ECOH: an enzyme commission number predictor using mutual information and a support vector machine

An explainability framework for deep learning on chemical reactions exemplified by enzyme-catalysed reaction classification

Assigning or proposing a catalysing enzyme given a chemical or biochemical reaction is of great interest to life sciences and chemistry alike. The exploration and design of metabolic pathways and the challenge of finding more sustainable enzyme-catalysed alternatives to traditional organic reactions are just two examples of tasks that require an association between reaction and enzyme. However, given the lack of large and balanced annotated data sets of enzyme-catalysed reactions, assigning an enzyme to a reaction still relies on expert-curated rules and databases. Here, we present a data-driven explainable human-in-the-loop machine learning approach to support and ultimately automate the association of a catalysing enzyme with a given biochemical reaction. In addition, the proposed method is capable of predicting enzymes as candidate catalysts for organic reactions amendable to biocatalysis. Finally, the introduced explainability and visualisation methods can easily be generalised to support other machine-learning approaches involving chemical and biochemical reactions.

Machine Learning-Guided Protein Engineering

Recent progress in engineering highly promising biocatalysts has increasingly involved machine learning methods. These methods leverage existing experimental and simulation data to aid in the discovery and annotation of promising enzymes, as well as in suggesting beneficial mutations for improving known targets. The field of machine learning for protein engineering is gathering steam, driven by recent success stories and notable progress in other areas. It already encompasses ambitious tasks such as understanding and predicting protein structure and function, catalytic efficiency, enantioselectivity, protein dynamics, stability, solubility, aggregation, and more. Nonetheless, the field is still evolving, with many challenges to overcome and questions to address. In this Perspective, we provide an overview of ongoing trends in this domain, highlight recent case studies, and examine the current limitations of machine learning-based methods. We emphasize the crucial importance of thorough experimental validation of emerging models before their use for rational protein design. We present our opinions on the fundamental problems and outline the potential directions for future research.

Predicting enzymatic function of protein sequences with attention

MOTIVATION

There is a growing number of available protein sequences, but only a limited amount has been manually annotated. For example, only 0.25% of all entries of UniProtKB are reviewed by human annotators. Further developing automatic tools to infer protein function from sequence alone can alleviate part of this gap. In this article, we investigate the potential of Transformer deep neural networks on a specific case of functional sequence annotation: the prediction of enzymatic classes.

RESULTS

We show that our EnzBert transformer models, trained to predict Enzyme Commission (EC) numbers by specialization of a protein language model, outperforms state-of-the-art tools for monofunctional enzyme class prediction based on sequences only. Accuracy is improved from 84% to 95% on the prediction of EC numbers at level two on the EC40 benchmark. To evaluate the prediction quality at level four, the most detailed level of EC numbers, we built two new time-based benchmarks for comparison with state-of-the-art methods ECPred and DeepEC: the macro-F1 score is respectively improved from 41% to 54% and from 20% to 26%. Finally, we also show that using a simple combination of attention maps is on par with, or better than, other classical interpretability methods on the EC prediction task. More specifically, important residues identified by attention maps tend to correspond to known catalytic sites. Quantitatively, we report a max F-Gain score of 96.05%, while classical interpretability methods reach 91.44% at best.

AVAILABILITY AND IMPLEMENTATION

Source code and datasets are respectively available at https://gitlab.inria.fr/nbuton/tfpc and https://doi.org/10.5281/zenodo.7253910.

