1
|
Shirvanizadeh N, Vihinen M. VariBench, new variation benchmark categories and data sets. FRONTIERS IN BIOINFORMATICS 2023; 3:1248732. [PMID: 37795169 PMCID: PMC10546188 DOI: 10.3389/fbinf.2023.1248732] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/27/2023] [Accepted: 09/08/2023] [Indexed: 10/06/2023] Open
Affiliation(s)
| | - Mauno Vihinen
- Department of Experimental Medical Science, Lund University, Lund, Sweden
| |
Collapse
|
2
|
Li K, Wu H, Yue Z, Sun Y, Xia C. A convolutional network and attention mechanism-based approach to predict protein-RNA binding residues. Comput Biol Chem 2023; 105:107901. [PMID: 37327559 DOI: 10.1016/j.compbiolchem.2023.107901] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/13/2023] [Revised: 05/29/2023] [Accepted: 05/31/2023] [Indexed: 06/18/2023]
Abstract
Protein-RNA interactions play a key role in various biological cellular processes, and many experimental and computational studies have been initiated to analyze their interactions. However, experimental determination is quite complex and expensive. Therefore, researchers have worked to develop efficient computational tools to detect protein-RNA binding residues. The accuracy of existing methods is limited by the features of the target and the performance of the computational models; there remains room for improvement. To solve the problem of the accurate detection of protein-RNA binding residues, we propose a convolutional network model named PBRPre based on improved MobileNet. First, by extracting the position information of the target complex and the 3-mer amino acid feature data, the position-specific scoring matrix (PSSM) is improved by using spatial neighbor smoothing processing and discrete wavelet transform to fully exploit the spatial structure information of the target and enrich the feature dataset. Second, the deep learning model MobileNet is used to integrate and optimize the potential features in the target complexes; then, by introducing the Vision Transformer (ViT) network classification layer, the deep-level information of the target is mined to enhance the processing ability of the model for global information and to improve the detection accuracy of the classifiers. The results show that the AUC value of the model can reach 0.866 in the independent testing dataset, which shows that PBRPre can effectively realize the detection of protein-RNA binding residues. All datasets and resource codes of PBRPre are available at https://github.com/linglewu/PBRPre for academic use.
Collapse
Affiliation(s)
- Ke Li
- School of Information & Computer, Anhui Agricultural University, Hefei, Anhui 230036, China; Information Materials and Intelligent Sensing Laboratory of Anhui Province, Anhui University, Hefei, Anhui 230601, China; Anhui Provincial Engineering Laboratory for Beidou Precision Agriculture Information, Anhui Agricultural University, Hefei, Anhui 230036, China.
| | - Hongwei Wu
- School of Information & Computer, Anhui Agricultural University, Hefei, Anhui 230036, China; Anhui Provincial Engineering Laboratory for Beidou Precision Agriculture Information, Anhui Agricultural University, Hefei, Anhui 230036, China
| | - Zhenyu Yue
- School of Information & Computer, Anhui Agricultural University, Hefei, Anhui 230036, China; Anhui Provincial Engineering Laboratory for Beidou Precision Agriculture Information, Anhui Agricultural University, Hefei, Anhui 230036, China
| | - Yu Sun
- School of Information & Computer, Anhui Agricultural University, Hefei, Anhui 230036, China; Anhui Provincial Engineering Laboratory for Beidou Precision Agriculture Information, Anhui Agricultural University, Hefei, Anhui 230036, China
| | - Chuan Xia
- Anhui Provincial Engineering Laboratory for Beidou Precision Agriculture Information, Anhui Agricultural University, Hefei, Anhui 230036, China
| |
Collapse
|
3
|
Bheemireddy S, Sandhya S, Srinivasan N, Sowdhamini R. Computational tools to study RNA-protein complexes. Front Mol Biosci 2022; 9:954926. [PMID: 36275618 PMCID: PMC9585174 DOI: 10.3389/fmolb.2022.954926] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/27/2022] [Accepted: 09/20/2022] [Indexed: 11/19/2022] Open
Abstract
RNA is the key player in many cellular processes such as signal transduction, replication, transport, cell division, transcription, and translation. These diverse functions are accomplished through interactions of RNA with proteins. However, protein–RNA interactions are still poorly derstood in contrast to protein–protein and protein–DNA interactions. This knowledge gap can be attributed to the limited availability of protein-RNA structures along with the experimental difficulties in studying these complexes. Recent progress in computational resources has expanded the number of tools available for studying protein-RNA interactions at various molecular levels. These include tools for predicting interacting residues from primary sequences, modelling of protein-RNA complexes, predicting hotspots in these complexes and insights into derstanding in the dynamics of their interactions. Each of these tools has its strengths and limitations, which makes it significant to select an optimal approach for the question of interest. Here we present a mini review of computational tools to study different aspects of protein-RNA interactions, with focus on overall application, development of the field and the future perspectives.
