DeepGO: predicting protein functions from sequence and interactions using a deep ontology-aware classifier

Integrating unsupervised language model with triplet neural networks for protein gene ontology prediction

Accurate identification of protein function is critical to elucidate life mechanisms and design new drugs. We proposed a novel deep-learning method, ATGO, to predict Gene Ontology (GO) attributes of proteins through a triplet neural-network architecture embedded with pre-trained language models from protein sequences. The method was systematically tested on 1068 non-redundant benchmarking proteins and 3328 targets from the third Critical Assessment of Protein Function Annotation (CAFA) challenge. Experimental results showed that ATGO achieved a significant increase of the GO prediction accuracy compared to the state-of-the-art approaches in all aspects of molecular function, biological process, and cellular component. Detailed data analyses showed that the major advantage of ATGO lies in the utilization of pre-trained transformer language models which can extract discriminative functional pattern from the feature embeddings. Meanwhile, the proposed triplet network helps enhance the association of functional similarity with feature similarity in the sequence embedding space. In addition, it was found that the combination of the network scores with the complementary homology-based inferences could further improve the accuracy of the predicted models. These results demonstrated a new avenue for high-accuracy deep-learning function prediction that is applicable to large-scale protein function annotations from sequence alone.

GOProFormer: A Multi-Modal Transformer Method for Gene Ontology Protein Function Prediction

Protein Language Models (PLMs) are shown to be capable of learning sequence representations useful for various prediction tasks, from subcellular localization, evolutionary relationships, family membership, and more. They have yet to be demonstrated useful for protein function prediction. In particular, the problem of automatic annotation of proteins under the Gene Ontology (GO) framework remains open. This paper makes two key contributions. It debuts a novel method that leverages the transformer architecture in two ways. A sequence transformer encodes protein sequences in a task-agnostic feature space. A graph transformer learns a representation of GO terms while respecting their hierarchical relationships. The learned sequence and GO terms representations are combined and utilized for multi-label classification, with the labels corresponding to GO terms. The method is shown superior over recent representative GO prediction methods. The second major contribution in this paper is a deep investigation of different ways of constructing training and testing datasets. The paper shows that existing approaches under- or over-estimate the generalization power of a model. A novel approach is proposed to address these issues, resulting in a new benchmark dataset to rigorously evaluate and compare methods and advance the state-of-the-art.

TripletProt: Deep Representation Learning of Proteins Based On Siamese Networks

Pretrained representations have recently gained attention in various machine learning applications. Nonetheless, the high computational costs associated with training these models have motivated alternative approaches for representation learning. Herein we introduce TripletProt, a new approach for protein representation learning based on the Siamese neural networks. Representation learning of biological entities which capture essential features can alleviate many of the challenges associated with supervised learning in bioinformatics. The most important distinction of our proposed method is relying on the protein-protein interaction (PPI) network. The computational cost of the generated representations for any potential application is significantly lower than comparable methods since the length of the representations is significantly smaller than that in other approaches. TripletProt offers great potentials for the protein informatics tasks and can be widely applied to similar tasks. We evaluate TripletProt comprehensively in protein functional annotation tasks including sub-cellular localization (14 categories) and gene ontology prediction (more than 2000 classes), which are both challenging multi-class, multi-label classification machine learning problems. We compare the performance of TripletProt with the state-of-the-art approaches including a recurrent language model-based approach (i.e., UniRep), as well as a protein-protein interaction (PPI) network and sequence-based method (i.e., DeepGO). Our TripletProt showed an overall improvement of F1 score in the above mentioned comprehensive functional annotation tasks, solely relying on the PPI network. Availability: The source code and datasets are available at https://github.com/EsmaeilNourani/TripletProt.

TripletGO: Integrating Transcript Expression Profiles with Protein Homology Inferences for Gene Function Prediction

Gene Ontology (GO) has been widely used to annotate functions of genes and gene products. Here, we proposed a new method, TripletGO, to deduce GO terms of protein-coding and non-coding genes, through the integration of four complementary pipelines built on transcript expression profile, genetic sequence alignment, protein sequence alignment, and naïve probability. TripletGO was tested on a large set of 5754 genes from 8 species (human, mouse, Arabidopsis, rat, fly, budding yeast, fission yeast, and nematoda) and 2433 proteins with available expression data from the third Critical Assessment of Protein Function Annotation challenge (CAFA3). Experimental results show that TripletGO achieves function annotation accuracy significantly beyond the current state-of-the-art approaches. Detailed analyses show that the major advantage of TripletGO lies in the coupling of a new triplet network-based profiling method with the feature space mapping technique, which can accurately recognize function patterns from transcript expression profiles. Meanwhile, the combination of multiple complementary models, especially those from transcript expression and protein-level alignments, improves the coverage and accuracy of the final GO annotation results. The standalone package and an online server of TripletGO are freely available at https://zhanggroup.org/TripletGO/.

Isoform function prediction by Gene Ontology embedding

MOTIVATION

High-resolution annotation of gene functions is a central task in functional genomics. Multiple proteoforms translated from alternatively spliced isoforms from a single gene are actual function performers and greatly increase the functional diversity. The specific functions of different isoforms can decipher the molecular basis of various complex diseases at a finer granularity. Multi-instance learning (MIL)-based solutions have been developed to distribute gene(bag)-level Gene Ontology (GO) annotations to isoforms(instances), but they simply presume that a particular annotation of the gene is responsible by only one isoform, neglect the hierarchical structures and semantics of massive GO terms (labels), or can only handle dozens of terms.

RESULTS

We propose an efficacy approach IsofunGO to differentiate massive functions of isoforms by GO embedding. Particularly, IsofunGO first introduces an attributed hierarchical network to model massive GO terms, and a GO network embedding strategy to learn compact representations of GO terms and project GO annotations of genes into compressed ones, this strategy not only explores and preserves hierarchy between GO terms but also greatly reduces the prediction load. Next, it develops an attention-based MIL network to fuse genomics and transcriptomics data of isoforms and predict isoform functions by referring to compressed annotations. Extensive experiments on benchmark datasets demonstrate the efficacy of IsofunGO. Both the GO embedding and attention mechanism can boost the performance and interpretability.