Bioinformatics approaches and big data analytics opportunities in improving fisheries and aquaculture

Aquaculture has witnessed an excellent growth rate during the last two decades and offers huge potential to provide nutritional as well as livelihood security. Genomic research has contributed significantly toward the development of beneficial technologies for aquaculture. The existing high throughput technologies like next-generation technologies generate oceanic data which requires extensive analysis using appropriate tools. Bioinformatics is a rapidly evolving science that involves integrating gene based information and computational technology to produce new knowledge for the benefit of aquaculture. Bioinformatics provides new opportunities as well as challenges for information and data processing in new generation aquaculture. Rapid technical advancements have opened up a world of possibilities for using current genomics to improve aquaculture performance. Understanding the genes that govern economically relevant characteristics, necessitates a significant amount of additional research. The various dimensions of data sources includes next-generation DNA sequencing, protein sequencing, RNA sequencing gene expression profiles, metabolic pathways, molecular markers, and so on. Appropriate bioinformatics tools are developed to mine the biologically relevant and commercially useful results. The purpose of this scoping review is to present various arms of diverse bioinformatics tools with special emphasis on practical translation to the aquaculture industry.

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.

IHEC_RAAC: a online platform for identifying human enzyme classes via reduced amino acid cluster strategy

Enzymes have been proven to play considerable roles in disease diagnosis and biological functions. The feature extraction that truly reflects the intrinsic properties of protein is the most critical step for the automatic identification of enzymes. Although lots of feature extraction methods have been proposed, some challenges remain. In this study, we developed a predictor called IHEC_RAAC, which has the capability to identify whether a protein is a human enzyme and distinguish the function of the human enzyme. To improve the feature representation ability, protein sequences were encoded by a new feature-vector called 'reduced amino acid cluster'. We calculated 673 amino acid reduction alphabets to determine the optimal feature representative scheme. The tenfold cross-validation test showed that the accuracy of IHEC_RAAC to identify human enzymes was 74.66% and further discriminate the human enzyme classes with an accuracy of 54.78%, which was 2.06% and 8.68% higher than the state-of-the-art predictors, respectively. Additionally, the results from the independent dataset indicated that IHEC_RAAC can effectively predict human enzymes and human enzyme classes to further provide guidance for protein research. A user-friendly web server, IHEC_RAAC, is freely accessible at http://bioinfor.imu.edu.cn/ihecraac .

iMPTCE-Hnetwork: A Multilabel Classifier for Identifying Metabolic Pathway Types of Chemicals and Enzymes with a Heterogeneous Network

Metabolic pathway is an important type of biological pathways. It produces essential molecules and energies to maintain the life of living organisms. Each metabolic pathway consists of a chain of chemical reactions, which always need enzymes to participate in. Thus, chemicals and enzymes are two major components for each metabolic pathway. Although several metabolic pathways have been uncovered, the metabolic pathway system is still far from complete. Some hidden chemicals or enzymes are not discovered in a certain metabolic pathway. Besides the traditional experiments to detect hidden chemicals or enzymes, an alternative pipeline is to design efficient computational methods. In this study, we proposed a powerful multilabel classifier, called iMPTCE-Hnetwork, to uniformly assign chemicals and enzymes to metabolic pathway types reported in KEGG. Such classifier adopted the embedding features derived from a heterogeneous network, which defined chemicals and enzymes as nodes and the interactions between chemicals and enzymes as edges, through a powerful network embedding algorithm, Mashup. The popular RAndom k-labELsets (RAKEL) algorithm was employed to construct the classifier, which incorporated the support vector machine (polynomial kernel) as the basic classifier. The ten-fold cross-validation results indicated that such a classifier had good performance with accuracy higher than 0.800 and exact match higher than 0.750. Several comparisons were done to indicate the superiority of the iMPTCE-Hnetwork.

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.

Enzyme annotation for orphan and novel reactions using knowledge of substrate reactive sites

Recent advances in synthetic biochemistry have resulted in a wealth of novel hypothetical enzymatic reactions that are not matched to protein-encoding genes, deeming them “orphan.” A large number of known metabolic enzymes are also orphan, leaving important gaps in metabolic network maps. Proposing genes for the catalysis of orphan reactions is critical for applications ranging from biotechnology to medicine. In this work, the computational method BridgIT identified potential enzymes of orphan reactions and nearly all theoretically possible biochemical transformations, providing candidate genes to catalyze these reactions to the research community. The BridgIT online tool will allow researchers to fill the knowledge gaps in metabolic networks and will act as a starting point for designing novel enzymes to catalyze nonnatural transformations.

