Method For Prediction Of Proteins Research Articles

The Development and Opportunities of Predictive Biotechnology.

Recent advances in bioeconomy allow a holistic view of existing and new process chains and enable novel production routines continuously advanced by academia and industry. All this progress benefits from a growing number of prediction tools that have found their way into the field. For example, automated genome annotations, tools for building model structures of proteins, and structural protein prediction methods such as AlphaFold2TM or RoseTTAFold have gained popularity in recent years. Recently, it has become apparent that more and more AI-based tools are being developed and used for biocatalysis and biotechnology. This is an excellent opportunity for academia and industry to accelerate advancements in the field further. Biotechnology, as a rapidly growing interdisciplinary field, stands to benefit greatly from these developments.

LBi-DBP, an accurate DNA-binding protein prediction method based lightweight interpretable BiLSTM network

Accurate prediction of the DNA-binding protein (DBP) helps to uncover the underlying mechanism of DNA-protein interaction, thus contributing to the understanding of many biochemical processes. Although the computational methods have made great process, there is still much room for improving the DBP prediction. In this study, a lightweight and interpretable sequence-based method, named LBi-DBP, is well-designed to improve the performance of DBP prediction. In LBi-DBP, five sequence feature sources are first employed to extract important sequence context information using a slight bidirectional long short-term memory (BiLSTM)-based neural network module; then, a multi-layer perceptron (MLP) module is utilized to learn the optimal/sub-optimal parameters with sequence context information extracted by BiLSTM module and pseudo sequence order feature as inputs. Experimental results on two independent test sets, i.e., UniSwiss-Tst and PDB2272, demonstrate that LBi-DBP can achieve Matthew’s correlation coefficient values of 0.762 and 0.574 respectively, which are both higher than those of the second-best existing DBP prediction method (0.741 and 0.424). Meanwhile, good prediction result on a disease-related test set also shows the value of LBi-DBP. Through interpretability analysis, we find that the BiLSTM module can focus on the residues in real native DNA-binding domain region without prior knowledge, thus improving DNA-binding protein prediction performance.

VirulentPred 2.0: An improved method for prediction of virulent proteins in bacterial pathogens.

Virulence proteins in pathogens are essential for causing disease in a host. They enable the pathogen to invade, survive and multiply within the host, thus enhancing its potential to cause disease while also causing evasion of host defense mechanisms. Identifying these factors, especially potential vaccine candidates or drug targets, is critical for vaccine or drug development research. In this context, we present an improved version of VirulentPred 1.0 for rapidly identifying virulent proteins. The VirulentPred 2.0 is based on training machine learning models with experimentally validated virulent protein sequences. VirulentPred 2.0 achieved 84.71% accuracy with the validation dataset and 85.18% on an independent test dataset. The models are trained and evaluated with the latest sequence datasets of virulent proteins, which are three times greater in number than the proteins used in the earlier version of VirulentPred. Moreover, a significant improvement of 11% in the prediction accuracy over the earlier version is achieved with the best position-specific scoring matrix (PSSM)-based model for the latest test dataset. VirulentPred 2.0 is available as a user-friendly web interface at https://bioinfo.icgeb.res.in/virulent2/ and a standalone application suitable for bulk predictions. With higher efficiency and availability as a standalone tool, VirulentPred 2.0 holds immense potential for high throughput yet efficient identification of virulent proteins in bacterial pathogens.

A Quantitative Prediction Method for the Human Pharmacokinetics of Fc-Fusion Proteins.

Fc fusion is an effective strategy for extending the half-lives of therapeutic proteins. This study aimed to evaluate the applicability of a human pharmacokinetics prediction method for Fc-fusion proteins by extending on reported methods for monoclonal antibodies (mAbs). To predict human pharmacokinetic profiles following intravenous (IV) dosing, the pharmacokinetic data for 11 Fc-fusion proteins in monkeys were analysed by two approaches: a species-invariant time method with a range of allometric exponents in clearance (CL, 0.7-1.0) and a two-compartment model reported for mAbs. The pharmacokinetic profiles following subcutaneous (SC) dosing were predicted by simple dose normalisation from monkeys or using the geometric means of the absorption rate constant (Ka) and bioavailability (BA) for mAbs or Fc-fusion proteins in humans and compared. In the case of IV administration, the area under the curve could be predicted for more than 85% of Fc-fusion proteins within a twofold difference from the observed value using the species-invariant time method (scaling exponent for CL, 0.95). For SC dosing, incorporating the geometric means of absorption parameters for both mAbs (BA 68.2%, Ka 0.287 day-1) and Fc-fusion proteins (BA 63.0%, Ka 0.209 day-1) in humans provided better accuracy than simple normalisation from monkeys. We have successfully predicted the human pharmacokinetic profiles of Fc-fusion proteins for both IV and SC administration within twofold of the observed value from monkey pharmacokinetic data by extending on reported methods for mAbs. This method will facilitate drug discovery and development of Fc-fusion proteins.