AVAILABILITYAND IMPLEMENTATION

The code of IsofunGO is available at http://www.sdu-idea.cn/codes.php?name=IsofunGO.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

PFP-GO: Integrating protein sequence, domain and protein-protein interaction information for protein function prediction using ranked GO terms

Protein function prediction is gradually emerging as an essential field in biological and computational studies. Though the latter has clinched a significant footprint, it has been observed that the application of computational information gathered from multiple sources has more significant influence than the one derived from a single source. Considering this fact, a methodology, PFP-GO, is proposed where heterogeneous sources like Protein Sequence, Protein Domain, and Protein-Protein Interaction Network have been processed separately for ranking each individual functional GO term. Based on this ranking, GO terms are propagated to the target proteins. While Protein sequence enriches the sequence-based information, Protein Domain and Protein-Protein Interaction Networks embed structural/functional and topological based information, respectively, during the phase of GO ranking. Performance analysis of PFP-GO is also based on Precision, Recall, and F-Score. The same was found to perform reasonably better when compared to the other existing state-of-art. PFP-GO has achieved an overall Precision, Recall, and F-Score of 0.67, 0.58, and 0.62, respectively. Furthermore, we check some of the top-ranked GO terms predicted by PFP-GO through multilayer network propagation that affect the 3D structure of the genome. The complete source code of PFP-GO is freely available at https://sites.google.com/view/pfp-go/.

Feature importance network reveals novel functional relationships between biological features in Arabidopsis thaliana

Understanding how the different cellular components are working together to form a living cell requires multidisciplinary approaches combining molecular and computational biology. Machine learning shows great potential in life sciences, as it can find novel relationships between biological features. Here, we constructed a dataset of 11,801 gene features for 31,522 Arabidopsis thaliana genes and developed a machine learning workflow to identify linked features. The detected linked features are visualised as a Feature Important Network (FIN), which can be mined to reveal a variety of novel biological insights pertaining to gene function. We demonstrate how FIN can be used to generate novel insights into gene function. To make this network easily accessible to the scientific community, we present the FINder database, available at finder.plant.tools.

SLPred: a multi-view subcellular localization prediction tool for multi-location human proteins

SUMMARY

Accurate prediction of the subcellular locations (SLs) of proteins is a critical topic in protein science. In this study, we present SLPred, an ensemble-based multi-view and multi-label protein subcellular localization prediction tool. For a query protein sequence, SLPred provides predictions for nine main SLs using independent machine-learning models trained for each location. We used UniProtKB/Swiss-Prot human protein entries and their curated SL annotations as our source data. We connected all disjoint terms in the UniProt SL hierarchy based on the corresponding term relationships in the cellular component category of Gene Ontology and constructed a training dataset that is both reliable and large scale using the re-organized hierarchy. We tested SLPred on multiple benchmarking datasets including our-in house sets and compared its performance against six state-of-the-art methods. Results indicated that SLPred outperforms other tools in the majority of cases.

AVAILABILITY AND IMPLEMENTATION

SLPred is available both as an open-access and user-friendly web-server (https://slpred.kansil.org) and a stand-alone tool (https://github.com/kansil/SLPred). All datasets used in this study are also available at https://slpred.kansil.org.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

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.

DeePROG: Deep Attention-Based Model for Diseased Gene Prognosis by Fusing Multi-Omics Data

An in-depth exploration of gene prognosis using different methodologies aids in understanding various biological regulations of genes in disease pathobiology and molecular functions. Interpreting gene functions at biological and molecular levels remains a daunting yet crucial task in domains such as drug design, personalized medicine, and next-generation diagnostics. Recent advancements in omics technologies have produced diverse heterogeneous genomic datasets like micro-array gene expression, miRNA expression, DNA sequence, 3D structures, which are significant resources for understanding the gene functions. In this paper, we propose a novel self-attention based deep multi-modal model, named DeePROG, for the prognosis of disease affected genes based on heterogeneous omics data. We use three NCBI datasets covering three modalities, namely gene expression profile, the underlying DNA sequence, and the 3D protein structures. To extract useful features from each modality, we develop several context-specific deep learning models. Besides, we develop three attention-based deep bi-modal architectures along with DeePROG to leverage the prognosis of the underlying biomedical data. We assess the performance of the models' in terms of computational assessment of function annotation (CAFA2) metrics. Moreover, we analyze the results in terms of receiver operating characteristics (ROC) curve in high-class imbalance data setting and perform statistical significance tests in terms of Welch's t-test. Experiment results show that DeePROG significantly outperforms baseline models across in terms of performance metrics. The source code and all preprocessed datasets used in this study are available at https://github.com/duttaprat/DeePROG.

An Ensemble Tf-Idf Based Approach to Protein Function Prediction via Sequence Segmentation

This paper explores the use of variants of tf-idf-based descriptors, namely length-normalized-tf-idf and log-normalized-tf-idf, combined with a segmentation technique, for efficient modeling of variable-length protein sequences. The proposed solution, ProtVecGen-Ensemble, is an ensemble of three models trained on differently segmented datasets constructed from an input dataset containing complete protein sequences. Evaluations using biological process (BP) and molecular function (MF) datasets demonstrate that the proposed feature set is not only superior to its contemporaries but also produces more consistent results with respect to variation in sequence lengths. Improvements of +6.07% (BP) and +7.56% (MF) over state-of-the-art tf-idf-based MLDA feature set were obtained. The best results were achieved when ProtVecGen-Ensemble was combined with ProtVecGen-Plus - the state-of-the-art method for protein function prediction - resulting in improvements of +8.90% (BP) and +11.28% (MF) over MLDA and +1.49% (BP) and +2.07% (MF) over ProtVecGen-Plus+MLDA. To capture the performance consistency with respect to sequence lengths, we have defined a variance-based metric, with lower values indicating better performance. On this metric, the proposed ProtVecGen-Ensemble+ProtVecGen-Plus framework resulted in reductions of 56.85 percent (BP) and 56.08 percent (MF) over MLDA and 10.37 percent (BP) and 26.48 percent (MF) over ProtVecGenPlus+MLDA.

