Accurate Function Prediction Research Articles

Predicting Functional Outcome After Nonoperative Treatment of Proximal Humeral Fractures Involving the Surgical Neck

BackgroundIn nonoperative treated proximal humeral fractures (PHF), uncertainty remains regarding functional outcome. Therefore, the aim was to identify predictors of functional outcome following nonoperative treatment of PHF and to develop predictive models. MethodsAdults with a nonoperatively treated PHF involving the surgical neck were followed for 1 year. Radiographic parameters included fracture configuration, displacement, bone quality, and the critical shoulder angle (CSA). The neck-shaft angle (NSA) and humeral head offset assessed displacement. The greater tuberosity index (GTI) addressed greater tuberosity (GT) displacement relative to the articular surface while the impingement index addressed GT-displacement relative to the acromion. Multivariate regression models determined predictors of impaired function as measured by Constant Score (CS), flexion, and external rotation (ER). Recursive partitioning created a decision tree model over dichotomized functional outcome that combined flexion and ER (good = flexion ≥ 120° and ER ≥ 40°). ResultsIn 272 consecutive patients (mean age: 66 years, 69% women and 31% men), mean functional outcomes were a CS of 74 points, 138° flexion, and 50° ER. Older age was a predictor for impaired CS, flexion, and ER (p < 0.001), while sex only influenced CS (p = 0.040). Every 10% increase in GTI assessed on the Y-view explained a decrease of 3 CS points, 7° of flexion, and 5° of ER (p < 0.001). Every 10° of varus angulation assessed on anterior-posterior (ap) views in internal rotation resulted in a decrease of 1 CS point and 4° of flexion (p ≤ 0.004). Subgroups of combined important GT-displacement (GTI ≥ 1.15) with age > 76 years or < 119° varus angulation demonstrated the worst function. Linear prediction models estimated flexion, ER, and CS with a mean difference of 10°, 3°, and 3 points, respectively. The decision tree model predicted good function with 80% accuracy (positive predictive value = 81%; negative predictive value = 78%). ConclusionDemographic and radiographic predictors were identified allowing for accurate functional prognosis of nonoperatively treated PHF involving the surgical neck. A combined assessment of the ap-view in internal rotation and Y-view is sufficient for accurate function prediction. The identified subgroups resulting in good or impaired function, and the predictive models may be useful for patient counseling and guidance of treatment-related expectations.

StarFunc: Accurate Protein Function Prediction Reveals Novel Human Proteins Involved in Ubiquitination

StarFunc: Accurate Protein Function Prediction Reveals Novel Human Proteins Involved in Ubiquitination

Environmental activity-based protein profiling for function-driven enzyme discovery from natural communities

BackgroundMicrobial communities are important drivers of global biogeochemical cycles, xenobiotic detoxification, as well as organic matter decomposition. Their major metabolic role in ecosystem functioning is ensured by a unique set of enzymes, providing a tremendous yet mostly hidden enzymatic potential. Exploring this enzymatic repertoire is therefore not only relevant for a better understanding of how microorganisms function in their natural environment, and thus for ecological research, but further turns microbial communities, in particular from extreme habitats, into a valuable resource for the discovery of novel enzymes with potential applications in biotechnology. Different strategies for their uncovering such as bioprospecting, which relies mainly on metagenomic approaches in combination with sequence-based bioinformatic analyses, have emerged; yet accurate function prediction of their proteomes and deciphering the in vivo activity of an enzyme remains challenging.ResultsHere, we present environmental activity-based protein profiling (eABPP), a multi-omics approach that extends genome-resolved metagenomics with mass spectrometry-based ABPP. This combination allows direct profiling of environmental community samples in their native habitat and the identification of active enzymes based on their function, even without sequence or structural homologies to annotated enzyme families. eABPP thus bridges the gap between environmental genomics, correct function annotation, and in vivo enzyme activity. As a showcase, we report the successful identification of active thermostable serine hydrolases from eABPP of natural microbial communities from two independent hot springs in Kamchatka, Russia.ConclusionsBy reporting enzyme activities within an ecosystem in their native state, we anticipate that eABPP will not only advance current methodological approaches to sequence homology-guided enzyme discovery from environmental ecosystems for subsequent biocatalyst development but also contributes to the ecological investigation of microbial community interactions by dissecting their underlying molecular mechanisms.

Leveraging conformal prediction to annotate enzyme function space with limited false positives.