IDRPro-SC: identifying DNA-binding proteins and RNA-binding proteins based on subfunction classifiers.

Nucleic acid-binding proteins are proteins that interact with DNA and RNA to regulate gene expression and transcriptional control. The pathogenesis of many human diseases is related to abnormal gene expression. Therefore, recognizing nucleic acid-binding proteins accurately and efficiently has important implications for disease research. To address this question, some scientists have proposed the method of using sequence information to identify nucleic acid-binding proteins. However, different types of nucleic acid-binding proteins have different subfunctions, and these methods ignore their internal differences, so the performance of the predictor can be further improved. In this study, we proposed a new method, called iDRPro-SC, to predict the type of nucleic acid-binding proteins based on the sequence information. iDRPro-SC considers the internal differences of nucleic acid-binding proteins and combines their subfunctions to build a complete dataset. Additionally, we used an ensemble learning to characterize and predict nucleic acid-binding proteins. The results of the test dataset showed that iDRPro-SC achieved the best prediction performance and was superior to the other existing nucleic acid-binding protein prediction methods. We have established a web server that can be accessed online: http://bliulab.net/iDRPro-SC.

RLPredictiOme, a Machine Learning-Derived Method for High-Throughput Prediction of Plant Receptor-like Proteins, Reveals Novel Classes of Transmembrane Receptors.

Cell surface receptors play essential roles in perceiving and processing external and internal signals at the cell surface of plants and animals. The receptor-like protein kinases (RLK) and receptor-like proteins (RLPs), two major classes of proteins with membrane receptor configuration, play a crucial role in plant development and disease defense. Although RLPs and RLKs share a similar single-pass transmembrane configuration, RLPs harbor short divergent C-terminal regions instead of the conserved kinase domain of RLKs. This RLP receptor structural design precludes sequence comparison algorithms from being used for high-throughput predictions of the RLP family in plant genomes, as has been extensively performed for RLK superfamily predictions. Here, we developed the RLPredictiOme, implemented with machine learning models in combination with Bayesian inference, capable of predicting RLP subfamilies in plant genomes. The ML models were simultaneously trained using six types of features, along with three stages to distinguish RLPs from non-RLPs (NRLPs), RLPs from RLKs, and classify new subfamilies of RLPs in plants. The ML models achieved high accuracy, precision, sensitivity, and specificity for predicting RLPs with relatively high probability ranging from 0.79 to 0.99. The prediction of the method was assessed with three datasets, two of which contained leucine-rich repeats (LRR)-RLPs from Arabidopsis and rice, and the last one consisted of the complete set of previously described Arabidopsis RLPs. In these validation tests, more than 90% of known RLPs were correctly predicted via RLPredictiOme. In addition to predicting previously characterized RLPs, RLPredictiOme uncovered new RLP subfamilies in the Arabidopsis genome. These include probable lipid transfer (PLT)-RLP, plastocyanin-like-RLP, ring finger-RLP, glycosyl-hydrolase-RLP, and glycerophosphoryldiester phosphodiesterase (GDPD, GDPDL)-RLP subfamilies, yet to be characterized. Compared to the only Arabidopsis GDPDL-RLK, molecular evolution studies confirmed that the ectodomain of GDPDL-RLPs might have undergone a purifying selection with a predominance of synonymous substitutions. Expression analyses revealed that predicted GDPGL-RLPs display a basal expression level and respond to developmental and biotic signals. The results of these biological assays indicate that these subfamily members have maintained functional domains during evolution and may play relevant roles in development and plant defense. Therefore, RLPredictiOme provides a framework for genome-wide surveys of the RLP superfamily as a foundation to rationalize functional studies of surface receptors and their relationships with different biological processes.