Enhancing Protein Function Prediction Performance by Utilizing AlphaFold-Predicted Protein Structures

The structure of a protein is of great importance in determining its functionality, and this characteristic can be leveraged to train data-driven prediction models. However, the limited number of available protein structures severely limits the performance of these models. AlphaFold2 and its open-source data set of predicted protein structures have provided a promising solution to this problem, and these predicted structures are expected to benefit the model performance by increasing the number of training samples. In this work, we constructed a new data set that acted as a benchmark and implemented a state-of-the-art structure-based approach for determining whether the performance of the function prediction model can be improved by putting additional AlphaFold-predicted structures into the training set and further compared the performance differences between two models separately trained with real structures only and AlphaFold-predicted structures only. Experimental results indicated that structure-based protein function prediction models could benefit from virtual training data consisting of AlphaFold-predicted structures. First, model performances were improved in all three categories of Gene Ontology terms (GO terms) after adding predicted structures as training samples. Second, the model trained only on AlphaFold-predicted virtual samples achieved comparable performances to the model based on experimentally solved real structures, suggesting that predicted structures were almost equally effective in predicting protein functionality.

Gene Ontology Capsule GAN: an improved architecture for protein function prediction

Proteins are the core of all functions pertaining to living things. They consist of an extended amino acid chain folding into a three-dimensional shape that dictates their behavior. Currently, convolutional neural networks (CNNs) have been pivotal in predicting protein functions based on protein sequences. While it is a technology crucial to the niche, the computation cost and translational invariance associated with CNN make it impossible to detect spatial hierarchies between complex and simpler objects. Therefore, this research utilizes capsule networks to capture spatial information as opposed to CNNs. Since capsule networks focus on hierarchical links, they have a lot of potential for solving structural biology challenges. In comparison to the standard CNNs, our results exhibit an improvement in accuracy. Gene Ontology Capsule GAN (GOCAPGAN) achieved an F1 score of 82.6%, a precision score of 90.4% and recall score of 76.1%.

Integration of Human Protein Sequence and Protein-Protein Interaction Data by Graph Autoencoder to Identify Novel Protein-Abnormal Phenotype Associations

Understanding gene functions and their associated abnormal phenotypes is crucial in the prevention, diagnosis and treatment against diseases. The Human Phenotype Ontology (HPO) is a standardized vocabulary for describing the phenotype abnormalities associated with human diseases. However, the current HPO annotations are far from completion, and only a small fraction of human protein-coding genes has HPO annotations. Thus, it is necessary to predict protein-phenotype associations using computational methods. Protein sequences can indicate the structure and function of the proteins, and interacting proteins are more likely to have same function. It is promising to integrate these features for predicting HPO annotations of human protein. We developed GraphPheno, a semi-supervised method based on graph autoencoders, which does not require feature engineering to capture deep features from protein sequences, while also taking into account the topological properties in the protein–protein interaction network to predict the relationships between human genes/proteins and abnormal phenotypes. Cross validation and independent dataset tests show that GraphPheno has satisfactory prediction performance. The algorithm is further confirmed on automatic HPO annotation for no-knowledge proteins under the benchmark of the second Critical Assessment of Functional Annotation, 2013–2014 (CAFA2), where GraphPheno surpasses most existing methods. Further bioinformatics analysis shows that predicted certain phenotype-associated genes using GraphPheno share similar biological properties with known ones. In a case study on the phenotype of abnormality of mitochondrial respiratory chain, top prioritized genes are validated by recent papers. We believe that GraphPheno will help to reveal more associations between genes and phenotypes, and contribute to the discovery of drug targets.

Proteome-Wide Profiling of the Covalent-Druggable Cysteines with a Structure-Based Deep Graph Learning Network

Covalent ligands have attracted increasing attention due to their unique advantages, such as long residence time, high selectivity, and strong binding affinity. They also show promise for targets where previous efforts to identify noncovalent small molecule inhibitors have failed. However, our limited knowledge of covalent binding sites has hindered the discovery of novel ligands. Therefore, developing in silico methods to identify covalent binding sites is highly desirable. Here, we propose DeepCoSI, the first structure-based deep graph learning model to identify ligandable covalent sites in the protein. By integrating the characterization of the binding pocket and the interactions between each cysteine and the surrounding environment, DeepCoSI achieves state-of-the-art predictive performances. The validation on two external test sets which mimic the real application scenarios shows that DeepCoSI has strong ability to distinguish ligandable sites from the others. Finally, we profiled the entire set of protein structures in the RCSB Protein Data Bank (PDB) with DeepCoSI to evaluate the ligandability of each cysteine for covalent ligand design, and made the predicted data publicly available on website.

Deep embeddings to comprehend and visualize microbiome protein space

Understanding the function of microbial proteins is essential to reveal the clinical potential of the microbiome. The application of high-throughput sequencing technologies allows for fast and increasingly cheaper acquisition of data from microbial communities. However, many of the inferred protein sequences are novel and not catalogued, hence the possibility of predicting their function through conventional homology-based approaches is limited, which indicates the need for further research on alignment-free methods. Here, we leverage a deep-learning-based representation of proteins to assess its utility in alignment-free analysis of microbial proteins. We trained a language model on the Unified Human Gastrointestinal Protein catalogue and validated the resulting protein representation on the bacterial part of the SwissProt database. Finally, we present a use case on proteins involved in SCFA metabolism. Results indicate that the deep learning model manages to accurately represent features related to protein structure and function, allowing for alignment-free protein analyses. Technologies that contextualize metagenomic data are a promising direction to deeply understand the microbiome.