Collapse
Affiliation(s)
- Sneha Bheemireddy
- Molecular Biophysics Unit, Indian Institute of Science, Bangalore, India
| | - Sankaran Sandhya
- Department of Biotechnology, Faculty of Life and Allied Health Sciences, M.S. Ramaiah University of Applied Sciences, Bengaluru, India
- *Correspondence: Sankaran Sandhya, ; Ramanathan Sowdhamini,
| | | | - Ramanathan Sowdhamini
- Molecular Biophysics Unit, Indian Institute of Science, Bangalore, India
- National Centre for Biological Sciences, TIFR, GKVK Campus, Bangalore, India
- Institute of Bioinformatics and Applied Biotechnology, Bangalore, India
- *Correspondence: Sankaran Sandhya, ; Ramanathan Sowdhamini,
| |
Collapse
|
4
|
Ovek D, Abali Z, Zeylan ME, Keskin O, Gursoy A, Tuncbag N. Artificial intelligence based methods for hot spot prediction. Curr Opin Struct Biol 2021; 72:209-218. [PMID: 34954608 DOI: 10.1016/j.sbi.2021.11.003] [Citation(s) in RCA: 8] [Impact Index Per Article: 2.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/05/2021] [Revised: 10/07/2021] [Accepted: 11/08/2021] [Indexed: 11/29/2022]
Abstract
Proteins interact through their interfaces to fulfill essential functions in the cell. They bind to their partners in a highly specific manner and form complexes that have a profound effect on understanding the biological pathways they are involved in. Any abnormal interactions may cause diseases. Therefore, the identification of small molecules which modulate protein interactions through their interfaces has high therapeutic potential. However, discovering such molecules is challenging. Most protein-protein binding affinity is attributed to a small set of amino acids found in protein interfaces known as hot spots. Recent studies demonstrate that drug-like small molecules specifically may bind to hot spots. Therefore, hot spot prediction is crucial. As experimental data accumulates, artificial intelligence begins to be used for computational hot spot prediction. First, we review machine learning and deep learning for computational hot spot prediction and then explain the significance of hot spots toward drug design.
Collapse
Affiliation(s)
- Damla Ovek
- College of Engineering, Koc University, 34450 Istanbul, Turkey
| | - Zeynep Abali
- College of Engineering, Koc University, 34450 Istanbul, Turkey
| | | | - Ozlem Keskin
- College of Engineering, Koc University, 34450 Istanbul, Turkey.
| | - Attila Gursoy
- College of Engineering, Koc University, 34450 Istanbul, Turkey.
| | - Nurcan Tuncbag
- College of Engineering, Koc University, 34450 Istanbul, Turkey; School of Medicine, Koc University, 34450 Istanbul, Turkey.