Prediction of protein mononucleotide binding sites using AlphaFold2 and machine learning

In this study, we developed a system that predicts the binding sites of proteins for five mononucleotides (AMP, ADP, ATP, GDP, and GTP). The system comprises two machine learning (ML)-based predictors using a convolutional neural network and a gradient boosting machine, two template-based predictors based on sequence and structure alignment, and a predictor that performs ensemble learning of these four predictors. In this study, data augmentation of ligand binding sites with similar ligand structures was performed. For example, in the prediction of ADP-binding sites using ML methods, the binding sites of AMP and ATP, which have similar structures, are considered. In addition, we constructed the structure models using AlphaFold2, a highly accurate protein prediction method. The secondary structure and dihedral angle information obtained using the model structures were used as ML predictor features. Additionally, in the template-based predictor, the structures of the binding sites were used as templates to be explored based on structure alignment to identify the binding site of the target. Consequently, the template-based predictor based on structure alignment showed the best performance among the four individual predictors, and the ensemble predictor achieved the best performance, with an area under the curve of 0.958 for all mononucleotides.

Cov_DOX: A Method for Structure Prediction of Covalent Protein-Ligand Bindings.

A handful of molecular docking tools have been extended to enable a covalent docking. However, all of them face the challenge brought by the covalent bond between proteins and ligands. Many covalent drug design scenarios still heavily rely on demanding crystallographic experiments for accurate binding structures. Aiming at filling the gap between covalent dockings and crystallographic experiments, we develop and validate a hybrid method, dubbed as Cov_DOX, in this work. Cov_DOX achieves an overall success rate of 81% with RMSD < 2 Å for the Top 1 pose prediction in the validation against a test set including 405 crystal structures for covalent protein-ligand complexes, covering various types of the warhead chemistry and receptors. Such accuracy is not far from the much more demanding crystallographic experiments, in sharp contrast to the performance of the covalent docking front runners (success rate: 40-60%).

SecProCT: In Silico Prediction of Human Secretory Proteins Based on Capsule Network and Transformer.

Identifying secretory proteins from blood, saliva or other body fluids has become an effective method of diagnosing diseases. Existing secretory protein prediction methods are mainly based on conventional machine learning algorithms and are highly dependent on the feature set from the protein. In this article, we propose a deep learning model based on the capsule network and transformer architecture, SecProCT, to predict secretory proteins using only amino acid sequences. The proposed model was validated using cross-validation and achieved 0.921 and 0.892 accuracy for predicting blood-secretory proteins and saliva-secretory proteins, respectively. Meanwhile, the proposed model was validated on an independent test set and achieved 0.917 and 0.905 accuracy for predicting blood-secretory proteins and saliva-secretory proteins, respectively, which are better than conventional machine learning methods and other deep learning methods for biological sequence analysis. The main contributions of this article are as follows: (1) a deep learning model based on a capsule network and transformer architecture is proposed for predicting secretory proteins. The results of this model are better than the those of existing conventional machine learning methods and deep learning methods for biological sequence analysis; (2) only amino acid sequences are used in the proposed model, which overcomes the high dependence of existing methods on the annotated protein features; (3) the proposed model can accurately predict most experimentally verified secretory proteins and cancer protein biomarkers in blood and saliva.

DeepBSP-a Machine Learning Method for Accurate Prediction of Protein-Ligand Docking Structures.

In recent years, machine-learning-based scoring functions have significantly improved the scoring power. However, many of these methods do not perform well in distinguishing the native structure from docked decoy poses due to the lack of decoy structural information in their training data. Here, we developed a machine-learning model, named DeepBSP, that can directly predict the root mean square deviation (rmsd) of a ligand docking pose with reference to its native binding pose. Unlike the binding affinity, the rmsd between the docking poses with reference to their native structures can be straightforwardly determined. By training on a generated data set with 11,925 native complexes and more than 165,000 docked poses, our model shows excellent docking power on our test set and also on the CASF-2016 docking decoy set compared to other major scoring functions. Thus, by combining molecular dockings that generate many poses with the application of DeepBSP, one can more accurately predict the best binding pose that is closest to the native complex structure. This DeepBSP model shall be very useful in picking out poses close to their natives from many poses generated from a dock application.

