INGA 2.0: improving protein function prediction for the dark proteome

A CNN-CBAM-BIGRU model for protein function prediction

Understanding a protein's function based solely on its amino acid sequence is a crucial but intricate task in bioinformatics. Traditionally, this challenge has proven difficult. However, recent years have witnessed the rise of deep learning as a powerful tool, achieving significant success in protein function prediction. Their strength lies in their ability to automatically learn informative features from protein sequences, which can then be used to predict the protein's function. This study builds upon these advancements by proposing a novel model: CNN-CBAM+BiGRU. It incorporates a Convolutional Block Attention Module (CBAM) alongside BiGRUs. CBAM acts as a spotlight, guiding the CNN to focus on the most informative parts of the protein data, leading to more accurate feature extraction. BiGRUs, a type of Recurrent Neural Network (RNN), excel at capturing long-range dependencies within the protein sequence, which are essential for accurate function prediction. The proposed model integrates the strengths of both CNN-CBAM and BiGRU. This study's findings, validated through experimentation, showcase the effectiveness of this combined approach. For the human dataset, the suggested method outperforms the CNN-BIGRU+ATT model by +1.0 % for cellular components, +1.1 % for molecular functions, and +0.5 % for biological processes. For the yeast dataset, the suggested method outperforms the CNN-BIGRU+ATT model by +2.4 % for the cellular component, +1.2 % for molecular functions, and +0.6 % for biological processes.

Large-scale predicting protein functions through heterogeneous feature fusion

As the volume of protein sequence and structure data grows rapidly, the functions of the overwhelming majority of proteins cannot be experimentally determined. Automated annotation of protein function at a large scale is becoming increasingly important. Existing computational prediction methods are typically based on expanding the relatively small number of experimentally determined functions to large collections of proteins with various clues, including sequence homology, protein-protein interaction, gene co-expression, etc. Although there has been some progress in protein function prediction in recent years, the development of accurate and reliable solutions still has a long way to go. Here we exploit AlphaFold predicted three-dimensional structural information, together with other non-structural clues, to develop a large-scale approach termed PredGO to annotate Gene Ontology (GO) functions for proteins. We use a pre-trained language model, geometric vector perceptrons and attention mechanisms to extract heterogeneous features of proteins and fuse these features for function prediction. The computational results demonstrate that the proposed method outperforms other state-of-the-art approaches for predicting GO functions of proteins in terms of both coverage and accuracy. The improvement of coverage is because the number of structures predicted by AlphaFold is greatly increased, and on the other hand, PredGO can extensively use non-structural information for functional prediction. Moreover, we show that over 205 000 ($\sim $100%) entries in UniProt for human are annotated by PredGO, over 186 000 ($\sim $90%) of which are based on predicted structure. The webserver and database are available at http://predgo.denglab.org/.

NetGO 3.0: Protein Language Model Improves Large-scale Functional Annotations

As one of the state-of-the-art automated function prediction (AFP) methods, NetGO 2.0 integrates multi-source information to improve the performance. However, it mainly utilizes the proteins with experimentally supported functional annotations without leveraging valuable information from a vast number of unannotated proteins. Recently, protein language models have been proposed to learn informative representations [e.g., Evolutionary Scale Modeling (ESM)-1b embedding] from protein sequences based on self-supervision. Here, we represented each protein by ESM-1b and used logistic regression (LR) to train a new model, LR-ESM, for AFP. The experimental results showed that LR-ESM achieved comparable performance with the best-performing component of NetGO 2.0. Therefore, by incorporating LR-ESM into NetGO 2.0, we developed NetGO 3.0 to improve the performance of AFP extensively. NetGO 3.0 is freely accessible at https://dmiip.sjtu.edu.cn/ng3.0.