Machine learning (ML) is increasingly being used to guide biological discovery in biomedicine such as prioritizing promising small molecules in drug discovery. In those applications, ML models are used to predict the properties of biological systems, and researchers use these predictions to prioritize candidates as new biological hypotheses for downstream experimental validations. However, when applied to unseen situations, these models can be overconfident and produce a large number of false positives. One solution to address this issue is to quantify the model's prediction uncertainty and provide a set of hypotheses with a controlled false discovery rate (FDR) pre-specified by researchers. We propose CPEC, an ML framework for FDR-controlled biological discovery. We demonstrate its effectiveness using enzyme function annotation as a case study, simulating the discovery process of identifying the functions of less-characterized enzymes. CPEC integrates a deep learning model with a statistical tool known as conformal prediction, providing accurate and FDR-controlled function predictions for a given protein enzyme. Conformal prediction provides rigorous statistical guarantees to the predictive model and ensures that the expected FDR will not exceed a user-specified level with high probability. Evaluation experiments show that CPEC achieves reliable FDR control, better or comparable prediction performance at a lower FDR than existing methods, and accurate predictions for enzymes under-represented in the training data. We expect CPEC to be a useful tool for biological discovery applications where a high yield rate in validation experiments is desired but the experimental budget is limited.

Prot2Text: Multimodal Protein’s Function Generation with GNNs and Transformers

In recent years, significant progress has been made in the field of protein function prediction with the development of various machine-learning approaches. However, most existing methods formulate the task as a multi-classification problem, i.e. assigning predefined labels to proteins. In this work, we propose a novel approach, Prot2Text, which predicts a protein's function in a free text style, moving beyond the conventional binary or categorical classifications. By combining Graph Neural Networks(GNNs) and Large Language Models(LLMs), in an encoder-decoder framework, our model effectively integrates diverse data types including protein sequence, structure, and textual annotation and description. This multimodal approach allows for a holistic representation of proteins' functions, enabling the generation of detailed and accurate functional descriptions. To evaluate our model, we extracted a multimodal protein dataset from SwissProt, and demonstrate empirically the effectiveness of Prot2Text. These results highlight the transformative impact of multimodal models, specifically the fusion of GNNs and LLMs, empowering researchers with powerful tools for more accurate function prediction of existing as well as first-to-see proteins.

Fast and accurate protein function prediction from sequence through pretrained language model and homology-based label diffusion.

Protein function prediction is an essential task in bioinformatics which benefits disease mechanism elucidation and drug target discovery. Due to the explosive growth of proteins in sequence databases and the diversity of their functions, it remains challenging to fast and accurately predict protein functions from sequences alone. Although many methods have integrated protein structures, biological networks or literature information to improve performance, these extra features are often unavailable for most proteins. Here, we propose SPROF-GO, a Sequence-based alignment-free PROtein Function predictor, which leverages a pretrained language model to efficiently extract informative sequence embeddings and employs self-attention pooling to focus on important residues. The prediction is further advanced by exploiting the homology information and accounting for the overlapping communities of proteins with related functions through the label diffusion algorithm. SPROF-GO was shown to surpass state-of-the-art sequence-based and even network-based approaches by more than 14.5, 27.3 and 10.1% in area under the precision-recall curve on the three sub-ontology test sets, respectively. Our method was also demonstrated to generalize well on non-homologous proteins and unseen species. Finally, visualization based on the attention mechanism indicated that SPROF-GO is able to capture sequence domains useful for function prediction. The datasets, source codes and trained models of SPROF-GO are available at https://github.com/biomed-AI/SPROF-GO. The SPROF-GO web server is freely available at http://bio-web1.nscc-gz.cn/app/sprof-go.

Tissue Specificity Based Isoform Function Prediction.

Alternative splicing enables a gene spliced into different isoforms and hence protein variants. Identifying individual functions of these isoforms help deciphering the functional diversity of proteins. Although much efforts have been made for automatic gene function prediction, few efforts have been moved toward computational isoform function prediction, mainly due to the unavailable (or scanty) functional annotations of isoforms. Existing efforts directly combine multiple RNA-seq datasets without account of the important tissue specificity of alternative splicing. To bridge this gap, we introduce a novel approach called TS-Isofun to predict the functions of isoforms by integrating multiple functional association networks with respect to tissue specificity. TS-Isofun first constructs tissue-specific isoform functional association networks using multiple RNA-seq datasets from tissue-wise. Next, TS-Isofun assigns weights to these networks and models the tissue specificity by selectively integrating them with adaptive weights. It then introduces a joint matrix factorization-based data fusion model to leverage the integrated network, gene-level data and functional annotations of genes to infer the functions of isoforms. To achieve coherent weight assignment and isoform function prediction, TS-Isofun jointly optimizes the weights of individual networks and the isoform function prediction in a unified objective function. Experimental results show that TS-Isofun significantly outperforms state-of-the-art methods and the account of tissue specificity contributes to more accurate isoform function prediction.

