plays an essential role in the regulation of many cellular processes. In the ubiquitin-proteasome pathway, E3 ligases play important roles by recognizing a specific protein substrate and catalyzing the attachment of ubiquitin to a lysine (K) residue. As more and more experimental data on ubiquitin conjugation sites become available, it becomes possible to develop prediction models that can be scaled to big data. However, no development that focuses on the investigation of ubiquitinated substrate specificities has existed. Herein, we present an approach that exploits an iteratively statistical method to identify ubiquitin conjugation sites with substrate site specificities.

Results

In this investigation, totally 6259 experimentally validated ubiquitinated proteins were obtained from dbPTM. After having filtered out homologous fragments with 40% sequence identity, the training data set contained 2658 ubiquitination sites (positive data) and 5532 non-ubiquitinated sites (negative data). Due to the difficulty in characterizing the substrate site specificities of E3 ligases by conventional sequence logo analysis, a recursively statistical method has been applied to obtain significant conserved motifs. The profile hidden Markov model (profile HMM) was adopted to construct the predictive models learned from the identified substrate motifs. A five-fold cross validation was then used to evaluate the predictive model, achieving sensitivity, specificity, and accuracy of 73.07%, 65.46%, and 67.93%, respectively. Additionally, an independent testing set, completely blind to the training data of the predictive model, was used to demonstrate that the proposed method could provide a promising accuracy (76.13%) and outperform other ubiquitination site prediction tool.

Conclusion

A case study demonstrated the effectiveness of the characterized substrate motifs for identifying ubiquitination sites. The proposed method presents a practical means of preliminary analysis and greatly diminishes the total number of potential targets required for further experimental confirmation. This method may help unravel their mechanisms and roles in E3 recognition and ubiquitin-mediated protein degradation.

Characterization and identification of protein O-GlcNAcylation sites with substrate specificity

Background

Protein O-GlcNAcylation, involving the attachment of single N-acetylglucosamine (GlcNAc) to the hydroxyl group of serine or threonine residues. Elucidation of O-GlcNAcylation sites on proteins is required in order to decipher its crucial roles in regulating cellular processes and aid in drug design. With an increasing number of O-GlcNAcylation sites identified by mass spectrometry (MS)-based proteomics, several methods have been proposed for the computational identification of O-GlcNAcylation sites. However, no development that focuses on the investigation of O-GlcNAcylated substrate motifs has existed. Thus, we were motivated to design a new method for the identification of protein O-GlcNAcylation sites with the consideration of substrate site specificity.

Results

In this study, 375 experimentally verified O-GlcNAcylation sites were collected from dbOGAP, which is an integrated resource for protein O-GlcNAcylation. Due to the difficulty in characterizing the substrate motifs by conventional sequence logo analysis, a recursively statistical method has been applied to obtain significant conserved motifs. To construct the predictive models learned from the identified substrate motifs, we adopted Support Vector Machines (SVMs). A five-fold cross validation was used to evaluate the predictive model, achieving sensitivity, specificity, and accuracy of 0.76, 0.80, and 0.78, respectively. Additionally, an independent testing set, which was really blind to the training data of predictive model, was used to demonstrate that the proposed method could provide a promising accuracy (0.94) and outperform three other O-GlcNAcylation site prediction tools.

Conclusion

This work proposed a computational method to identify informative substrate motifs for O-GlcNAcylation sites. The evaluation of cross validation and independent testing indicated that the identified motifs were effective in the identification of O-GlcNAcylation sites. A case study demonstrated that the proposed method could be a feasible means of conducting preliminary analyses of protein O-GlcNAcylation. We also anticipated that the revealed substrate motif may facilitate the study of extensive crosstalk between O-GlcNAcylation and phosphorylation. This method may help unravel their mechanisms and roles in signaling, transcription, chronic disease, and cancer.

An intelligent system for identifying acetylated lysine on histones and nonhistone proteins

Lysine acetylation is an important and ubiquitous posttranslational modification conserved in prokaryotes and eukaryotes. This process, which is dynamically and temporally regulated by histone acetyltransferases and deacetylases, is crucial for numerous essential biological processes such as transcriptional regulation, cellular signaling, and stress response. Since the experimental identification of lysine acetylation sites within proteins is time-consuming and laboratory-intensive, several computational approaches have been developed to identify candidates for experimental validation. In this work, acetylated protein data collected from UniProtKB were categorized into histone or nonhistone proteins. Support vector machines (SVMs) were applied to build predictive models by using amino acid pair composition (AAPC) as a feature in a histone model. We combined BLOSUM62 and AAPC features in a nonhistone model. Furthermore, using maximal dependence decomposition (MDD) clustering can enhance the performance of the model on a fivefold cross-validation evaluation to yield a sensitivity of 0.863, specificity of 0.885, accuracy of 0.880, and MCC of 0.706. Additionally, the proposed method is evaluated using independent test sets resulting in a predictive accuracy of 74%. This indicates that the performance of our method is comparable with that of other acetylation prediction methods.

ViralPhos: incorporating a recursively statistical method to predict phosphorylation sites on virus proteins

Background

The phosphorylation of virus proteins by host kinases is linked to viral replication. This leads to an inhibition of normal host-cell functions. Further elucidation of phosphorylation in virus proteins is required in order to aid in drug design and treatment. However, only a few studies have investigated substrate motifs in identifying virus phosphorylation sites. Additionally, existing bioinformatics tool do not consider potential host kinases that may initiate the phosphorylation of a virus protein.