Bingham deep neural and oppositional fish swarm optimized protein structure prediction

It is familiar that essential proteins take part in managing cellular activities in living organisms. Moreover, protein structure prediction from its amino acid sequence is advantageous to the comprehending of cellular functions. Formerly, several essential protein prediction methods have been proposed. However, those existing prediction methods were not satisfactory because to low sensitivity to imbalance characteristics. To address this issue, this paper presents a novel secondary protein structure prediction method, called, Bingham Deep Convolutional-based Oppositional Artificial Fish Optimized (BDC-OAFO). First, a protein structure identification framework, called, Bingham Distributed Deep Convolutional (BDDC) is designed to identify the essential proteins by eliminating the imbalanced learning issue. Next, secondary structure prediction framework, called, Oppositional Artificial Fish Swarm Optimization is proposed that obtain precise prediction results. Then, predicting secondary protein structure by emulating three biological behaviors of artificial fishes, including foraging behavior, following behavior, swarming behavior in which process, proximal count, oppositional function and Gaussian function are utilized. To evaluate the performance of BDC-OAFO method, we conduct experiments on Protein Data Bank dataset the experimental results show that our method BDC-OAFO achieves a better performance for identifying essential proteins and precise prediction in comparison with several other well-known prediction methods, which confirms the significance of BDC-OAFO. Communicated by Ramaswamy H. Sarma

RecSNO: Prediction of Protein S-Nitrosylation Sites Using a Recurrent Neural Network

S-Nitrosylation modification is one of the most important post-translational modifications; it plays a critical role in a vast variety of biological processes and is related to various diseases. Identification of S-Nitrosylation sites in proteins is crucial for understanding and controlling basic biological processes. The conventional experimental identification methods are laborious and cost in-efficient. To overcome these issues, computational biological approaches are under consideration, including use of machine learning and deep learning algorithms. All existing S-Nitrosylation predictors use the handicraft feature extraction method and could be improved upon. We propose an end-to-end deep learning based S-Nitrosylation site predictor with an embedded layer and bidirectional long short-term memory. The proposed method uses protein sequences as inputs without any need for complex features interventions. This sequence-based protein prediction method is associated with a significant improvement in identification of S-Nitrosylation sites. More specifically, the best prediction of the proposed architecture showed an improvement of in MCC 3% on 5-fold cross validation and 5% on an independent test dataset. Finally, the user-friendly publicly available webserver is accessible at http://nsclbio.jbnu.ac.kr/tools/RecSNO/.

Improving Multi-type Gram-negative Bacterial Secreted Protein Prediction via Protein Evolutionary Information and Feature Ranking

Background: Gram-negative bacteria interact with their environment by secreting a wide range of particular substrates (such as proteins) across two lipid bilayers from the cytoplasm to the extracellular space. Determining the types of secreted proteins is beneficial for further research on secreted proteins and secretion systems. Objective: As an essential alternative for experimental methods, an accurate machine learningbased multi-type Gram-negative bacterial secreted protein prediction method was proposed in this study. Methods: The main contribution is combining auto-cross-correlation analysis and feature ranking technology to build an effective support vector machine-based multi-type Gram-negative bacterial secreted protein predictor. The specifically designed auto-cross-correlation descriptor can capture evolutionary correlation information between amino acid pairs along protein sequence from position specific scoring matrices. Feature ranking technique was used to analyze and select the most informative features for building prediction model. Results: Several kinds of prediction accuracies obtained by independent dataset test are reported on two benchmark datasets. Compared with the state-of-the-art prediction methods, the proposed method improves overall accuracies by 2.91% and 2.25%, respectively. Conclusion: Our study will provide an important guide to utilize protein evolutionary information for further research on bacterial secreted proteins.