DeephageTP: a convolutional neural network framework for identifying phage-specific proteins from metagenomic sequencing data

Bacteriophages (phages) are the most abundant and diverse biological entity on Earth. Due to the lack of universal gene markers and database representatives, there about 50-90% of genes of phages are unable to assign functions. This makes it a challenge to identify phage genomes and annotate functions of phage genes efficiently by homology search on a large scale, especially for newly phages. Portal (portal protein), TerL (large terminase subunit protein), and TerS (small terminase subunit protein) are three specific proteins of Caudovirales phage. Here, we developed a CNN (convolutional neural network)-based framework, DeephageTP, to identify the three specific proteins from metagenomic data. The framework takes one-hot encoding data of original protein sequences as the input and automatically extracts predictive features in the process of modeling. To overcome the false positive problem, a cutoff-loss-value strategy is introduced based on the distributions of the loss values of protein sequences within the same category. The proposed model with a set of cutoff-loss-values demonstrates high performance in terms of Precision in identifying TerL and Portal sequences (94% and 90%, respectively) from the mimic metagenomic dataset. Finally, we tested the efficacy of the framework using three real metagenomic datasets, and the results shown that compared to the conventional alignment-based methods, our proposed framework had a particular advantage in identifying the novel phage-specific protein sequences of portal and TerL with remote homology to their counterparts in the training datasets. In summary, our study for the first time develops a CNN-based framework for identifying the phage-specific protein sequences with high complexity and low conservation, and this framework will help us find novel phages in metagenomic sequencing data. The DeephageTP is available at https://github.com/chuym726/DeephageTP.

A neural network-based framework to understand the type 2 diabetes-related alteration of the human gut microbiome

The identification of microbial markers adequate to delineate the disease-related microbiome alterations from the complex human gut microbiota is of great interest. Here, we develop a framework combining neural network (NN) and random forest, resulting in 40 marker species and 90 marker genes identified from the metagenomic data set (185 healthy and 183 type 2 diabetes [T2D] samples), respectively. In terms of these markers, the NN model obtained higher accuracy in classifying the T2D-related samples than other methods; the interaction network analyses identified the key species and functional modules; the regression analysis determined that fasting blood glucose is the most significant factor (p < 0.05) in the T2D-related alteration of the human gut microbiome. We also observed that those marker species varied little across the case and control samples greatly shift in the different stages of the T2D development, suggestive of their important roles in the T2D-related microbiome alteration. Our study provides a new way of identifying the disease-related biomarkers and analyzing the role they may play in the development of the disease.

Using deep learning to annotate the protein universe

Understanding the relationship between amino acid sequence and protein function is a long-standing challenge with far-reaching scientific and translational implications. State-of-the-art alignment-based techniques cannot predict function for one-third of microbial protein sequences, hampering our ability to exploit data from diverse organisms. Here, we train deep learning models to accurately predict functional annotations for unaligned amino acid sequences across rigorous benchmark assessments built from the 17,929 families of the protein families database Pfam. The models infer known patterns of evolutionary substitutions and learn representations that accurately cluster sequences from unseen families. Combining deep models with existing methods significantly improves remote homology detection, suggesting that the deep models learn complementary information. This approach extends the coverage of Pfam by >9.5%, exceeding additions made over the last decade, and predicts function for 360 human reference proteome proteins with no previous Pfam annotation. These results suggest that deep learning models will be a core component of future protein annotation tools.

Protein-protein interaction networks as miners of biological discovery

Protein-protein interactions (PPIs) form the basis of a myriad of biological pathways and mechanism, such as the formation of protein-complexes or the components of signaling cascades. Here, we reviewed experimental methods for identifying PPI pairs, including yeast two-hybrid, mass spectrometry, co-localization, and co-immunoprecipitation. Furthermore, a range of computational methods leveraging biochemical properties, evolution history, protein structures and more have enabled identification of additional PPIs. Given the wealth of known PPIs, we reviewed important network methods to construct and analyze networks of PPIs. These methods aid biological discovery through identifying hub genes and dynamic changes in the network, and have been thoroughly applied in various fields of biological research. Lastly, we discussed the challenges and future direction of research utilizing the power of PPI networks. This article is protected by copyright. All rights reserved.

CrowdGO: Machine learning and semantic similarity guided consensus Gene Ontology annotation

Characterising gene function for the ever-increasing number and diversity of species with annotated genomes relies almost entirely on computational prediction methods. These software are also numerous and diverse, each with different strengths and weaknesses as revealed through community benchmarking efforts. Meta-predictors that assess consensus and conflict from individual algorithms should deliver enhanced functional annotations. To exploit the benefits of meta-approaches, we developed CrowdGO, an open-source consensus-based Gene Ontology (GO) term meta-predictor that employs machine learning models with GO term semantic similarities and information contents. By re-evaluating each gene-term annotation, a consensus dataset is produced with high-scoring confident annotations and low-scoring rejected annotations. Applying CrowdGO to results from a deep learning-based, a sequence similarity-based, and two protein domain-based methods, delivers consensus annotations with improved precision and recall. Furthermore, using standard evaluation measures CrowdGO performance matches that of the community’s best performing individual methods. CrowdGO therefore offers a model-informed approach to leverage strengths of individual predictors and produce comprehensive and accurate gene functional annotations.

New technologies mean that we are able to read the genetic blueprints in the form of complete genome sequences from many different species. We are also able to use computational methods combined with evidence from experiments to map out the locations in the genomes of many thousands of genes and other important regions. However, discovering and characterising the biological functions of all these genes and their protein products requires considerably more experimental work. In order to gain insights into the possible functions of the many genes currently lacking functional information from experiments we must therefore rely on methods that computationally predict protein functions. Many different software tools have been developed to tackle this challenge, each with their own strengths and weaknesses as shown by several community-based competitions that assess the performance of the predictors. Taking advantage of powerful modern machine learning techniques, we developed CrowdGO, a new software that aims to combine predictions from several tools and produce comprehensive and accurate gene functional annotations. CrowdGO is able to computationally assess agreements and conflicts amongst annotations from different predictors to then re-evaluate the results and deliver enhanced predictions of protein functions.