Putative Protein Discovery from Microalgal Genomes as a Synthetic Biology Protein Library for Heavy Metal Bio-Removal.

Simple SummaryNowadays, heavy metal polluted wastewater is one of the global challenges that leads to an insufficient supply of clean water. Taking advantage of what nature has to offer, several organisms, including microalgae, can natively bioremediate these heavy metals. However, the effectiveness of such processes does not meet expectations, especially with the increasing amount of pollution in today’s world. Therefore, with the goal of creating effective strains, synthetic biology via bioengineering is widely used as a strategy to enhance the heavy metal bio-removing capability, either by directly engineering the native ability of organisms or by transferring the ability to a more suitable host. In order to do so, a list of genes or proteins involved in the processes is crucial for stepwise engineering. Yet, a large amount of information remains to be discovered. In this work, a comprehensive library of putative proteins that are involved in heavy metal bio-removal from microalgae was constructed. Moreover, with the development of machine learning, the 3D structures of these proteins are also predicted, using machine learning-based methods, to aid the use of synthetic biology further.Synthetic biology is a principle that aims to create new biological systems with particular functions or to redesign the existing ones through bioengineering. Therefore, this principle is often utilized as a tool to put the knowledge learned to practical use in actual fields. However, there is still a great deal of information remaining to be found, and this limits the possible utilization of synthetic biology, particularly on the topic that is the focus of the present work—heavy metal bio-removal. In this work, we aim to construct a comprehensive library of putative proteins that might support heavy metal bio-removal. Hypothetical proteins were discovered from Chlorella and Scenedesmus genomes and extensively annotated. The protein structures of these putative proteins were also modeled through Alphafold2. Although a portion of this workflow has previously been demonstrated to annotate hypothetical proteins from whole genome sequences, the adaptation of such steps is yet to be done for library construction purposes. We also demonstrated further downstream steps that allow a more accurate function prediction of the hypothetical proteins by subjecting the models generated to structure-based annotation. In conclusion, a total of 72 newly discovered putative proteins were annotated with ready-to-use predicted structures available for further investigation.

EnsembleFam: towards more accurate protein family prediction in the twilight zone

BackgroundCurrent protein family modeling methods like profile Hidden Markov Model (pHMM), k-mer based methods, and deep learning-based methods do not provide very accurate protein function prediction for proteins in the twilight zone, due to low sequence similarity to reference proteins with known functions.ResultsWe present a novel method EnsembleFam, aiming at better function prediction for proteins in the twilight zone. EnsembleFam extracts the core characteristics of a protein family using similarity and dissimilarity features calculated from sequence homology relations. EnsembleFam trains three separate Support Vector Machine (SVM) classifiers for each family using these features, and an ensemble prediction is made to classify novel proteins into these families. Extensive experiments are conducted using the Clusters of Orthologous Groups (COG) dataset and G Protein-Coupled Receptor (GPCR) dataset. EnsembleFam not only outperforms state-of-the-art methods on the overall dataset but also provides a much more accurate prediction for twilight zone proteins.ConclusionsEnsembleFam, a machine learning method to model protein families, can be used to better identify members with very low sequence homology. Using EnsembleFam protein functions can be predicted using just sequence information with better accuracy than state-of-the-art methods.

SPOT-Contact-LM: improving single-sequence-based prediction of protein contact map using a transformer language model.

MotivationAccurate prediction of protein contact-map is essential for accurate protein structure and function prediction. As a result, many methods have been developed for protein contact map prediction. However, most methods rely on protein-sequence-evolutionary information, which may not exist for many proteins due to lack of naturally occurring homologous sequences. Moreover, generating evolutionary profiles is computationally intensive. Here, we developed a contact-map predictor utilizing the output of a pre-trained language model ESM-1b as an input along with a large training set and an ensemble of residual neural networks.ResultsWe showed that the proposed method makes a significant improvement over a single-sequence-based predictor SSCpred with 15% improvement in the F1-score for the independent CASP14-FM test set. It also outperforms evolutionary-profile-based methods trRosetta and SPOT-Contact with 48.7% and 48.5% respective improvement in the F1-score on the proteins without homologs (Neff = 1) in the independent SPOT-2018 set. The new method provides a much faster and reasonably accurate alternative to evolution-based methods, useful for large-scale prediction.Availability and implementationStand-alone-version of SPOT-Contact-LM is available at https://github.com/jas-preet/SPOT-Contact-Single. Direct prediction can also be made at https://sparks-lab.org/server/spot-contact-single. The datasets used in this research can also be downloaded from the GitHub.Supplementary information Supplementary data are available at Bioinformatics online.