Prediction of human phosphorylated proteins by extracting multi-perspective discriminative features from the evolutionary profile and physicochemical properties through LFDA

Protein phosphorylation is an emerging post-translational modification, which critically involved in the intracellular process of the human body by controlling diverse functions ranging from cell growth to metabolism. The existing experimental methods for identifying phosphorylated proteins are overpriced and resource-intensive; thus, it is necessary to develop a fast and accurate computational method to address the problem. Here we report a novel predictor HPhosPPred, a phosphorylated protein prediction method that is incorporating highly discriminative evolutionary and physicochemical information conserved in protein primary motifs, namely pseudo-position specific scoring matrix, the auto-covariance transformation of the position-specific scoring matrix and normalized moreau-broto auto-correlation. Further, to boost up the generalization capability of HPhosPPred, we used local fisher discriminant analysis as a dominant feature selection strategy for eliminating redundant and noise patterns from the extracted features. Finally, the optimized features feed to support vector machine with radial basis function kernel to predict phosphorylated proteins. As evident from the results, the proposed method achieved promising performance with an accuracy of 80.68%, sensitivity of 84.63%, specificity of 73.67%, and Matthew’s correlation coefficient of 0.581 using rigorous leave-one-out-cross-validation test and 10-fold cross-validation test. The empirical outcomes demonstrate that the developed model outperformed the existing state-of-the-art methods. Furthermore, our analysis reveals that the proposed tool can help detect unseen phosphorylated proteins in particular and proteomics research in general. The source code and dataset are publicly available at https://github.com/saeed344/HPhosPPred.

A Method for Prediction of Thermophilic Protein Based on Reduced Amino Acids and Mixed Features.

The thermostability of proteins is a key factor considered during enzyme engineering, and finding a method that can identify thermophilic and non-thermophilic proteins will be helpful for enzyme design. In this study, we established a novel method combining mixed features and machine learning to achieve this recognition task. In this method, an amino acid reduction scheme was adopted to recode the amino acid sequence. Then, the physicochemical characteristics, auto-cross covariance (ACC), and reduced dipeptides were calculated and integrated to form a mixed feature set, which was processed using correlation analysis, feature selection, and principal component analysis (PCA) to remove redundant information. Finally, four machine learning methods and a dataset containing 500 random observations out of 915 thermophilic proteins and 500 random samples out of 793 non-thermophilic proteins were used to train and predict the data. The experimental results showed that 98.2% of thermophilic and non-thermophilic proteins were correctly identified using 10-fold cross-validation. Moreover, our analysis of the final reserved features and removed features yielded information about the crucial, unimportant and insensitive elements, it also provided essential information for enzyme design.

MsDBP: Exploring DNA-Binding Proteins by Integrating Multiscale Sequence Information via Chou's Five-Step Rule.

DNA-binding proteins are crucial to alternative splicing, methylation, and the structural composition of the DNA. The existing experimental methods for identifying DNA-binding proteins are expensive and time-consuming; thus, it is necessary to develop a fast and accurate computational method to address the problem. In this Article, we report a novel predictor MsDBP, a DNA-binding protein prediction method that combines the multiscale sequence feature into a deep neural network. First of all, instead of developing a narrow-application structured-based method, we are committed to a sequenced-based predictor. Second, instead of characterizing the whole protein directly, we divide the protein into subsequences with different lengths and then encode them into a vector based on composition information. In this way, the multiscale sequence feature can be obtained. Finally, a branch of dense layers is applied for learning multilevel abstract features to discriminate DNA-binding proteins. When MsDBP is tested on the independent data set PDB2272, it achieves an overall accuracy of 66.99% with the SE of 70.69%. In addition, we also perform extensive experiments to compare the proposed method with other existing methods. The results indicate that MsDBP would be a useful tool for the identification of DNA-binding proteins. MsDBP is freely available at a web server on http://47.100.203.218/MsDBP/ .

Inaccurate secondary structure predictions often indicate protein fold switching.

Although most proteins conform to the classical one‐structure/one‐function paradigm, an increasing number of proteins with dual structures and functions have been discovered. In response to cellular stimuli, such proteins undergo structural changes sufficiently dramatic to remodel even their secondary structures and domain organization. This “fold‐switching” capability fosters protein multi‐functionality, enabling cells to establish tight control over various biochemical processes. Accurate predictions of fold‐switching proteins could both suggest underlying mechanisms for uncharacterized biological processes and reveal potential drug targets. Recently, we developed a prediction method for fold‐switching proteins using structure‐based thermodynamic calculations and discrepancies between predicted and experimentally determined protein secondary structure (Porter and Looger, Proc Natl Acad Sci U S A 2018; 115:5968–5973). Here we seek to leverage the negative information found in these secondary structure prediction discrepancies. To do this, we quantified secondary structure prediction accuracies of 192 known fold‐switching regions (FSRs) within solved protein structures found in the Protein Data Bank (PDB). We find that the secondary structure prediction accuracies for these FSRs vary widely. Inaccurate secondary structure predictions are strongly associated with fold‐switching proteins compared to equally long segments of non‐fold‐switching proteins selected at random. These inaccurate predictions are enriched in helix‐to‐strand and strand‐to‐coil discrepancies. Finally, we find that most proteins with inaccurate secondary structure predictions are underrepresented in the PDB compared with their alternatively folded cognates, suggesting that unequal representation of fold‐switching conformers within the PDB could be an important cause of inaccurate secondary structure predictions. These results demonstrate that inconsistent secondary structure predictions can serve as a useful preliminary marker of fold switching.