| |
Collapse
|
5
|
Nguyen TB, Myung Y, de Sá AGC, Pires DEV, Ascher DB. mmCSM-NA: accurately predicting effects of single and multiple mutations on protein-nucleic acid binding affinity. NAR Genom Bioinform 2021; 3:lqab109. [PMID: 34805992 PMCID: PMC8600011 DOI: 10.1093/nargab/lqab109] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/11/2021] [Revised: 09/20/2021] [Accepted: 10/27/2021] [Indexed: 02/02/2023] Open
Abstract
While protein-nucleic acid interactions are pivotal for many crucial biological processes, limited experimental data has made the development of computational approaches to characterise these interactions a challenge. Consequently, most approaches to understand the effects of missense mutations on protein-nucleic acid affinity have focused on single-point mutations and have presented a limited performance on independent data sets. To overcome this, we have curated the largest dataset of experimentally measured effects of mutations on nucleic acid binding affinity to date, encompassing 856 single-point mutations and 141 multiple-point mutations across 155 experimentally solved complexes. This was used in combination with an optimized version of our graph-based signatures to develop mmCSM-NA (http://biosig.unimelb.edu.au/mmcsm_na), the first scalable method capable of quantitatively and accurately predicting the effects of multiple-point mutations on nucleic acid binding affinities. mmCSM-NA obtained a Pearson's correlation of up to 0.67 (RMSE of 1.06 Kcal/mol) on single-point mutations under cross-validation, and up to 0.65 on independent non-redundant datasets of multiple-point mutations (RMSE of 1.12 kcal/mol), outperforming similar tools. mmCSM-NA is freely available as an easy-to-use web-server and API. We believe it will be an invaluable tool to shed light on the role of mutations affecting protein-nucleic acid interactions in diseases.
Collapse
Affiliation(s)
- Thanh Binh Nguyen
- Computational Biology and Clinical Informatics, Baker Heart and Diabetes Institute, Melbourne, Victoria, Australia
- School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Australia
- Systems and Computational Biology, Bio21 Institute, University of Melbourne, Melbourne, Victoria, Australia
| | - Yoochan Myung
- Computational Biology and Clinical Informatics, Baker Heart and Diabetes Institute, Melbourne, Victoria, Australia
- Systems and Computational Biology, Bio21 Institute, University of Melbourne, Melbourne, Victoria, Australia
| | - Alex G C de Sá
- Computational Biology and Clinical Informatics, Baker Heart and Diabetes Institute, Melbourne, Victoria, Australia
- School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Australia
- Systems and Computational Biology, Bio21 Institute, University of Melbourne, Melbourne, Victoria, Australia
| | - Douglas E V Pires
- Computational Biology and Clinical Informatics, Baker Heart and Diabetes Institute, Melbourne, Victoria, Australia
- Systems and Computational Biology, Bio21 Institute, University of Melbourne, Melbourne, Victoria, Australia
- School of Computing and Information Systems, University of Melbourne, Melbourne, Victoria, Australia
| | - David B Ascher
- Computational Biology and Clinical Informatics, Baker Heart and Diabetes Institute, Melbourne, Victoria, Australia
- School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Australia
- Systems and Computational Biology, Bio21 Institute, University of Melbourne, Melbourne, Victoria, Australia
- Department of Biochemistry, University of Cambridge, Cambridge, UK
| |
Collapse
|
6
|
Liu J, Liu S, Liu C, Zhang Y, Pan Y, Wang Z, Wang J, Wen T, Deng L. Nabe: an energetic database of amino acid mutations in protein-nucleic acid binding interfaces. DATABASE-THE JOURNAL OF BIOLOGICAL DATABASES AND CURATION 2021; 2021:6352208. [PMID: 34389843 PMCID: PMC8363842 DOI: 10.1093/database/baab050] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 04/01/2021] [Revised: 07/23/2021] [Accepted: 07/29/2021] [Indexed: 12/17/2022]
Abstract