Neural Network and Random Forest Models in Protein Function Prediction

Over the past decade, the demand for automated protein function prediction has increased due to the volume of newly sequenced proteins. In this paper, we address the function prediction task by developing an ensemble system automatically assigning Gene Ontology (GO) terms to the given input protein sequence. We develop an ensemble system which combines the GO predictions made by random forest (RF) and neural network (NN) classifiers. Both RF and NN models rely on features derived from BLAST sequence alignments, taxonomy and protein signature analysis tools. In addition, we report on experiments with a NN model that directly analyzes the amino acid sequence as its sole input, using a convolutional layer. The Swiss-Prot database is used as the training and evaluation data. In the CAFA3 evaluation, which relies on experimental verification of the functional predictions, our submitted ensemble model demonstrates competitive performance ranking among top-10 best-performing systems out of over 100 submitted systems. In this paper, we evaluate and further improve the CAFA3-submitted system. Our machine learning models together with the data pre-processing and feature generation tools are publicly available as an open source software at https://github.com/TurkuNLP/CAFA3.

Modeling multi-scale data via a network of networks

MOTIVATION

Prediction of node and graph labels are prominent network science tasks. Data analyzed in these tasks are sometimes related: entities represented by nodes in a higher-level (higher scale) network can themselves be modeled as networks at a lower level. We argue that systems involving such entities should be integrated with a 'network of networks' (NoNs) representation. Then, we ask whether entity label prediction using multi-level NoN data via our proposed approaches is more accurate than using each of single-level node and graph data alone, i.e. than traditional node label prediction on the higher-level network and graph label prediction on the lower-level networks. To obtain data, we develop the first synthetic NoN generator and construct a real biological NoN. We evaluate accuracy of considered approaches when predicting artificial labels from the synthetic NoNs and proteins' functions from the biological NoN.

RESULTS

For the synthetic NoNs, our NoN approaches outperform or are as good as node- and network-level ones depending on the NoN properties. For the biological NoN, our NoN approaches outperform the single-level approaches for just under half of the protein functions, and for 30% of the functions, only our NoN approaches make meaningful predictions, while node- and network-level ones achieve random accuracy. So, NoN-based data integration is important.

AVAILABILITY AND IMPLEMENTATION

The software and data are available at https://nd.edu/~cone/NoNs.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Current progress and open challenges for applying deep learning across the biosciences

Deep Learning (DL) has recently enabled unprecedented advances in one of the grand challenges in computational biology: the half-century-old problem of protein structure prediction. In this paper we discuss recent advances, limitations, and future perspectives of DL on five broad areas: protein structure prediction, protein function prediction, genome engineering, systems biology and data integration, and phylogenetic inference. We discuss each application area and cover the main bottlenecks of DL approaches, such as training data, problem scope, and the ability to leverage existing DL architectures in new contexts. To conclude, we provide a summary of the subject-specific and general challenges for DL across the biosciences.

PFmulDL: a novel strategy enabling multi-class and multi-label protein function annotation by integrating diverse deep learning methods

Bioinformatic annotation of protein function is essential but extremely sophisticated, which asks for extensive efforts to develop effective prediction method. However, the existing methods tend to amplify the representativeness of the families with large number of proteins by misclassifying the proteins in the families with small number of proteins. That is to say, the ability of the existing methods to annotate proteins in the 'rare classes' remains limited. Herein, a new protein function annotation strategy, PFmulDL, integrating multiple deep learning methods, was thus constructed. First, the recurrent neural network was integrated, for the first time, with the convolutional neural network to facilitate the function annotation. Second, a transfer learning method was introduced to the model construction for further improving the prediction performances. Third, based on the latest data of Gene Ontology, the newly constructed model could annotate the largest number of protein families comparing with the existing methods. Finally, this newly constructed model was found capable of significantly elevating the prediction performance for the 'rare classes' without sacrificing that for the 'major classes'. All in all, due to the emerging requirements on improving the prediction performance for the proteins in 'rare classes', this new strategy would become an essential complement to the existing methods for protein function prediction. All the models and source codes are freely available and open to all users at: https://github.com/idrblab/PFmulDL.

Ontology-aware neural network: a general framework for pattern mining from microbiome data

With the rapid accumulation of microbiome data around the world, numerous computational bioinformatics methods have been developed for pattern mining from such paramount microbiome data. Current microbiome data mining methods, such as gene and species mining, rely heavily on sequence comparison. Most of these methods, however, have a clear trade-off, particularly, when it comes to big-data analytical efficiency and accuracy. Microbiome entities are usually organized in ontology structures, and pattern mining methods that have considered ontology structures could offer advantages in mining efficiency and accuracy. Here, we have summarized the ontology-aware neural network (ONN) as a novel framework for microbiome data mining. We have discussed the applications of ONN in multiple contexts, including gene mining, species mining and microbial community dynamic pattern mining. We have then highlighted one of the most important characteristics of ONN, namely, novel knowledge discovery, which makes ONN a standout among all microbiome data mining methods. Finally, we have provided several applications to showcase the advantage of ONN over other methods in microbiome data mining. In summary, ONN represents a paradigm shift for pattern mining from microbiome data: from traditional machine learning approach to ontology-aware and model-based approach, which has found its broad application scenarios in microbiome data mining.