Thousands of biochemical reactions with characterized activities are “orphan,” meaning they cannot be assigned to a specific enzyme, leaving gaps in metabolic pathways. Novel reactions predicted by pathway-generation tools also lack associated sequences, limiting protein engineering applications. Associating orphan and novel reactions with known biochemistry and suggesting enzymes to catalyze them is a daunting problem. We propose the method BridgIT to identify candidate genes and catalyzing proteins for these reactions. This method introduces information about the enzyme binding pocket into reaction-similarity comparisons. BridgIT assesses the similarity of two reactions, one orphan and one well-characterized nonorphan reaction, using their substrate reactive sites, their surrounding structures, and the structures of the generated products to suggest enzymes that catalyze the most-similar nonorphan reactions as candidates for also catalyzing the orphan ones. We performed two large-scale validation studies to test BridgIT predictions against experimental biochemical evidence. For the 234 orphan reactions from the Kyoto Encyclopedia of Genes and Genomes (KEGG) 2011 (a comprehensive enzymatic-reaction database) that became nonorphan in KEGG 2018, BridgIT predicted the exact or a highly related enzyme for 211 of them. Moreover, for 334 of 379 novel reactions in 2014 that were later cataloged in KEGG 2018, BridgIT predicted the exact or highly similar enzymes. BridgIT requires knowledge about only four connecting bonds around the atoms of the reactive sites to correctly annotate proteins for 93% of analyzed enzymatic reactions. Increasing to seven connecting bonds allowed for the accurate identification of a sequence for nearly all known enzymatic reactions.

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 .

Multiclassification Prediction of Enzymatic Reactions for Oxidoreductases and Hydrolases Using Reaction Fingerprints and Machine Learning Methods

Drug metabolism is a complex procedure in the human body, including a series of enzymatically catalyzed reactions. However, it is costly and time consuming to investigate drug metabolism experimentally; computational methods are hence developed to predict drug metabolism and have shown great advantages. As the first step, classification of metabolic reactions and enzymes is highly desirable for drug metabolism prediction. In this study, we developed multiclassification models for prediction of reaction types catalyzed by oxidoreductases and hydrolases, in which three reaction fingerprints were used to describe the reactions and seven machine learnings algorithms were employed for model building. Data retrieved from KEGG containing 1055 hydrolysis and 2510 redox reactions were used to build the models, respectively. The external validation data consisted of 213 hydrolysis and 512 redox reactions extracted from the Rhea database. The best models were built by neural network or logistic regression with a 2048-bit transformation reaction fingerprint. The predictive accuracies of the main class, subclass, and superclass classification models on external validation sets were all above 90%. This study will be very helpful for enzymatic reaction annotation and further study on metabolism prediction.

Metabolic pathway reconstruction strategies for central metabolism and natural product biosynthesis

Metabolic pathway reconstruction presents a challenge for understanding metabolic pathways in organisms of interest. Different strategies, i.e., reference-based vs. de novo, must be used for pathway reconstruction depending on the availability of well-characterized enzymatic reactions. If at least one enzyme is already known to catalyze a reaction, its amino acid sequence can be used as a reference for identifying homologous enzymes in the genome of an organism of interest. Where there is no known enzyme able to catalyze a corresponding reaction, however, the reaction and the corresponding enzyme must be predicted de novo from chemical transformations of the putative substrate-product pair. This review summarizes studies involving reference-based and de novo metabolic pathway reconstruction and discusses the importance of the classification and structure-function relationships of enzymes.

Simultaneous prediction of enzyme orthologs from chemical transformation patterns for de novo metabolic pathway reconstruction

MOTIVATION

Metabolic pathways are an important class of molecular networks consisting of compounds, enzymes and their interactions. The understanding of global metabolic pathways is extremely important for various applications in ecology and pharmacology. However, large parts of metabolic pathways remain unknown, and most organism-specific pathways contain many missing enzymes.