Protein–nucleic acid complexes play essential roles in regulating transcription, translation, DNA replication, repair and recombination, RNA processing and translocation. Site-directed mutagenesis has been extremely useful in understanding the principles of protein–DNA and protein–RNA interactions, and experimentally determined mutagenesis data are prerequisites for designing effective algorithms for predicting the binding affinity change upon mutation. However, a vital challenge in this area is the lack of sufficient public experimentally recognized mutation data, which leads to difficulties in developing computational prediction methods. In this article, we present Nabe, an integrated database of amino acid mutations and their effects on the binding free energy in protein–DNA and protein–RNA interactions for which binding affinities have been experimentally determined. Compared with existing databases and data sets, Nabe is the largest protein–nucleic acid mutation database, containing 2506 mutations in 473 protein–DNA and protein–RNA complexes, and of that 1751 are alanine mutations in 405 protein–nucleic acid complexes. For researchers to conveniently utilize the data, Nabe assembles protein–DNA and protein–RNA benchmark databases by adopting the data-processing procedures in the majority of models. To further facilitate users to query data, Nabe provides a searchable and graphical web page. Database URL: http://nabe.denglab.org
Collapse
Affiliation(s)
- Junyi Liu
- School of Computer Science and Engineering, Central South University, 22 Shaoshan South Road, Changsha 410075, China.,Viterbi School of Engineering, University of Southern California, 3650 McClintock Ave. OHE 106, Los Angeles, CA 90089, USA
| | - Siyu Liu
- School of Computer Science and Engineering, Central South University, 22 Shaoshan South Road, Changsha 410075, China
| | - Chenzhe Liu
- School of Computer Science and Engineering, Central South University, 22 Shaoshan South Road, Changsha 410075, China
| | - Yaping Zhang
- School of Computer Science and Engineering, Central South University, 22 Shaoshan South Road, Changsha 410075, China
| | - Yuliang Pan
- School of Computer Science and Engineering, Central South University, 22 Shaoshan South Road, Changsha 410075, China
| | - Zixiang Wang
- School of Computer Science and Engineering, Central South University, 22 Shaoshan South Road, Changsha 410075, China
| | - Jiacheng Wang
- School of Computer Science and Engineering, Central South University, 22 Shaoshan South Road, Changsha 410075, China
| | - Ting Wen
- School of Computer Science and Engineering, Central South University, 22 Shaoshan South Road, Changsha 410075, China
| | - Lei Deng
- School of Computer Science and Engineering, Central South University, 22 Shaoshan South Road, Changsha 410075, China
| |
Collapse
|
7
|
Mei LC, Hao GF, Yang GF. Computational methods for predicting hotspots at protein-RNA interfaces. WILEY INTERDISCIPLINARY REVIEWS-RNA 2021; 13:e1675. [PMID: 34080311 DOI: 10.1002/wrna.1675] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/11/2021] [Revised: 05/13/2021] [Accepted: 05/14/2021] [Indexed: 11/10/2022]
Abstract
Protein-RNA interactions play essential roles in many critical biological events. A comprehensive understanding of the mechanisms underlying these interactions is helpful when studying cellular activities and therapeutic applications. Hotspots are a small portion of residues contributing much toward protein-RNA binding affinity. In pharmaceutical research, the hotspot residues are seen as the best option for designing small molecules to target proteins of therapeutic interest. With the accumulation of experimental data about protein-RNA interactions, computational methods have been produced for hotspot prediction on a large scale. In this review, we first present an overview of the existing databases for protein-RNA binding data. Furthermore, we outline the most adopted computational methods for hotspots prediction in protein-RNA interactions. Finally, we discuss the applications of hotspot prediction. This article is categorized under: RNA Interactions with Proteins and Other Molecules > Protein-RNA Recognition RNA Interactions with Proteins and Other Molecules > Protein-RNA Interactions: Functional Implications RNA Methods > RNA Analyses In Vitro and In Silico.