Deep learning in prediction of intrinsic disorder in proteins

Intrinsic disorder prediction is an active area that has developed over 100 predictors. We identify and investigate a recent trend towards the development of deep neural network (DNN)-based methods. The first DNN-based method was released in 2013 and since 2019 deep learners account for majority of the new disorder predictors. We find that the 13 currently available DNN-based predictors are diverse in their topologies, sizes of their networks and the inputs that they utilize. We empirically show that the deep learners are statistically more accurate than other types of disorder predictors using the blind test dataset from the recent community assessment of intrinsic disorder predictions (CAID). We also identify several well-rounded DNN-based predictors that are accurate, fast and/or conveniently available. The popularity, favorable predictive performance and architectural flexibility suggest that deep networks are likely to fuel the development of future disordered predictors. Novel hybrid designs of deep networks could be used to adequately accommodate for diversity of types and flavors of intrinsic disorder. We also discuss scarcity of the DNN-based methods for the prediction of disordered binding regions and the need to develop more accurate methods for this prediction.

A deep learning model to detect novel pore-forming proteins

Many pore-forming proteins originating from pathogenic bacteria are toxic against agricultural pests. They are the key ingredients in several pesticidal products for agricultural use, including transgenic crops. There is an urgent need to identify novel pore-forming proteins to combat development of resistance in pests to existing products, and to develop products that are effective against a broader range of pests. Existing computational methodologies to search for these proteins rely on sequence homology-based approaches. These approaches are based on similarities between protein sequences, and thus are limited in their usefulness for discovering novel proteins. In this paper, we outline a novel deep learning model trained on pore-forming proteins from the public domain. We compare different ways of encoding protein information during training, and contrast it with traditional approaches. We show that our model is capable of identifying known pore formers with no sequence similarity to the proteins used to train the model, and therefore holds promise for identifying novel pore formers.

PANDA2: protein function prediction using graph neural networks

High-throughput sequencing technologies have generated massive protein sequences, but the annotations of protein sequences highly rely on the low-throughput and expensive biological experiments. Therefore, accurate and fast computational alternatives are needed to infer functional knowledge from protein sequences. The gene ontology (GO) directed acyclic graph (DAG) contains the hierarchical relationships between GO terms but is hard to be integrated into machine learning algorithms for functional predictions. We developed a deep learning system named PANDA2 to predict protein functions, which used the cutting-edge graph neural network to model the topology of the GO DAG and integrated the features generated by transformer protein language models. Compared with the top 10 methods in CAFA3, PANDA2 ranked first in cellular component ontology (CCO), tied first in biological process ontology (BPO) but had a higher coverage rate, and second in molecular function ontology (MFO). Compared with other recently-developed cutting-edge predictors DeepGOPlus, GOLabeler, and DeepText2GO, and benchmarked on another independent dataset, PANDA2 ranked first in CCO, first in BPO, and second in MFO. PANDA2 can be freely accessed from http://dna.cs.miami.edu/PANDA2/.

Protein-DNA/RNA interactions: Machine intelligence tools and approaches in the era of artificial intelligence and big data

With the development of artificial intelligence technologies and the availability of large amounts of biological data, computational methods for proteomics have undergone a developmental process from traditional machine learning to deep learning. This review focuses on computational approaches and tools for the prediction of protein-DNA/RNA interactions using machine intelligence techniques. We provide an overview of the development progress of computational methods and summarize the advantages and shortcomings of these methods. We further compiled applications in tasks related to the protein-DNA/RNA interactions, and pointed out possible future application trends. Moreover, biological sequence-digitizing representation strategies used in different types of computational methods are also summarized and discussed. This article is protected by copyright. All rights reserved.

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.

Deep neural learning based protein function prediction

It is vital for the annotation of uncharacterized proteins by protein function prediction. At present, Deep Neural Network based protein function prediction is mainly carried out for dataset of small scale proteins or Gene Ontology, and usually explore the relationships between single protein feature and function tags. The practical methods for large-scale multi-features protein prediction still need to be studied in depth. This paper proposes a DNN based protein function prediction approach IGP-DNN. This method uses Grasshopper Optimization Algorithm (GOA) and Intuitionistic Fuzzy c-Means clustering (IFCM) based protein function modules extracting algorithm to extract the features of protein modules, utilizing Kernel Principal Component Analysis (KPCA) method to reduce the dimensionality of the protein attribute information, and integrating module features and attribute features. Inputting integrated data into DNN through multiple hidden layers to classify proteins and predict protein functions. In the experiments, the F-measure value of IGP-DNN on the DIP dataset reaches 0.4436, which shows better performance.

AMP0: Species-Specific Prediction of Anti-microbial Peptides Using Zero and Few Shot Learning

Evolution of drug-resistant microbial species is one of the major challenges to global health. Development of new antimicrobial treatments such as antimicrobial peptides needs to be accelerated to combat this threat. However, the discovery of novel antimicrobial peptides is hampered by low-throughput biochemical assays. Computational techniques can be used for rapid screening of promising antimicrobial peptide candidates prior to testing in the wet lab. The vast majority of existing antimicrobial peptide predictors are non-targeted in nature, i.e., they can predict whether a given peptide sequence is antimicrobial, but they are unable to predict whether the sequence can target a particular microbial species. In this work, we have used zero and few shot machine learning to develop a targeted antimicrobial peptide activity predictor called AMP₀. The proposed predictor takes the sequence of a peptide and any N/C-termini modifications together with the genomic sequence of a microbial species to generate targeted predictions. Cross-validation results show that the proposed scheme is particularly effective for targeted antimicrobial prediction in comparison to existing approaches and can be used for screening potential antimicrobial peptides in a targeted manner with only a small number of training examples for novel species. AMP₀ webserver is available at http://ampzero.pythonanywhere.com.

Deep learning program to predict protein functions based on sequence information

•

A new deep learning program to predict protein functions in silico.

•

Requirement of nothing more than the protein sequence information.

•

A sequence segmentation to improve the efficiency of prediction.

•

Prediction of the clinical impact of mutations or polymorphisms.