RESULTS

In this study we propose a novel method to predict the enzyme orthologs that catalyze the putative reactions to facilitate the de novo reconstruction of metabolic pathways from metabolome-scale compound sets. The algorithm detects the chemical transformation patterns of substrate-product pairs using chemical graph alignments, and constructs a set of enzyme-specific classifiers to simultaneously predict all the enzyme orthologs that could catalyze the putative reactions of the substrate-product pairs in the joint learning framework. The originality of the method lies in its ability to make predictions for thousands of enzyme orthologs simultaneously, as well as its extraction of enzyme-specific chemical transformation patterns of substrate-product pairs. We demonstrate the usefulness of the proposed method by applying it to some ten thousands of metabolic compounds, and analyze the extracted chemical transformation patterns that provide insights into the characteristics and specificities of enzymes. The proposed method will open the door to both primary (central) and secondary metabolism in genomics research, increasing research productivity to tackle a wide variety of environmental and public health matters.

AVAILABILITY AND IMPLEMENTATION

CONTACT

: maskot@bio.titech.ac.jp.

Identification of Enzyme Genes Using Chemical Structure Alignments of Substrate-Product Pairs

Although there are several databases that contain data on many metabolites and reactions in biochemical pathways, there is still a big gap in the numbers between experimentally identified enzymes and metabolites. It is supposed that many catalytic enzyme genes are still unknown. Although there are previous studies that estimate the number of candidate enzyme genes, these studies required some additional information aside from the structures of metabolites such as gene expression and order in the genome. In this study, we developed a novel method to identify a candidate enzyme gene of a reaction using the chemical structures of the substrate-product pair (reactant pair). The proposed method is based on a search for similar reactant pairs in a reference database and offers ortholog groups that possibly mediate the given reaction. We applied the proposed method to two experimentally validated reactions. As a result, we confirmed that the histidine transaminase was correctly identified. Although our method could not directly identify the asparagine oxo-acid transaminase, we successfully found the paralog gene most similar to the correct enzyme gene. We also applied our method to infer candidate enzyme genes in the mesaconate pathway. The advantage of our method lies in the prediction of possible genes for orphan enzyme reactions where any associated gene sequences are not determined yet. We believe that this approach will facilitate experimental identification of genes for orphan enzymes.

Sequence-based identification of recombination spots using pseudo nucleic acid representation and recursive feature extraction by linear kernel SVM

Background

Identification of the recombination hot/cold spots is critical for understanding the mechanism of recombination as well as the genome evolution process. However, experimental identification of recombination spots is both time-consuming and costly. Developing an accurate and automated method for reliably and quickly identifying recombination spots is thus urgently needed.

Results

Here we proposed a novel approach by fusing features from pseudo nucleic acid composition (PseNAC), including NAC, n-tier NAC and pseudo dinucleotide composition (PseDNC). A recursive feature extraction by linear kernel support vector machine (SVM) was then used to rank the integrated feature vectors and extract optimal features. SVM was adopted for identifying recombination spots based on these optimal features. To evaluate the performance of the proposed method, jackknife cross-validation test was employed on a benchmark dataset. The overall accuracy of this approach was 84.09%, which was higher (from 0.37% to 3.79%) than those of state-of-the-art tools.

Conclusions

Comparison results suggested that linear kernel SVM is a useful vehicle for identifying recombination hot/cold spots.