Collapse
Affiliation(s)
- Long-Can Mei
- Key Laboratory of Pesticide and Chemical Biology, Ministry of Education, College of Chemistry, Central China Normal University, Wuhan, China.,International Joint Research Center for Intelligent Biosensor Technology and Health, Central China Normal University, Wuhan, China
| | - Ge-Fei Hao
- Key Laboratory of Pesticide and Chemical Biology, Ministry of Education, College of Chemistry, Central China Normal University, Wuhan, China.,International Joint Research Center for Intelligent Biosensor Technology and Health, Central China Normal University, Wuhan, China.,State Key Laboratory Breeding Base of Green Pesticide and Agricultural Bioengineering, Key Laboratory of Green Pesticide and Agricultural Bioengineering, Ministry of Education, Research and Development Center for Fine Chemicals, Guizhou University, Guiyang, China
| | - Guang-Fu Yang
- Key Laboratory of Pesticide and Chemical Biology, Ministry of Education, College of Chemistry, Central China Normal University, Wuhan, China.,International Joint Research Center for Intelligent Biosensor Technology and Health, Central China Normal University, Wuhan, China.,Collaborative Innovation Center of Chemical Science and Engineering, Tianjin, China
| |
Collapse
|
8
|
Jiang Y, Liu HF, Liu R. Systematic comparison and prediction of the effects of missense mutations on protein-DNA and protein-RNA interactions. PLoS Comput Biol 2021; 17:e1008951. [PMID: 33872313 PMCID: PMC8084330 DOI: 10.1371/journal.pcbi.1008951] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/19/2021] [Revised: 04/29/2021] [Accepted: 04/08/2021] [Indexed: 12/30/2022] Open
Abstract
The binding affinities of protein-nucleic acid interactions could be altered due to missense mutations occurring in DNA- or RNA-binding proteins, therefore resulting in various diseases. Unfortunately, a systematic comparison and prediction of the effects of mutations on protein-DNA and protein-RNA interactions (these two mutation classes are termed MPDs and MPRs, respectively) is still lacking. Here, we demonstrated that these two classes of mutations could generate similar or different tendencies for binding free energy changes in terms of the properties of mutated residues. We then developed regression algorithms separately for MPDs and MPRs by introducing novel geometric partition-based energy features and interface-based structural features. Through feature selection and ensemble learning, similar computational frameworks that integrated energy- and nonenergy-based models were established to estimate the binding affinity changes resulting from MPDs and MPRs, but the selected features for the final models were different and therefore reflected the specificity of these two mutation classes. Furthermore, the proposed methodology was extended to the identification of mutations that significantly decreased the binding affinities. Extensive validations indicated that our algorithm generally performed better than the state-of-the-art methods on both the regression and classification tasks. The webserver and software are freely available at http://liulab.hzau.edu.cn/PEMPNI and https://github.com/hzau-liulab/PEMPNI. Protein-nucleic acid interactions play important roles in various cellular processes. Missense mutations occurring in DNA- or RNA-binding proteins (termed MPDs and MPRs, respectively) could change the binding affinities of these interactions. Previous studies have compared protein-DNA and protein-RNA interactions from multifaceted viewpoints, but less attention has been given to the similarities and specific differences between the effects of MPDs and MPRs and between the methodologies for predicting the affinity changes induced by the two mutation classes. Therefore, we systematically compared their impacts and demonstrated that MPDs and MPRs could have specific preferences for binding affinity changes. These observations motivated us to construct regression models separately for MPDs and MPRs by introducing novel energy and nonenergy descriptors. Although similar frameworks were developed to estimate these two categories of mutation effects, different descriptors were selected in the regression models and further revealed the specificity of mutation classes. The interplay between the energy and nonenergy modules effectively improved prediction performance. Our algorithm can also be adopted to disentangle mutations significantly decreasing binding affinities from other mutations.
Collapse
Affiliation(s)
- Yao Jiang
- Hubei Key Laboratory of Agricultural Bioinformatics, College of Informatics, Huazhong Agricultural University, Wuhan, P. R. China
| | - Hui-Fang Liu
- Hubei Key Laboratory of Agricultural Bioinformatics, College of Informatics, Huazhong Agricultural University, Wuhan, P. R. China
| | - Rong Liu
- Hubei Key Laboratory of Agricultural Bioinformatics, College of Informatics, Huazhong Agricultural University, Wuhan, P. R. China
| |
Collapse
|