Machine learning for the identification of respiratory viral attachment machinery from sequences data

At the outset of an emergent viral respiratory pandemic, sequence data is among the first molecular information available. As viral attachment machinery is a key target for therapeutic and prophylactic interventions, rapid identification of viral "spike" proteins from sequence can significantly accelerate the development of medical countermeasures. For six families of respiratory viruses, covering the vast majority of airborne and droplet-transmitted diseases, host cell entry is mediated by the binding of viral surface glycoproteins that interact with a host cell receptor. In this report it is shown that sequence data for an unknown virus belonging to one of the six families above provides sufficient information to identify the protein(s) responsible for viral attachment. Random forest models that take as input a set of respiratory viral sequences can classify the protein as "spike" vs. non-spike based on predicted secondary structure elements alone (with 97.3% correctly classified) or in combination with N-glycosylation related features (with 97.0% correctly classified). Models were validated through 10-fold cross-validation, bootstrapping on a class-balanced set, and an out-of-sample extra-familial validation set. Surprisingly, we showed that secondary structural elements and N-glycosylation features were sufficient for model generation. The ability to rapidly identify viral attachment machinery directly from sequence data holds the potential to accelerate the design of medical countermeasures for future pandemics. Furthermore, this approach may be extendable for the identification of other potential viral targets and for viral sequence annotation in general in the future.

A systematic review of state-of-the-art strategies for machine learning-based protein function prediction

New drug discovery is inseparable from the discovery of drug targets, and the vast majority of the known targets are proteins. At the same time, proteins are essential structural and functional elements of living cells necessary for the maintenance of all forms of life. Therefore, protein functions have become the focus of many pharmacological and biological studies. Traditional experimental techniques are no longer adequate for rapidly growing annotation of protein sequences, and approaches to protein function prediction using computational methods have emerged and flourished. A significant trend has been to use machine learning to achieve this goal. In this review, approaches to protein function prediction based on the sequence, structure, protein-protein interaction (PPI) networks, and fusion of multi-information sources are discussed. The current status of research on protein function prediction using machine learning is considered, and existing challenges and prominent breakthroughs are discussed to provide ideas and methods for future studies.

I-TASSER-MTD: a deep-learning-based platform for multi-domain protein structure and function prediction

Most proteins in cells are composed of multiple folding units (or domains) to perform complex functions in a cooperative manner. Relative to the rapid progress in single-domain structure prediction, there are few effective tools available for multi-domain protein structure assembly, mainly due to the complexity of modeling multi-domain proteins, which involves higher degrees of freedom in domain-orientation space and various levels of continuous and discontinuous domain assembly and linker refinement. To meet the challenge and the high demand of the community, we developed I-TASSER-MTD to model the structures and functions of multi-domain proteins through a progressive protocol that combines sequence-based domain parsing, single-domain structure folding, inter-domain structure assembly and structure-based function annotation in a fully automated pipeline. Advanced deep-learning models have been incorporated into each of the steps to enhance both the domain modeling and inter-domain assembly accuracy. The protocol allows for the incorporation of experimental cross-linking data and cryo-electron microscopy density maps to guide the multi-domain structure assembly simulations. I-TASSER-MTD is built on I-TASSER but substantially extends its ability and accuracy in modeling large multi-domain protein structures and provides meaningful functional insights for the targets at both the domain- and full-chain levels from the amino acid sequence alone.

A tensor-based bi-random walks model for protein function prediction

Background

The accurate characterization of protein functions is critical to understanding life at the molecular level and has a huge impact on biomedicine and pharmaceuticals. Computationally predicting protein function has been studied in the past decades. Plagued by noise and errors in protein–protein interaction (PPI) networks, researchers have undertaken to focus on the fusion of multi-omics data in recent years. A data model that appropriately integrates network topologies with biological data and preserves their intrinsic characteristics is still a bottleneck and an aspirational goal for protein function prediction.