Validation of RetroPath, a computer-aided design tool for metabolic pathway engineering

Metabolic engineering has succeeded in biosynthesis of numerous commodity or high value compounds. However, the choice of pathways and enzymes used for production was many times made ad hoc, or required expert knowledge of the specific biochemical reactions. In order to rationalize the process of engineering producer strains, we developed the computer-aided design (CAD) tool RetroPath that explores and enumerates metabolic pathways connecting the endogenous metabolites of a chassis cell to the target compound. To experimentally validate our tool, we constructed 12 top-ranked enzyme combinations producing the flavonoid pinocembrin, four of which displayed significant yields. Namely, our tool queried the enzymes found in metabolic databases based on their annotated and predicted activities. Next, it ranked pathways based on the predicted efficiency of the available enzymes, the toxicity of the intermediate metabolites and the calculated maximum product flux. To implement the top-ranking pathway, our procedure narrowed down a list of nine million possible enzyme combinations to 12, a number easily assembled and tested. One round of metabolic network optimization based on RetroPath output further increased pinocembrin titers 17-fold. In total, 12 out of the 13 enzymes tested in this work displayed a relative performance that was in accordance with its predicted score. These results validate the ranking function of our CAD tool, and open the way to its utilization in the biosynthesis of novel compounds.

A hybrid method for prediction and repositioning of drug Anatomical Therapeutic Chemical classes

In the Anatomical Therapeutic Chemical (ATC) classification system, therapeutic drugs are divided into 14 main classes according to the organ or system on which they act and their chemical, pharmacological and therapeutic properties. This system, recommended by the World Health Organization (WHO), provides a global standard for classifying medical substances and serves as a tool for international drug utilization research to improve quality of drug use. In view of this, it is necessary to develop effective computational prediction methods to identify the ATC-class of a given drug, which thereby could facilitate further analysis of this system. In this study, we initiated an attempt to develop a prediction method and to gain insights from it by utilizing ontology information of drug compounds. Since only about one-fourth of drugs in the ATC classification system have ontology information, a hybrid prediction method combining the ontology information, chemical interaction information and chemical structure information of drug compounds was proposed for the prediction of drug ATC-classes. As a result, by using the Jackknife test, the 1st prediction accuracies for identifying the 14 main ATC-classes in the training dataset, the internal validation dataset and the external validation dataset were 75.90%, 75.70% and 66.36%, respectively. Analysis of some samples with false-positive predictions in the internal and external validation datasets indicated that some of them may even have a relationship with the false-positive predicted ATC-class, suggesting novel uses of these drugs. It was conceivable that the proposed method could be used as an efficient tool to identify ATC-classes of novel drugs or to discover novel uses of known drugs.

Prediction of cancer drugs by chemical-chemical interactions

Cancer, which is a leading cause of death worldwide, places a big burden on health-care system. In this study, an order-prediction model was built to predict a series of cancer drug indications based on chemical-chemical interactions. According to the confidence scores of their interactions, the order from the most likely cancer to the least one was obtained for each query drug. The 1(st) order prediction accuracy of the training dataset was 55.93%, evaluated by Jackknife test, while it was 55.56% and 59.09% on a validation test dataset and an independent test dataset, respectively. The proposed method outperformed a popular method based on molecular descriptors. Moreover, it was verified that some drugs were effective to the 'wrong' predicted indications, indicating that some 'wrong' drug indications were actually correct indications. Encouraged by the promising results, the method may become a useful tool to the prediction of drugs indications.

Prediction of drugs target groups based on ChEBI ontology

Most drugs have beneficial as well as adverse effects and exert their biological functions by adjusting and altering the functions of their target proteins. Thus, knowledge of drugs target proteins is essential for the improvement of therapeutic effects and mitigation of undesirable side effects. In the study, we proposed a novel prediction method based on drug/compound ontology information extracted from ChEBI to identify drugs target groups from which the kind of functions of a drug may be deduced. By collecting data in KEGG, a benchmark dataset consisting of 876 drugs, categorized into four target groups, was constructed. To evaluate the method more thoroughly, the benchmark dataset was divided into a training dataset and an independent test dataset. It is observed by jackknife test that the overall prediction accuracy on the training dataset was 83.12%, while it was 87.50% on the test dataset-the predictor exhibited an excellent generalization. The good performance of the method indicates that the ontology information of the drugs contains rich information about their target groups, and the study may become an inspiration to solve the problems of this sort and bridge the gap between ChEBI ontology and drugs target groups.