Accurate protein function prediction via graph attention networks with predicted structure information.

Experimental protein function annotation does not scale with the fast-growing sequence databases. Only a tiny fraction (<0.1%) of protein sequences has experimentally determined functional annotations. Computational methods may predict protein function very quickly, but their accuracy is not very satisfactory. Based upon recent breakthroughs in protein structure prediction and protein language models, we develop GAT-GO, a graph attention network (GAT) method that may substantially improve protein function prediction by leveraging predicted structure information and protein sequence embedding. Our experimental results show that GAT-GO greatly outperforms the latest sequence- and structure-based deep learning methods. On the PDB-mmseqs testset where the train and test proteins share <15% sequence identity, our GAT-GO yields Fmax (maximum F-score) 0.508, 0.416, 0.501, and area under the precision-recall curve (AUPRC) 0.427, 0.253, 0.411 for the MFO, BPO, CCO ontology domains, respectively, much better than the homology-based method BLAST (Fmax 0.117, 0.121, 0.207 and AUPRC 0.120, 0.120, 0.163) that does not use any structure information. On the PDB-cdhit testset where the training and test proteins are more similar, although using predicted structure information, our GAT-GO obtains Fmax 0.637, 0.501, 0.542 for the MFO, BPO, CCO ontology domains, respectively, and AUPRC 0.662, 0.384, 0.481, significantly exceeding the just-published method DeepFRI that uses experimental structures, which has Fmax 0.542, 0.425, 0.424 and AUPRC only 0.313, 0.159, 0.193.

Probing Isoform Switching Events in Various Cancer Types: Lessons From Pan-Cancer Studies.

Alternative splicing is an essential regulatory mechanism for gene expression in mammalian cells contributing to protein, cellular, and species diversity. In cancer, alternative splicing is frequently disturbed, leading to changes in the expression of alternatively spliced protein isoforms. Advances in sequencing technologies and analysis methods led to new insights into the extent and functional impact of disturbed alternative splicing events. In this review, we give a brief overview of the molecular mechanisms driving alternative splicing, highlight the function of alternative splicing in healthy tissues and describe how alternative splicing is disrupted in cancer. We summarize current available computational tools for analyzing differential transcript usage, isoform switching events, and the pathogenic impact of cancer-specific splicing events. Finally, the strategies of three recent pan-cancer studies on isoform switching events are compared. Their methodological similarities and discrepancies are highlighted and lessons learned from the comparison are listed. We hope that our assessment will lead to new and more robust methods for cancer-specific transcript detection and help to produce more accurate functional impact predictions of isoform switching events.

Methods for Studying Bacterial–Fungal Interactions in the Microenvironments of Soil

Due to their small size, microorganisms directly experience only a tiny portion of the environmental heterogeneity manifested in the soil. The microscale variations in soil properties constrain the distribution of fungi and bacteria, and the extent to which they can interact with each other, thereby directly influencing their behavior and ecological roles. Thus, to obtain a realistic understanding of bacterial–fungal interactions, the spatiotemporal complexity of their microenvironments must be accounted for. The objective of this review is to further raise awareness of this important aspect and to discuss an overview of possible methodologies, some of easier applicability than others, that can be implemented in the experimental design in this field of research. The experimental design can be rationalized in three different scales, namely reconstructing the physicochemical complexity of the soil matrix, identifying and locating fungi and bacteria to depict their physical interactions, and, lastly, analyzing their molecular environment to describe their activity. In the long term, only relevant experimental data at the cell-to-cell level can provide the base for any solid theory or model that may serve for accurate functional prediction at the ecosystem level. The way to this level of application is still long, but we should all start small.