Results

In this paper, we propose the RWRT (Random Walks with Restart on Tensor) method to accomplish protein function prediction by applying bi-random walks on the tensor. RWRT firstly constructs a functional similarity tensor by combining protein interaction networks with multi-omics data derived from domain annotation and protein complex information. After this, RWRT extends the bi-random walks algorithm from a two-dimensional matrix to the tensor for scoring functional similarity between proteins. Finally, RWRT filters out possible pretenders based on the concept of cohesiveness coefficient and annotates target proteins with functions of the remaining functional partners. Experimental results indicate that RWRT performs significantly better than the state-of-the-art methods and improves the area under the receiver-operating curve (AUROC) by no less than 18%.

Conclusions

The functional similarity tensor offers us an alternative, in that it is a collection of networks sharing the same nodes; however, the edges belong to different categories or represent interactions of different nature. We demonstrate that the tensor-based random walk model can not only discover more partners with similar functions but also free from the constraints of errors in protein interaction networks effectively. We believe that the performance of function prediction depends greatly on whether we can extract and exploit proper functional similarity information on protein correlations.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04747-2.

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.

Identification of Intrinsically Disordered Proteins and Regions in a Non-Model Insect Species Ostrinia nubilalis (Hbn.)

Research in previous decades has shown that intrinsically disordered proteins (IDPs) and regions in proteins (IDRs) are as ubiquitous as highly ordered proteins. Despite this, research on IDPs and IDRs still has many gaps left to fill. Here, we present an approach that combines wet lab methods with bioinformatics tools to identify and analyze intrinsically disordered proteins in a non-model insect species that is cold-hardy. Due to their known resilience to the effects of extreme temperatures, these proteins likely play important roles in this insect's adaptive mechanisms to sub-zero temperatures. The approach involves IDP enrichment by sample heating and double-digestion of proteins, followed by peptide and protein identification. Next, proteins are bioinformatically analyzed for disorder content, presence of long disordered regions, amino acid composition, and processes they are involved in. Finally, IDP detection is validated with an in-house 2D PAGE. In total, 608 unique proteins were identified, with 39 being mostly disordered, 100 partially disordered, 95 nearly ordered, and 374 ordered. One-third contain at least one long disordered segment. Functional information was available for only 90 proteins with intrinsic disorders out of 312 characterized proteins. Around half of the 90 proteins are cytoskeletal elements or involved in translational processes.

PANNZER-A practical tool for protein function prediction

The facility of next-generation sequencing has led to an explosion of gene catalogs for novel genomes, transcriptomes and metagenomes, which are functionally uncharacterized. Computational inference has emerged as a necessary substitute for first-hand experimental evidence. PANNZER (Protein ANNotation with Z-scoRE) is a high-throughput functional annotation web server that stands out among similar publically accessible web servers in supporting submission of up to 100,000 protein sequences at once and providing both Gene Ontology (GO) annotations and free text description predictions. Here, we demonstrate the use of PANNZER and discuss future plans and challenges. We present two case studies to illustrate problems related to data quality and method evaluation. Some commonly used evaluation metrics and evaluation datasets promote methods that favor unspecific and broad functional classes over more informative and specific classes. We argue that this can bias the development of automated function prediction methods. The PANNZER web server and source code are available at http://ekhidna2.biocenter.helsinki.fi/sanspanz/.

Semantic similarity and machine learning with ontologies

Ontologies have long been employed in the life sciences to formally represent and reason over domain knowledge and they are employed in almost every major biological database. Recently, ontologies are increasingly being used to provide background knowledge in similarity-based analysis and machine learning models. The methods employed to combine ontologies and machine learning are still novel and actively being developed. We provide an overview over the methods that use ontologies to compute similarity and incorporate them in machine learning methods; in particular, we outline how semantic similarity measures and ontology embeddings can exploit the background knowledge in ontologies and how ontologies can provide constraints that improve machine learning models. The methods and experiments we describe are available as a set of executable notebooks, and we also provide a set of slides and additional resources at https://github.com/bio-ontology-research-group/machine-learning-with-ontologies.

