Identification of Multi-Functional Enzyme with Multi-Label Classifier

PredictEFC: a fast and efficient multi-label classifier for predicting enzyme family classes

BACKGROUND

Enzymes play an irreplaceable and important role in maintaining the lives of living organisms. The Enzyme Commission (EC) number of an enzyme indicates its essential functions. Correct identification of the first digit (family class) of the EC number for a given enzyme is a hot topic in the past twenty years. Several previous methods adopted functional domain composition to represent enzymes. However, it would lead to dimension disaster, thereby reducing the efficiency of the methods. On the other hand, most previous methods can only deal with enzymes belonging to one family class. In fact, several enzymes belong to two or more family classes.

RESULTS

In this study, a fast and efficient multi-label classifier, named PredictEFC, was designed. To construct this classifier, a novel feature extraction scheme was designed for processing functional domain information of enzymes, which counting the distribution of each functional domain entry across seven family classes in the training dataset. Based on this scheme, each training or test enzyme was encoded into a 7-dimenion vector by fusing its functional domain information and above statistical results. Random k-labelsets (RAKEL) was adopted to build the classifier, where random forest was selected as the base classification algorithm. The two tenfold cross-validation results on the training dataset shown that the accuracy of PredictEFC can reach 0.8493 and 0.8370. The independent test on two datasets indicated the accuracy values of 0.9118 and 0.8777.

CONCLUSION

The performance of PredictEFC was slightly lower than the classifier directly using functional domain composition. However, its efficiency was sharply improved. The running time was less than one-tenth of the time of the classifier directly using functional domain composition. In additional, the utility of PredictEFC was superior to the classifiers using traditional dimensionality reduction methods and some previous methods, and this classifier can be transplanted for predicting enzyme family classes of other species. Finally, a web-server available at http://124.221.158.221/ was set up for easy usage.

Enzyme Substrate Prediction from Three-Dimensional Feature Representations Using Space-Filling Curves

Compact and interpretable structural feature representations are required for accurately predicting properties and function of proteins. In this work, we construct and evaluate three-dimensional feature representations of protein structures based on space-filling curves (SFCs). We focus on the problem of enzyme substrate prediction, using two ubiquitous enzyme families as case studies: the short-chain dehydrogenase/reductases (SDRs) and the S-adenosylmethionine-dependent methyltransferases (SAM-MTases). Space-filling curves such as the Hilbert curve and the Morton curve generate a reversible mapping from discretized three-dimensional to one-dimensional representations and thus help to encode three-dimensional molecular structures in a system-independent way and with only a few adjustable parameters. Using three-dimensional structures of SDRs and SAM-MTases generated using AlphaFold2, we assess the performance of the SFC-based feature representations in predictions on a new benchmark database of enzyme classification tasks including their cofactor and substrate selectivity. Gradient-boosted tree classifiers yield binary prediction accuracy of 0.77-0.91 and area under curve (AUC) characteristics of 0.83-0.92 for the classification tasks. We investigate the effects of amino acid encoding, spatial orientation, and (the few) parameters of SFC-based encodings on the accuracy of the predictions. Our results suggest that geometry-based approaches such as SFCs are promising for generating protein structural representations and are complementary to the existing protein feature representations such as evolutionary scale modeling (ESM) sequence embeddings.

Machine learning discovery of missing links that mediate alternative branches to plant alkaloids

Engineering the microbial production of secondary metabolites is limited by the known reactions of correctly annotated enzymes. Therefore, the machine learning discovery of specialized enzymes offers great potential to expand the range of biosynthesis pathways. Benzylisoquinoline alkaloid production is a model example of metabolic engineering with potential to revolutionize the paradigm of sustainable biomanufacturing. Existing bacterial studies utilize a norlaudanosoline pathway, whereas plants contain a more stable norcoclaurine pathway, which is exploited in yeast. However, committed aromatic precursors are still produced using microbial enzymes that remain elusive in plants, and additional downstream missing links remain hidden within highly duplicated plant gene families. In the current study, machine learning is applied to predict and select plant missing link enzymes from homologous candidate sequences. Metabolomics-based characterization of the selected sequences reveals potential aromatic acetaldehyde synthases and phenylpyruvate decarboxylases in reconstructed plant gene-only benzylisoquinoline alkaloid pathways from tyrosine. Synergistic application of the aryl acetaldehyde producing enzymes results in enhanced benzylisoquinoline alkaloid production through hybrid norcoclaurine and norlaudanosoline pathways.