Deep learning technologies have been adopted to predict the functions of newly identified proteins in silico. However, most current models are not suitable for poorly characterized proteins because they require diverse information on target proteins. We designed a binary classification deep learning program requiring only sequence information. This program was named ‘FUTUSA’ (function teller using sequence alone). It applied sequence segmentation during the sequence feature extraction process, by a convolution neural network, to train the regional sequence patterns and their relationship. This segmentation process improved the predictive performance by 49% than the full-length process. Compared with a baseline method, our approach achieved higher performance in predicting oxidoreductase activity. In addition, FUTUSA also showed dramatic performance in predicting acetyltransferase and demethylase activities. Next, we tested the possibility that FUTUSA can predict the functional consequence of point mutation. After trained for monooxygenase activity, FUTUSA successfully predicted the impact of point mutations on phenylalanine hydroxylase, which is responsible for an inherited metabolic disease PKU. This deep-learning program can be used as the first-step tool for characterizing newly identified or poorly studied proteins.•

We proposed new deep learning program to predict protein functions in silico that requires nothing more than the protein sequence information.

•

Due to application of sequence segmentation, the efficiency of prediction is improved.

•

This method makes prediction of the clinical impact of mutations or polymorphisms possible.

OUP accepted manuscript

Motivation

Protein functions are often described using the Gene Ontology (GO) which is an ontology consisting of over 50 000 classes and a large set of formal axioms. Predicting the functions of proteins is one of the key challenges in computational biology and a variety of machine learning methods have been developed for this purpose. However, these methods usually require a significant amount of training data and cannot make predictions for GO classes that have only few or no experimental annotations.

Results

We developed DeepGOZero, a machine learning model which improves predictions for functions with no or only a small number of annotations. To achieve this goal, we rely on a model-theoretic approach for learning ontology embeddings and combine it with neural networks for protein function prediction. DeepGOZero can exploit formal axioms in the GO to make zero-shot predictions, i.e., predict protein functions even if not a single protein in the training phase was associated with that function. Furthermore, the zero-shot prediction method employed by DeepGOZero is generic and can be applied whenever associations with ontology classes need to be predicted.

Availability and implementation

http://github.com/bio-ontology-research-group/deepgozero.

Supplementary information

Supplementary data are available at Bioinformatics online.

Graph representation learning for structural proteomics

The field of structural proteomics, which is focused on studying the structure-function relationship of proteins and protein complexes, is experiencing rapid growth. Since the early 2000s, structural databases such as the Protein Data Bank are storing increasing amounts of protein structural data, in addition to modeled structures becoming increasingly available. This, combined with the recent advances in graph-based machine-learning models, enables the use of protein structural data in predictive models, with the goal of creating tools that will advance our understanding of protein function. Similar to using graph learning tools to molecular graphs, which currently undergo rapid development, there is also an increasing trend in using graph learning approaches on protein structures. In this short review paper, we survey studies that use graph learning techniques on proteins, and examine their successes and shortcomings, while also discussing future directions.

Gene prediction of aging-related diseases based on DNN and Mashup

Background

At present, the bioinformatics research on the relationship between aging-related diseases and genes is mainly through the establishment of a machine learning multi-label model to classify each gene. Most of the existing methods for predicting pathogenic genes mainly rely on specific types of gene features, or directly encode multiple features with different dimensions, use the same encoder to concatenate and predict the final results, which will be subject to many limitations in the applicability of the algorithm. Possible shortcomings of the above include: incomplete coverage of gene features by a single type of biomics data, overfitting of small dimensional datasets by a single encoder, or underfitting of larger dimensional datasets.

Methods

We use the known gene disease association data and gene descriptors, such as gene ontology terms (GO), protein interaction data (PPI), PathDIP, Kyoto Encyclopedia of genes and genomes Genes (KEGG), etc, as input for deep learning to predict the association between genes and diseases. Our innovation is to use Mashup algorithm to reduce the dimensionality of PPI, GO and other large biological networks, and add new pathway data in KEGG database, and then combine a variety of biological information sources through modular Deep Neural Network (DNN) to predict the genes related to aging diseases.

Result and conclusion

The results show that our algorithm is more effective than the standard neural network algorithm (the Area Under the ROC curve from 0.8795 to 0.9153), gradient enhanced tree classifier and logistic regression classifier. In this paper, we firstly use DNN to learn the similar genes associated with the known diseases from the complex multi-dimensional feature space, and then provide the evidence that the assumed genes are associated with a certain disease.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04518-5.

Improved Prediction Model of Protein Lysine Crotonylation Sites Using Bidirectional Recurrent Neural Networks

Histone lysine crotonylation (Kcr) is a post-translational modification of histone proteins that is involved in the regulation of gene transcription, acute and chronic kidney injury, spermatogenesis, depression, cancer, and so forth. The identification of Kcr sites in proteins is important for characterizing and regulating primary biological mechanisms. The use of computational approaches such as machine learning and deep learning algorithms have emerged in recent years as the traditional wet-lab experiments are time-consuming and costly. We propose as part of this study a deep learning model based on a recurrent neural network (RNN) termed as Sohoko-Kcr for the prediction of Kcr sites. Through the embedded encoding of the peptide sequences, we investigate the efficiency of RNN-based models such as long short-term memory (LSTM), bidirectional LSTM (BiLSTM), and bidirectional gated recurrent unit (BiGRU) networks using cross-validation and independent tests. We also established the comparison between Sohoko-Kcr and other published tools to verify the efficiency of our model based on 3-fold, 5-fold, and 10-fold cross-validations using independent set tests. The results then show that the BiGRU model has consistently displayed outstanding performance and computational efficiency. Based on the proposed model, a webserver called Sohoko-Kcr was deployed for free use and is accessible at https://sohoko-research-9uu23.ondigitalocean.app.