NetGO 2.0: improving large-scale protein function prediction with massive sequence, text, domain, family and network information

With the explosive growth of protein sequences, large-scale automated protein function prediction (AFP) is becoming challenging. A protein is usually associated with dozens of gene ontology (GO) terms. Therefore, AFP is regarded as a problem of large-scale multi-label classification. Under the learning to rank (LTR) framework, our previous NetGO tool integrated massive networks and multi-type information about protein sequences to achieve good performance by dealing with all possible GO terms (>44 000). In this work, we propose the updated version as NetGO 2.0, which further improves the performance of large-scale AFP. NetGO 2.0 also incorporates literature information by logistic regression and deep sequence information by recurrent neural network (RNN) into the framework. We generate datasets following the critical assessment of functional annotation (CAFA) protocol. Experiment results show that NetGO 2.0 outperformed NetGO significantly in biological process ontology (BPO) and cellular component ontology (CCO). In particular, NetGO 2.0 achieved a 12.6% improvement over NetGO in terms of area under precision-recall curve (AUPR) in BPO and around 2.6% in terms of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} }{}$\mathbf {F_{max}}$\end{document} in CCO. These results demonstrate the benefits of incorporating text and deep sequence information for the functional annotation of BPO and CCO. The NetGO 2.0 web server is freely available at http://issubmission.sjtu.edu.cn/ng2/.

A thorough analysis of the contribution of experimental, derived and sequence-based predicted protein-protein interactions for functional annotation of proteins

Physical interaction between two proteins is strong evidence that the proteins are involved in the same biological process, making Protein-Protein Interaction (PPI) networks a valuable data resource for predicting the cellular functions of proteins. However, PPI networks are largely incomplete for non-model species. Here, we tested to what extent these incomplete networks are still useful for genome-wide function prediction. We used two network-based classifiers to predict Biological Process Gene Ontology terms from protein interaction data in four species: Saccharomyces cerevisiae, Escherichia coli, Arabidopsis thaliana and Solanum lycopersicum (tomato). The classifiers had reasonable performance in the well-studied yeast, but performed poorly in the other species. We showed that this poor performance can be considerably improved by adding edges predicted from various data sources, such as text mining, and that associations from the STRING database are more useful than interactions predicted by a neural network from sequence-based features.

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.

NPF:network propagation for protein function prediction

Background

The accurate annotation of protein functions is of great significance in elucidating the phenomena of life, treating disease and developing new medicines. Various methods have been developed to facilitate the prediction of these functions by combining protein interaction networks (PINs) with multi-omics data. However, it is still challenging to make full use of multiple biological to improve the performance of functions annotation.

Results

We presented NPF (Network Propagation for Functions prediction), an integrative protein function predicting framework assisted by network propagation and functional module detection, for discovering interacting partners with similar functions to target proteins. NPF leverages knowledge of the protein interaction network architecture and multi-omics data, such as domain annotation and protein complex information, to augment protein-protein functional similarity in a propagation manner. We have verified the great potential of NPF for accurately inferring protein functions. According to the comprehensive evaluation of NPF, it delivered a better performance than other competing methods in terms of leave-one-out cross-validation and ten-fold cross validation.

Conclusions

We demonstrated that network propagation, together with multi-omics data, can both discover more partners with similar function, and is unconstricted by the “small-world” feature of protein interaction networks. We conclude that the performance of function prediction depends greatly on whether we can extract and exploit proper functional information of similarity from protein correlations.

Predicting Functions of Uncharacterized Human Proteins: From Canonical to Proteoforms

Despite tremendous efforts in genomics, transcriptomics, and proteomics communities, there is still no comprehensive data about the exact number of protein-coding genes, translated proteoforms, and their function. In addition, by now, we lack functional annotation for 1193 genes, where expression was confirmed at the proteomic level (uPE1 proteins). We re-analyzed results of AP-MS experiments from the BioPlex 2.0 database to predict functions of uPE1 proteins and their splice forms. By building a protein-protein interaction network for 12 ths. identified proteins encoded by 11 ths. genes, we were able to predict Gene Ontology categories for a total of 387 uPE1 genes. We predicted different functions for canonical and alternatively spliced forms for four uPE1 genes. In total, functional differences were revealed for 62 proteoforms encoded by 31 genes. Based on these results, it can be carefully concluded that the dynamics and versatility of the interactome is ensured by changing the dominant splice form. Overall, we propose that analysis of large-scale AP-MS experiments performed for various cell lines and under various conditions is a key to understanding the full potential of genes role in cellular processes.