Producing plant secondary metabolites by microbes is limited by the known enzymatic reactions. Here, the authors apply machine learning to predict missing link enzymes of benzylisoquinoline alkaloid (BIA) biosynthesis in Papaver somniferum, and validate the specialized activities through heterologous production.

Antibody Class(es) Predictor for Epitopes (AbCPE): A Multi-Label Classification Algorithm

Development of vaccines and therapeutic antibodies to deal with infectious and other diseases are the most perceptible scientific interventions that have had huge impact on public health including that in the current Covid-19 pandemic. From inactivation methodologies to reverse vaccinology, vaccine development strategies of 21st century have undergone several transformations and are moving towards rational design approaches. These developments are driven by data as the combinatorials involved in antigenic diversity of pathogens and immune repertoire of hosts are enormous. The computational prediction of epitopes is central to these developments and numerous B-cell epitope prediction methods developed over the years in the field of immunoinformatics have contributed enormously. Most of these methods predict epitopes that could potentially bind to an antibody regardless of its type and only a few account for antibody class specific epitope prediction. Recent studies have provided evidence of more than one class of antibodies being associated with a particular disease. Therefore, it is desirable to predict and prioritize ‘peptidome’ representing B-cell epitopes that can potentially bind to multiple classes of antibodies, as an open problem in immunoinformatics. To address this, AbCPE, a novel algorithm based on multi-label classification approach has been developed for prediction of antibody class(es) to which an epitope can potentially bind. The epitopes binding to one or more antibody classes (IgG, IgE, IgA and IgM) have been used as a knowledgebase to derive features for prediction. Multi-label algorithms, Binary Relevance and Label Powerset were applied along with Random Forest and AdaBoost. Classifier performance was assessed using evaluation measures like Hamming Loss, Precision, Recall and F1 score. The Binary Relevance model based on dipeptide composition, Random Forest and AdaBoost achieved the best results with Hamming Loss of 0.1121 and 0.1074 on training and test sets respectively. The results obtained by AbCPE are promising. To the best of our knowledge, this is the first multi-label method developed for prediction of antibody class(es) for sequential B-cell epitopes and is expected to bring a paradigm shift in the field of immunoinformatics and immunotherapeutic developments in synthetic biology. The AbCPE web server is available at http://bioinfo.unipune.ac.in/AbCPE/Home.html.

Machine learning differentiates enzymatic and non-enzymatic metals in proteins

Metalloenzymes are 40% of all enzymes and can perform all seven classes of enzyme reactions. Because of the physicochemical similarities between the active sites of metalloenzymes and inactive metal binding sites, it is challenging to differentiate between them. Yet distinguishing these two classes is critical for the identification of both native and designed enzymes. Because of similarities between catalytic and non-catalytic  metal binding sites, finding physicochemical features that distinguish these two types of metal sites can indicate aspects that are critical to enzyme function. In this work, we develop the largest structural dataset of enzymatic and non-enzymatic metalloprotein sites to date. We then use a decision-tree ensemble machine learning model to classify metals bound to proteins as enzymatic or non-enzymatic with 92.2% precision and 90.1% recall. Our model scores electrostatic and pocket lining features as more important than pocket volume, despite the fact that volume is the most quantitatively different feature between enzyme and non-enzymatic sites. Finally, we find our model has overall better performance in a side-to-side comparison against other methods that differentiate enzymatic from non-enzymatic sequences. We anticipate that our model's ability to correctly identify which metal sites are responsible for enzymatic activity could enable identification of new enzymatic mechanisms and de novo enzyme design.

A hierarchical deep learning based approach for multi-functional enzyme classification

Enzymes are critical proteins in every organism. They speed up essential chemical reactions, help fight diseases, and have a wide use in the pharmaceutical and manufacturing industries. Wet lab experiments to figure out an enzyme's function are time consuming and expensive. Therefore, the need for computational approaches to address this problem are becoming necessary. Usually, an enzyme is extremely specific in performing its function. However, there exist enzymes that can perform multiple functions. A multi-functional enzyme has vast potential as it reduces the need to discover/use different enzymes for different functions. We propose an approach to predict a multi-functional enzyme's function up to the most specific fourth level of the hierarchy of the Enzyme Commission (EC) number. Previous studies can only predict the function of the enzyme till level 1. Using a dataset of 2,583 multi-functional enzymes, we achieved a hierarchical subset accuracy of 71.4% and a Macro F₁ Score of 96.1% at the fourth level. The robustness of the network was further tested on a multi-functional isoforms dataset. Our method is broadly applicable and may be used to discover better enzymes. The web-server can be freely accessed at http://hecnet.cbrlab.org/.