TRNAscan-SE 2.0: improved detection and functional classification of transfer RNA genes.

tRNAscan-SE has been widely used for transfer RNA (tRNA) gene prediction for over twenty years, developed just as the first genomes were decoded. With the massive increase in quantity and phylogenetic diversity of genomes, the accurate detection and functional prediction of tRNAs has become more challenging. Utilizing a vastly larger training set, we created nearly one hundred specialized isotype- and clade-specific models, greatly improving tRNAscan-SE’s ability to identify and classify both typical and atypical tRNAs. We employ a new comparative multi-model strategy where predicted tRNAs are scored against a full set of isotype-specific covariance models, allowing functional prediction based on both the anticodon and the highest-scoring isotype model. Comparative model scoring has also enhanced the program's ability to detect tRNA-derived SINEs and other likely pseudogenes. For the first time, tRNAscan-SE also includes fast and highly accurate detection of mitochondrial tRNAs using newly developed models. Overall, tRNA detection sensitivity and specificity is improved for all isotypes, particularly those utilizing specialized models for selenocysteine and the three subtypes of tRNA genes encoding a CAU anticodon. These enhancements will provide researchers with more accurate and detailed tRNA annotation for a wider variety of tRNAs, and may direct attention to tRNAs with novel traits.

Efficacy of computational predictions of the functional effect of idiosyncratic pharmacogenetic variants.

BackgroundPharmacogenetic variation is important to drug responses through diverse and complex mechanisms. Predictions of the functional impact of missense pharmacogenetic variants primarily rely on the degree of sequence conservation between species as a primary discriminator. However, idiosyncratic or off-target drug-variant interactions sometimes involve effects that are peripheral or accessory to the central systems in which a gene functions. Given the importance of sequence conservation to functional prediction tools—these idiosyncratic pharmacogenetic variants may violate the assumptions of predictive software commonly used to infer their effect.MethodsHere we exhaustively assess the effectiveness of eleven missense mutation functional inference tools on all known pharmacogenetic missense variants contained in the Pharmacogenomics Knowledgebase (PharmGKB) repository. We categorize PharmGKB entries into sub-classes to catalog likely off-target interactions, such that we may compare predictions across different variant annotations.ResultsAs previously demonstrated, functional inference tools perform variably across the complete set of PharmGKB variants, with large numbers of variants incorrectly classified as ‘benign’. However, we find substantial differences amongst PharmGKB variant sub-classes, particularly in variants known to cause off-target, type B adverse drug reactions, that are largely unrelated to the main pharmacological action of the drug. Specifically, variants associated with off-target effects (hence referred to as off-target variants) were most often incorrectly classified as ‘benign’. These results highlight the importance of understanding the underlying mechanism of pharmacogenetic variants and how variants associated with off-target effects will ultimately require new predictive algorithms.ConclusionIn this work we demonstrate that functional inference tools perform poorly on pharmacogenetic variants, particularly on subsets enriched for variants causing off-target, type B adverse drug reactions. We describe how to identify variants associated with off-target effects within PharmGKB in order to generate a training set of variants that is needed to develop new algorithms specifically for this class of variant. Development of such tools will lead to more accurate functional predictions and pave the way for the increased wide-spread adoption of pharmacogenetics in clinical practice.

DeepGOWeb: fast and accurate protein function prediction on the (Semantic) Web.

Understanding the functions of proteins is crucial to understand biological processes on a molecular level. Many more protein sequences are available than can be investigated experimentally. DeepGOPlus is a protein function prediction method based on deep learning and sequence similarity. DeepGOWeb makes the prediction model available through a website, an API, and through the SPARQL query language for interoperability with databases that rely on Semantic Web technologies. DeepGOWeb provides accurate and fast predictions and ensures that predicted functions are consistent with the Gene Ontology; it can provide predictions for any protein and any function in Gene Ontology. DeepGOWeb is freely available at https://deepgo.cbrc.kaust.edu.sa/.

SPOT-1D-Single: improving the single-sequence-based prediction of protein secondary structure, backbone angles, solvent accessibility and half-sphere exposures using a large training set and ensembled deep learning.