DisProt: intrinsic protein disorder annotation in 2020

The Database of Protein Disorder (DisProt, URL: https://disprot.org) provides manually curated annotations of intrinsically disordered proteins from the literature. Here we report recent developments with DisProt (version 8), including the doubling of protein entries, a new disorder ontology, improvements of the annotation format and a completely new website. The website includes a redesigned graphical interface, a better search engine, a clearer API for programmatic access and a new annotation interface that integrates text mining technologies. The new entry format provides a greater flexibility, simplifies maintenance and allows the capture of more information from the literature. The new disorder ontology has been formalized and made interoperable by adopting the OWL format, as well as its structure and term definitions have been improved. The new annotation interface has made the curation process faster and more effective. We recently showed that new DisProt annotations can be effectively used to train and validate disorder predictors. We believe the growth of DisProt will accelerate, contributing to the improvement of function and disorder predictors and therefore to illuminate the ‘dark’ proteome.

SDN2GO: An Integrated Deep Learning Model for Protein Function Prediction

The assignment of function to proteins at a large scale is essential for understanding the molecular mechanism of life. However, only a very small percentage of the more than 179 million proteins in UniProtKB have Gene Ontology (GO) annotations supported by experimental evidence. In this paper, we proposed an integrated deep-learning-based classification model, named SDN2GO, to predict protein functions. SDN2GO applies convolutional neural networks to learn and extract features from sequences, protein domains, and known PPI networks, and then utilizes a weight classifier to integrate these features and achieve accurate predictions of GO terms. We constructed the training set and the independent test set according to the time-delayed principle of the Critical Assessment of Function Annotation (CAFA) and compared it with two highly competitive methods and the classic BLAST method on the independent test set. The results show that our method outperforms others on each sub-ontology of GO. We also investigated the performance of using protein domain information. We learned from the Natural Language Processing (NLP) to process domain information and pre-trained a deep learning sub-model to extract the comprehensive features of domains. The experimental results demonstrate that the domain features we obtained are much improved the performance of our model. Our deep learning models together with the data pre-processing scripts are publicly available as an open source software at https://github.com/Charrick/SDN2GO.

The CAFA challenge reports improved protein function prediction and new functional annotations for hundreds of genes through experimental screens

BACKGROUND

The Critical Assessment of Functional Annotation (CAFA) is an ongoing, global, community-driven effort to evaluate and improve the computational annotation of protein function.

RESULTS

Here, we report on the results of the third CAFA challenge, CAFA3, that featured an expanded analysis over the previous CAFA rounds, both in terms of volume of data analyzed and the types of analysis performed. In a novel and major new development, computational predictions and assessment goals drove some of the experimental assays, resulting in new functional annotations for more than 1000 genes. Specifically, we performed experimental whole-genome mutation screening in Candida albicans and Pseudomonas aureginosa genomes, which provided us with genome-wide experimental data for genes associated with biofilm formation and motility. We further performed targeted assays on selected genes in Drosophila melanogaster, which we suspected of being involved in long-term memory.

CONCLUSION

We conclude that while predictions of the molecular function and biological process annotations have slightly improved over time, those of the cellular component have not. Term-centric prediction of experimental annotations remains equally challenging; although the performance of the top methods is significantly better than the expectations set by baseline methods in C. albicans and D. melanogaster, it leaves considerable room and need for improvement. Finally, we report that the CAFA community now involves a broad range of participants with expertise in bioinformatics, biological experimentation, biocuration, and bio-ontologies, working together to improve functional annotation, computational function prediction, and our ability to manage big data in the era of large experimental screens.