Machine learning for enzyme engineering, selection and design

Machine learning is a useful computational tool for large and complex tasks such as those in the field of enzyme engineering, selection and design. In this review, we examine enzyme-related applications of machine learning. We start by comparing tools that can identify the function of an enzyme and the site responsible for that function. Then we detail methods for optimizing important experimental properties, such as the enzyme environment and enzyme reactants. We describe recent advances in enzyme systems design and enzyme design itself. Throughout we compare and contrast the data and algorithms used for these tasks to illustrate how the algorithms and data can be best used by future designers.

Prediction of oxidoreductase subfamily classes based on RFE-SND-CC-PSSM and machine learning methods

Oxidoreductase is an enzyme that widely exists in organisms. It plays an important role in cellular energy metabolism and biotransformation processes. Oxidoreductases have many subclasses with different functions, creating an important classification task in bioinformatics. In this paper, a dataset of 2640 oxidoreductase sequences was used to perform an analysis and comparison. The idea of dipeptides was introduced to process the Position Specific Score Matrix (PSSM), since each dipeptide consists of two amino acids and each column of PSSM corresponds to the information of one amino acid. Two kinds of dipeptide scores were proposed, the Standardization Normal Distribution PSSM (SND-PSSM) and the Correlation Coefficient PSSM (CC-PSSM). Recursive Feature Elimination (RFE) is used to extract features from the SND-PSSM and CC-PSSM, and the two sets of extracted features are combined to form a new feature matrix, the RFE-SND-CC-PSSM. The results show that, with the proposed method and a kernel-based nonlinear SVM classifier, the accuracy can reach 95.56% by the Jackknife test. Our method greatly improves the accuracy of oxidoreductase subclass prediction. Using this method to predict the categories of the 6 major types of enzymes effectively improves its prediction accuracy to 94.54%, indicating that this method has general applicability to other protein problems. The results show that our method is effective and universally applicable, and might be complementary to the existing methods.

Bio2Rxn: sequence-based enzymatic reaction predictions by a consensus strategy

Summary

The development of sequencing technologies has generated large amounts of protein sequence data. The automated prediction of the enzymatic reactions of uncharacterized proteins is a major challenge in the field of bioinformatics. Here, we present Bio2Rxn as a web-based tool to provide putative enzymatic reaction predictions for uncharacterized protein sequences. Bio2Rxn adopts a consensus strategy by incorporating six types of enzyme prediction tools. It allows for the efficient integration of these computational resources to maximize the accuracy and comprehensiveness of enzymatic reaction predictions, which facilitates the characterization of the functional roles of target proteins in metabolism. Bio2Rxn further links the enzyme function prediction with more than 300 000 enzymatic reactions, which were manually curated by more than 100 people over the past 9 years from more than 580 000 publications.

Availability and implementation

Bio2Rxn is available at: http://design.rxnfinder.org/bio2rxn/.

Contact

qnhu@sibs.ac.cn

Supplementary information

Supplementary data are available at Bioinformatics online.

Exploration and Evaluation of Machine Learning-Based Models for Predicting Enzymatic Reactions

Unannotated gene sequences in databases are increasing due to sequencing advances. Therefore, computational methods to predict functions of unannotated genes are needed. Moreover, novel enzyme discovery for metabolic engineering applications further encourages annotation of sequences. Here, enzyme functions are predicted using two general approaches, each including several machine learning algorithms. First, Enzyme-models (E-models) predict Enzyme Commission (EC) numbers from amino acid sequence information. Second, Substrate-Enzyme models (SE-models) are built to predict substrates of enzymatic reactions together with EC numbers, and Substrate-Enzyme-Product models (SEP-models) are built to predict substrates, products, and EC numbers. While accuracy of E-models is not optimal, SE-models and SEP-models predict EC numbers and reactions with high accuracy using all tested machine learning-based methods. For example, a single Random Forests-based SEP-model predicts EC first digits with an Average AUC score of over 0.94. Various metrics indicate that the current strategy of combining sequence and chemical structure information is effective at improving enzyme reaction prediction.