Knowing protein secondary and other one-dimensional structural properties are essential for accurate protein structure and function prediction. As a result, many methods have been developed for predicting these one-dimensional structural properties. However, most methods relied on evolutionary information that may not exist for many proteins due to a lack of sequence homologs. Moreover, it is computationally intensive for obtaining evolutionary information as the library of protein sequences continues to expand exponentially. Here, we developed a new single-sequence method called SPOT-1D-Single based on a large training dataset of 39 120 proteins deposited prior to 2016 and an ensemble of hybrid long-short-term-memory bidirectional neural network and convolutional neural network. We showed that SPOT-1D-Single consistently improves over SPIDER3-Single and ProteinUnet for secondary structure, solvent accessibility, contact number and backbone angles prediction for all seven independent test sets (TEST2018, SPOT-2016, SPOT-2016-HQ, SPOT-2018, SPOT-2018-HQ, CASP12 and CASP13 free-modeling targets). For example, the predicted three-state secondary structure's accuracy ranges from 72.12% to 74.28% by SPOT-1D-Single, compared to 69.1-72.6% by SPIDER3-Single and 70.6-73% by ProteinUnet. SPOT-1D-Single also predicts SS3 and SS8 with 6.24% and 6.98% better accuracy than SPOT-1D on SPOT-2018 proteins with no homologs (Neff = 1), respectively. The new method's improvement over existing techniques is due to a larger training set combined with ensembled learning. Standalone-version of SPOT-1D-Single is available at https://github.com/jas-preet/SPOT-1D-Single. Direct prediction can also be made at https://sparks-lab.org/server/spot-1d-single. The datasets used in this research can also be downloaded from GitHub. Supplementary data are available at Bioinformatics online.

DeepIII: Predicting Isoform-Isoform Interactions by Deep Neural Networks and Data Fusion

Alternative splicing enables a gene translating into different isoforms and into the corresponding proteoforms, which actually accomplish various biological functions of a living body. Isoform-isoform interactions (IIIs) provide a higher resolution interactome to explore the cellular processes and disease mechanisms than the canonically studied protein-protein interactions (PPIs), which are often recorded at the coarse gene level. The knowledge of IIIs is critical to map pathways, understand protein complexity and functional diversity, but the known IIIs are very scanty. In this paper, we propose a deep learning based method called DeepIII to systematically predict genome-wide IIIs by integrating diverse data sources, including RNA-seq datasets of different human tissues, exon array data, domain-domain interactions (DDIs) of proteins, nucleotide sequences and amino acid sequences. Particularly, DeepIII fuses these data to learn the representation of isoform pairs with a four-layer deep neural networks, and then performs binary classification on the learnt representation to achieve the prediction of IIIs. Experimental results show that DeepIII achieves a superior prediction performance to the state-of-the-art solutions and the III network constructed by DeepIII gives more accurate isoform function prediction. Case studies further confirm that DeepIII can differentiate the individual interaction partners of different isoforms spliced from the same gene. The code and datasets of DeepIII are available at http://mlda.swu.edu.cn/codes.php?name=DeepIII.

Automatic Gene Function Prediction in the 2020's.

The current rate at which new DNA and protein sequences are being generated is too fast to experimentally discover the functions of those sequences, emphasizing the need for accurate Automatic Function Prediction (AFP) methods. AFP has been an active and growing research field for decades and has made considerable progress in that time. However, it is certainly not solved. In this paper, we describe challenges that the AFP field still has to overcome in the future to increase its applicability. The challenges we consider are how to: (1) include condition-specific functional annotation, (2) predict functions for non-model species, (3) include new informative data sources, (4) deal with the biases of Gene Ontology (GO) annotations, and (5) maximally exploit the GO to obtain performance gains. We also provide recommendations for addressing those challenges, by adapting (1) the way we represent proteins and genes, (2) the way we represent gene functions, and (3) the algorithms that perform the prediction from gene to function. Together, we show that AFP is still a vibrant research area that can benefit from continuing advances in machine learning with which AFP in the 2020s can again take a large step forward reinforcing the power of computational biology.

EnzyMine: a comprehensive database for enzyme function annotation with enzymatic reaction chemical feature.

Addition of chemical structural information in enzymatic reactions has proven to be significant for accurate enzyme function prediction. However, such chemical data lack systematic feature mining and hardly exist in enzyme-related databases. Therefore, global mining of enzymatic reactions will offer a unique landscape for researchers to understand the basic functional mechanisms of natural bioprocesses and facilitate enzyme function annotation. Here, we established a new knowledge base called EnzyMine, through which we propose to elucidate enzymatic reaction features and then link them with sequence and structural annotations. EnzyMine represents an advanced database that extends enzyme knowledge by incorporating reaction chemical feature strategies, strengthening the connectivity between enzyme and metabolic reactions. Therefore, it has the potential to reveal many new metabolic pathways involved with given enzymes, as well as expand enzyme function annotation. Database URL: http://www.rxnfinder.org/enzymine/.