Results

329 experimentally verified phosphorylation fragments on 111 virus proteins were collected from virPTM. These were clustered into subgroups of significantly conserved motifs using a recursively statistical method. Two-layered Support Vector Machines (SVMs) were then applied to train a predictive model for the identified substrate motifs. The SVM models were evaluated using a five-fold cross validation which yields an average accuracy of 0.86 for serine, and 0.81 for threonine. Furthermore, the proposed method is shown to perform at par with three other phosphorylation site prediction tools: PPSP, KinasePhos 2.0 and GPS 2.1.

Conclusion

In this study, we propose a computational method, ViralPhos, which aims to investigate virus substrate site motifs and identify potential phosphorylation sites on virus proteins. We identified informative substrate motifs that matched with several well-studied kinase groups as potential catalytic kinases for virus protein substrates. The identified substrate motifs were further exploited to identify potential virus phosphorylation sites. The proposed method is shown to be capable of predicting virus phosphorylation sites and has been implemented as a web server http://csb.cse.yzu.edu.tw/ViralPhos/.

DbPTM 3.0: an informative resource for investigating substrate site specificity and functional association of protein post-translational modifications

Protein modification is an extremely important post-translational regulation that adjusts the physical and chemical properties, conformation, stability and activity of a protein; thus altering protein function. Due to the high throughput of mass spectrometry (MS)-based methods in identifying site-specific post-translational modifications (PTMs), dbPTM (http://dbPTM.mbc.nctu.edu.tw/) is updated to integrate experimental PTMs obtained from public resources as well as manually curated MS/MS peptides associated with PTMs from research articles. Version 3.0 of dbPTM aims to be an informative resource for investigating the substrate specificity of PTM sites and functional association of PTMs between substrates and their interacting proteins. In order to investigate the substrate specificity for modification sites, a newly developed statistical method has been applied to identify the significant substrate motifs for each type of PTMs containing sufficient experimental data. According to the data statistics in dbPTM, >60% of PTM sites are located in the functional domains of proteins. It is known that most PTMs can create binding sites for specific protein-interaction domains that work together for cellular function. Thus, this update integrates protein–protein interaction and domain–domain interaction to determine the functional association of PTM sites located in protein-interacting domains. Additionally, the information of structural topologies on transmembrane (TM) proteins is integrated in dbPTM in order to delineate the structural correlation between the reported PTM sites and TM topologies. To facilitate the investigation of PTMs on TM proteins, the PTM substrate sites and the structural topology are graphically represented. Also, literature information related to PTMs, orthologous conservations and substrate motifs of PTMs are also provided in the resource. Finally, this version features an improved web interface to facilitate convenient access to the resource.

Computational prediction and analysis of protein γ-carboxylation sites based on a random forest method

The glutamate γ-carboxylation plays a pivotal part in a number of important human diseases. However, traditional protein γ-carboxylation site detection by experimental approaches are often laborious and time-consuming. In this study, we initiated an attempt for the computational prediction of protein γ-carboxylation sites. We developed a new method for predicting the γ-carboxylation sites based on a Random Forest method. As a result, 90.44% accuracy and 0.7739 MCC value were obtained for the training dataset, and 89.83% accuracy and 0.7448 MCC value for the testing dataset. Our method considered several features including sequence conservation, residual disorder, secondary structures, solvent accessibility, physicochemical/biochemical properties and amino acid occurrence frequencies. By means of the feature selection algorithm, an optimal set of 327 features were selected; these features were considered as the ones that contributed significantly to the prediction of protein γ-carboxylation sites. Analysis of the optimal feature set indicated several important factors in determining the γ-carboxylation and a possible consensus sequence of the γ-carboxylation recognition site (γ-CRS) was suggested. These may shed some light on the in-depth understanding of the mechanisms of γ-carboxylation, providing guidelines for experimental validation.

Identifying protein phosphorylation sites with kinase substrate specificity on human viruses

Viruses infect humans and progress inside the body leading to various diseases and complications. The phosphorylation of viral proteins catalyzed by host kinases plays crucial regulatory roles in enhancing replication and inhibition of normal host-cell functions. Due to its biological importance, there is a desire to identify the protein phosphorylation sites on human viruses. However, the use of mass spectrometry-based experiments is proven to be expensive and labor-intensive. Furthermore, previous studies which have identified phosphorylation sites in human viruses do not include the investigation of the responsible kinases. Thus, we are motivated to propose a new method to identify protein phosphorylation sites with its kinase substrate specificity on human viruses. The experimentally verified phosphorylation data were extracted from virPTM – a database containing 301 experimentally verified phosphorylation data on 104 human kinase-phosphorylated virus proteins. In an attempt to investigate kinase substrate specificities in viral protein phosphorylation sites, maximal dependence decomposition (MDD) is employed to cluster a large set of phosphorylation data into subgroups containing significantly conserved motifs. The experimental human phosphorylation sites are collected from Phospho.ELM, grouped according to its kinase annotation, and compared with the virus MDD clusters. This investigation identifies human kinases such as CK2, PKB, CDK, and MAPK as potential kinases for catalyzing virus protein substrates as confirmed by published literature. Profile hidden Markov model is then applied to learn a predictive model for each subgroup. A five-fold cross validation evaluation on the MDD-clustered HMMs yields an average accuracy of 84.93% for Serine, and 78.05% for Threonine. Furthermore, an independent testing data collected from UniProtKB and Phospho.ELM is used to make a comparison of predictive performance on three popular kinase-specific phosphorylation site prediction tools. In the independent testing, the high sensitivity and specificity of the proposed method demonstrate the predictive effectiveness of the identified substrate motifs and the importance of investigating potential kinases for viral protein phosphorylation sites.