Alignment-Free Method to Predict Enzyme Classes and Subclasses

The Enzyme Classification (EC) number is a numerical classification scheme for enzymes, established using the chemical reactions they catalyze. This classification is based on the recommendation of the Nomenclature Committee of the International Union of Biochemistry and Molecular Biology. Six enzyme classes were recognised in the first Enzyme Classification and Nomenclature List, reported by the International Union of Biochemistry in 1961. However, a new enzyme group was recently added as the six existing EC classes could not describe enzymes involved in the movement of ions or molecules across membranes. Such enzymes are now classified in the new EC class of translocases (EC 7). Several computational methods have been developed in order to predict the EC number. However, due to this new change, all such methods are now outdated and need updating. In this work, we developed a new multi-task quantitative structure-activity relationship (QSAR) method aimed at predicting all 7 EC classes and subclasses. In so doing, we developed an alignment-free model based on artificial neural networks that proved to be very successful.

mlDEEPre: Multi-Functional Enzyme Function Prediction With Hierarchical Multi-Label Deep Learning

As a great challenge in bioinformatics, enzyme function prediction is a significant step toward designing novel enzymes and diagnosing enzyme-related diseases. Existing studies mainly focus on the mono-functional enzyme function prediction. However, the number of multi-functional enzymes is growing rapidly, which requires novel computational methods to be developed. In this paper, following our previous work, DEEPre, which uses deep learning to annotate mono-functional enzyme's function, we propose a novel method, mlDEEPre, which is designed specifically for predicting the functionalities of multi-functional enzymes. By adopting a novel loss function, associated with the relationship between different labels, and a self-adapted label assigning threshold, mlDEEPre can accurately and efficiently perform multi-functional enzyme prediction. Extensive experiments also show that mlDEEPre can outperform the other methods in predicting whether an enzyme is a mono-functional or a multi-functional enzyme (mono-functional vs. multi-functional), as well as the main class prediction across different criteria. Furthermore, due to the flexibility of mlDEEPre and DEEPre, mlDEEPre can be incorporated into DEEPre seamlessly, which enables the updated DEEPre to handle both mono-functional and multi-functional predictions without human intervention.

iNuc-ext-PseTNC: an efficient ensemble model for identification of nucleosome positioning by extending the concept of Chou's PseAAC to pseudo-tri-nucleotide composition

Nucleosome is a central element of eukaryotic chromatin, which composes of histone proteins and DNA molecules. It performs vital roles in many eukaryotic intra-nuclear processes, for instance, chromatin structure and transcriptional regulation formation. Identification of nucleosome positioning via wet lab is difficult; so, the attention is diverted towards the accurate intelligent automated prediction. In this regard, a novel intelligent automated model "iNuc-ext-PseTNC" is developed to identify the nucleosome positioning in genomes accurately. In this predictor, the sequences of DNA are mathematically represented by two different discrete feature extraction techniques, namely pseudo-tri-nucleotide composition (PseTNC) and pseudo-di-nucleotide composition. Several contemporary machine learning algorithms were examined. Further, the predictions of individual classifiers were integrated through an evolutionary genetic algorithm. The success rates of the ensemble model are higher than individual classifiers. After analyzing the prediction results, it is noticed that iNuc-ext-PseTNC model has achieved better performance in combination with PseTNC feature space, which are 94.3%, 93.14%, and 88.60% of accuracies using six-fold cross-validation test for the three benchmark datasets S₁, S₂, and S₃, respectively. The achieved outcomes exposed that the results of iNuc-ext-PseTNC model are prominent compared to the existing methods so far notifiable in the literature. It is ascertained that the proposed model might be more fruitful and a practical tool for rudimentary academia and research.

ECPred: a tool for the prediction of the enzymatic functions of protein sequences based on the EC nomenclature

BACKGROUND

The automated prediction of the enzymatic functions of uncharacterized proteins is a crucial topic in bioinformatics. Although several methods and tools have been proposed to classify enzymes, most of these studies are limited to specific functional classes and levels of the Enzyme Commission (EC) number hierarchy. Besides, most of the previous methods incorporated only a single input feature type, which limits the applicability to the wide functional space. Here, we proposed a novel enzymatic function prediction tool, ECPred, based on ensemble of machine learning classifiers.