Protein folding prediction in the HP model using ions motion optimization with a greedy algorithm

BackgroundThe function of a protein is determined by its native protein structure. Among many protein prediction methods, the Hydrophobic-Polar (HP) model, an ab initio method, simplifies the protein folding prediction process in order to reduce the prediction complexity.ResultsIn this study, the ions motion optimization (IMO) algorithm was combined with the greedy algorithm (namely IMOG) and implemented to the HP model for the protein folding prediction based on the 2D-triangular-lattice model. Prediction results showed that the integration method IMOG provided a better prediction efficiency in a HP model. Compared to others, our proposed method turned out as superior in its prediction ability and resilience for most of the test sequences. The efficiency of the proposed method was verified by the prediction results. The global search capability and the ability to escape from the local best solution of IMO combined with a local search (greedy algorithm) to the new algorithm IMOG greatly improve the search for the best solution with reliable protein folding prediction.ConclusionOverall, the HP model integrated with IMO and a greedy algorithm as IMOG provides an improved way of protein structure prediction of high stability, high efficiency, and outstanding performance.

StackDPPred: a stacking based prediction of DNA-binding protein from sequence.

Identification of DNA-binding proteins from only sequence information is one of the most challenging problems in the field of genome annotation. DNA-binding proteins play an important role in various biological processes such as DNA replication, repair, transcription and splicing. Existing experimental techniques for identifying DNA-binding proteins are time-consuming and expensive. Thus, prediction of DNA-binding proteins from sequences alone using computational methods can be useful to quickly annotate and guide the experimental process. Most of the methods developed for predicting DNA-binding proteins use the information from the evolutionary profile, called the position-specific scoring matrix (PSSM) profile, alone and the accuracies of such methods have been limited. Here, we propose a method, called StackDPPred, which utilizes features extracted from PSSM and residue specific contact-energy to help train a stacking based machine learning method for the effective prediction of DNA-binding proteins. Based on benchmark sequences of 1063 (518 DNA-binding and 545 non DNA-binding) proteins and using jackknife validation, StackDPPred achieved an ACC of 89.96%, MCC of 0.799 and AUC of 94.50%. This outcome outperforms several state-of-the-art approaches. Furthermore, when tested on recently designed two independent test datasets, StackDPPred outperforms existing approaches consistently. The proposed StackDPPred can be used for effective prediction of DNA-binding proteins from sequence alone. Online server is at http://bmll.cs.uno.edu/add and code-data is at http://cs.uno.edu/∼tamjid/Software/StackDPPred/code_data.zip. Supplementary data are available at Bioinformatics online.

Prediction of Protein Configurational Entropy (Popcoen).

A knowledge-based method for configurational entropy prediction of proteins is presented; this methodology is extremely fast, compared to previous approaches, because it does not involve any type of configurational sampling. Instead, the configurational entropy of a query fold is estimated by evaluating an artificial neural network, which was trained on molecular-dynamics simulations of ∼1000 proteins. The predicted entropy can be incorporated into a large class of protein software based on cost-function minimization/evaluation, in which configurational entropy is currently neglected for performance reasons. Software of this type is used for all major protein tasks such as structure predictions, proteins design, NMR and X-ray refinement, docking, and mutation effect predictions. Integrating the predicted entropy can yield a significant accuracy increase as we show exemplarily for native-state identification with the prominent protein software FoldX. The method has been termed Popcoen for Prediction of Protein Configurational Entropy. An implementation is freely available at http://fmc.ub.edu/popcoen/ .