A roadmap for metagenomic enzyme discovery

Covering: up to 2021Metagenomics has yielded massive amounts of sequencing data offering a glimpse into the biosynthetic potential of the uncultivated microbial majority. While genome-resolved information about microbial communities from nearly every environment on earth is now available, the ability to accurately predict biocatalytic functions directly from sequencing data remains challenging. Compared to primary metabolic pathways, enzymes involved in secondary metabolism often catalyze specialized reactions with diverse substrates, making these pathways rich resources for the discovery of new enzymology. To date, functional insights gained from studies on environmental DNA (eDNA) have largely relied on PCR- or activity-based screening of eDNA fragments cloned in fosmid or cosmid libraries. As an alternative, shotgun metagenomics holds underexplored potential for the discovery of new enzymes directly from eDNA by avoiding common biases introduced through PCR- or activity-guided functional metagenomics workflows. However, inferring new enzyme functions directly from eDNA is similar to searching for a 'needle in a haystack' without direct links between genotype and phenotype. The goal of this review is to provide a roadmap to navigate shotgun metagenomic sequencing data and identify new candidate biosynthetic enzymes. We cover both computational and experimental strategies to mine metagenomes and explore protein sequence space with a spotlight on natural product biosynthesis. Specifically, we compare in silico methods for enzyme discovery including phylogenetics, sequence similarity networks, genomic context, 3D structure-based approaches, and machine learning techniques. We also discuss various experimental strategies to test computational predictions including heterologous expression and screening. Finally, we provide an outlook for future directions in the field with an emphasis on meta-omics, single-cell genomics, cell-free expression systems, and sequence-independent methods.

Protein function prediction using functional inter-relationship

With the growth of high throughput sequencing techniques, the generation of protein sequences has become fast and cheap, leading to a huge increase in the number of known proteins. However, it is challenging to identify the functions being performed by these newly discovered proteins. Machine learning techniques have improved traditional methods' efficiency by suggesting relevant functions but fails to perform well when the number of functions to be predicted becomes large. In this work, we propose a machine learning-based approach to predict huge set of protein functions that use the inter-relationships between functions to improve the model's predictability. These inter-relationships of functions is used to reduce the redundancy caused by highly correlated functions. The proposed model is trained on the reduced set of non-redundant functions hindering the ambiguity caused due to inter-related functions. Here, we use two statistical approaches 1) Pearson's correlation coefficient 2) Jaccard similarity coefficient, as a measure of correlation to remove redundant functions. To have a fair evaluation of the proposed model, we recreate our original function set by inverse transforming the reduced set using the two proposed approaches: Direct mapping and Ensemble approach. The model is tested using different feature sets and function sets of biological processes and molecular functions to get promising results on DeepGO and CAFA3 dataset. The proposed model is able to predict specific functions for the test data which were unpredictable by other compared methods. The experimental models, code and other relevant data are available at https://github.com/richadhanuka/PFP-using-Functional-interrelationship.

A Deep Learning Framework for Gene Ontology Annotations With Sequence- and Network-Based Information

Knowledge of protein functions plays an important role in biology and medicine. With the rapid development of high-throughput technologies, a huge number of proteins have been discovered. However, there are a great number of proteins without functional annotations. A protein usually has multiple functions and some functions or biological processes require interactions of a plurality of proteins. Additionally, Gene Ontology provides a useful classification for protein functions and contains more than 40,000 terms. We propose a deep learning framework called DeepGOA to predict protein functions with protein sequences and protein-protein interaction (PPI) networks. For protein sequences, we extract two types of information: sequence semantic information and subsequence-based features. We use the word2vec technique to numerically represent protein sequences, and utilize a Bi-directional Long and Short Time Memory (Bi-LSTM) and multi-scale convolutional neural network (multi-scale CNN) to obtain the global and local semantic features of protein sequences, respectively. Additionally, we use the InterPro tool to scan protein sequences for extracting subsequence-based information, such as domains and motifs. Then, the information is plugged into a neural network to generate high-quality features. For the PPI network, the Deepwalk algorithm is applied to generate its embedding information of PPI. Then the two types of features are concatenated together to predict protein functions. To evaluate the performance of DeepGOA, several different evaluation methods and metrics are utilized. The experimental results show that DeepGOA outperforms DeepGO and BLAST.

DeepTrio: a ternary prediction system for protein-protein interaction using mask multiple parallel convolutional neural networks

MOTIVATION

Protein-protein interaction (PPI), as a relative property, is determined by two binding proteins, which brings a great challenge to design an expert model with an unbiased learning architecture and a superior generalization performance. Additionally, few efforts have been made to allow PPI predictors to discriminate between relative properties and intrinsic properties.

RESULTS

We present a sequence-based approach, DeepTrio, for PPI prediction using mask multiple parallel convolutional neural networks. Experimental evaluations show that DeepTrio achieves a better performance over several state-of-the-art methods in terms of various quality metrics. Besides, DeepTrio is extended to provide additional insights into the contribution of each input neuron to the prediction results.

AVAILABILITY AND IMPLEMENTATION

We provide an online application at http://bis.zju.edu.cn/deeptrio. The DeepTrio models and training data are deposited at https://github.com/huxiaoti/deeptrio.git.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Sequence-Based Prediction of Plant Protein-Protein Interactions by Combining Discrete Sine Transformation With Rotation Forest

Protein-protein interactions (PPIs) in plants are essential for understanding the regulation of biological processes. Although high-throughput technologies have been widely used to identify PPIs, they are usually laborious, expensive, and suffer from high false-positive rates. Therefore, it is imperative to develop novel computational approaches as a supplement tool to detect PPIs in plants. In this work, we presented a method, namely DST-RoF, to identify PPIs in plants by combining an ensemble learning classifier-Rotation Forest (RoF) with discrete sine transformation (DST). Specifically, plant protein sequence is firstly converted into Position-Specific Scoring Matrix (PSSM). Then, the discrete sine transformation was employed to extract effective features for obtaining the evolutionary information of proteins. Finally, these optimal features were fed into the RoF classifier for training and prediction. When performed on the plant datasets Arabidopsis, Rice, and Maize, DST-RoF yielded high prediction accuracy of 82.95%, 88.82%, and 93.70%, respectively. To further evaluate the prediction ability of our approach, we compared it with 4 state-of-the-art classifiers and 3 different feature extraction methods. Comprehensive experimental results anticipated that our method is feasible and robust for predicting potential plant-protein interacted pairs.