RESULTS

In ECPred, each EC number constituted an individual class and therefore, had an independent learning model. Enzyme vs. non-enzyme classification is incorporated into ECPred along with a hierarchical prediction approach exploiting the tree structure of the EC nomenclature. ECPred provides predictions for 858 EC numbers in total including 6 main classes, 55 subclass classes, 163 sub-subclass classes and 634 substrate classes. The proposed method is tested and compared with the state-of-the-art enzyme function prediction tools by using independent temporal hold-out and no-Pfam datasets constructed during this study.

CONCLUSIONS

ECPred is presented both as a stand-alone and a web based tool to provide probabilistic enzymatic function predictions (at all five levels of EC) for uncharacterized protein sequences. Also, the datasets of this study will be a valuable resource for future benchmarking studies. ECPred is available for download, together with all of the datasets used in this study, at: https://github.com/cansyl/ECPred . ECPred webserver can be accessed through http://cansyl.metu.edu.tr/ECPred.html .

Sequence based predictor for discrimination of enhancer and their types by applying general form of Chou's trinucleotide composition

BACKGROUND AND OBJECTIVES

Enhancers are pivotal DNA elements, which are widely used in eukaryotes for activation of transcription genes. On the basis of enhancer strength, they are further classified into two groups; strong enhancers and weak enhancers. Due to high availability of huge amount of DNA sequences, it is needed to develop fast, reliable and robust intelligent computational method, which not only identify enhancers but also determines their strength. Considerable progress has been achieved in this regard; however, timely and precisely identification of enhancers is still a challenging task.

METHODS

Two-level intelligent computational model for identification of enhancers and their subgroups is proposed. Two different feature extraction techniques including di-nucleotide composition and tri-nucleotide composition were adopted for extraction of numerical descriptors. Four classification methods including probabilistic neural network, support vector machine, k-nearest neighbor and random forest were utilized for classification.

RESULTS

The proposed method yielded 77.25% of accuracy for dataset S₁ contains enhancers and non-enhancers, whereas 64.70% of accuracy for dataset S₂ comprises of strong enhancer and weak enhancer sequences using jackknife cross-validation test.

CONCLUSION

The predictive results validated that the proposed method is better than that of existing approaches so far reported in the literature. It is thus highly observed that the developed method will be useful and expedient for basic research and academia.

Fuzzy Decision Making Approach to Identify Optimum Enzyme Targets and Drug Dosage for Remedying Presynaptic Dopamine Deficiency

Model-based optimization approaches are valuable in developing new drugs for human metabolic disorders. The core objective in most optimal drug designs is positive therapeutic effects. In this study, we considered the effects of therapeutic, adverse, and target variation simultaneously. A fuzzy optimization method was applied to formulate a multiobjective drug design problem for detecting enzyme targets in the presynaptic dopamine metabolic network to remedy two types of enzymopathies caused by deficiencies of vesicular monoamine transporter 2 (VMAT2) and tyrosine hydroxylase (TH). The fuzzy membership approach transforms a two-stage drug discovery problem into a unified decision-making problem. We developed a nested hybrid differential evolution algorithm to efficiently identify a set of potential drug targets. Furthermore, we also simulated the effects of current clinical drugs for Parkinson’s disease (PD) in this model and tried to clarify the possible causes of neurotoxic and neuroprotective effects. The optimal drug design could yield 100% satisfaction grade when both therapeutic effect and the number of targets were considered in the objective. This scenario required regulating one to three and one or two enzyme targets for 50%–95% and 50%–100% VMAT2 and TH deficiencies, respectively. However, their corresponding adverse and target variation effect grades were less satisfactory. For the most severe deficiencies of VMAT2 and TH, a compromise design could be obtained when the effects of therapeutic, adverse, and target variation were simultaneously applied to the optimal drug discovery problem. Such a trade-off design followed the no free lunch theorem for optimization; that is, a more serious dopamine deficiency required more enzyme targets and lower satisfaction grade. In addition, the therapeutic effects of current clinical medications for PD could be enhanced in combination with new enzyme targets. The increase of toxic metabolites after treatment might be the cause of neurotoxic effects of some current PD medications.