Article
Biochemical Research Methods
Binjie Guo, Hanyu Zheng, Haohan Jiang, Xiaodan Li, Naiyu Guan, Yanming Zuo, Yicheng Zhang, Hengfu Yang, Xuhua Wang
Summary: Due to the lack of an efficient method to represent multimodal information of proteins, predicting compound-protein binding affinity (CPA) has low accuracy with machine-learning methods. In this study, we develop a novel end-to-end architecture called FeatNN to represent both structure and sequence features of proteins, and optimize mathematical models for CPA prediction using a coevolutionary strategy. We also propose a rational method to optimize the accuracy and generalization ability of FeatNN in CPA prediction tasks by utilizing both high- and low-quality databases. FeatNN outperforms the state-of-the-art baseline in virtual drug evaluation tasks, demonstrating its feasibility for practical use.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Biochemical Research Methods
Zhong-Ze Yu, Chun-Xiang Peng, Jun Liu, Biao Zhang, Xiao-Gen Zhou, Gui-Jun Zhang
Summary: This study develops a sequence-based protein domain boundary prediction method called DomBpred. It classifies the input sequence as either a single-domain or multi-domain protein using an effective sequence metric and a constructed single-domain sequence library. For multi-domain proteins, a domain-residue clustering algorithm is used to cluster residues based on their distances. The unclassified residues and residues at the cluster edge are adjusted using secondary structure information to create potential cut points. A domain boundary scoring function is then used to evaluate these potential cut points and generate the domain boundary.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Ahmad Haseeb, Maryam Bashir, Aamir Wali
Summary: This study proposes BERTDom, a method for segmenting protein sequences, which utilizes biological language modeling and deep learning techniques to improve protein domain boundary prediction.
COMPUTING AND INFORMATICS
(2023)
Article
Biochemical Research Methods
Fengqi Ge, Chunxiang Peng, Xinyue Cui, Yuhao Xia, Guijun Zhang
Summary: AlphaFold2 achieves breakthrough in protein structure prediction with an end-to-end deep learning method, accurately predicting single-domain proteins. However, accuracy in predicting full-chain proteins is lower due to incorrect domain interactions. This study introduces DeepIDDP, an inter-domain distance prediction method, which incorporates attention mechanisms and new inter-domain features to enhance capturing domain interactions. Integration of DeepIDDP into the SADA domain assembly method improves inter-domain distance prediction accuracy by 11.3% and 21.6% compared to trRosettaX and trRosetta, and the domain assembly model outperforms SADA by 2.5%. Additionally, using DeepIDDP to reassemble human multi-domain protein models enhances average TM-score by 11.8%. The online server can be found at .
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Cell Biology
Sharzil Haris Khan, Hilal Tayara, Kil To Chong
Summary: Protein-protein interactions are essential for biological processes and can help in drug development. Computational methods that use sequential information of proteins have been proposed to predict binding sites. A neural network-based model called ProB-site uses evolutionary and structural information to generate feature sets, which are then classified.
Article
Biochemical Research Methods
Sajid Mahmud, Zhiye Guo, Farhan Quadir, Jian Liu, Jianlin Cheng
Summary: In this study, a deep learning method called DistDom is developed to accurately predict protein domain boundaries using 1D sequence features and predicted 2D inter-residue distance map. The method outperforms the state-of-the-art techniques in terms of accuracy and F1 measure on both single-domain and multi-domain proteins.
BMC BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Yang Li, Guanyu Qiao, Xin Gao, Guohua Wang
Summary: The identification of drug-target interactions (DTIs) is crucial in drug discovery and repositioning. However, limited and expensive labeled data hinder the accuracy of traditional methods. In this study, we propose an end-to-end supervised graph co-contrastive learning model that leverages contrastive learning to improve the accuracy and reliability of DTI prediction.
Article
Biochemical Research Methods
Yiming Li, Min Zeng, Yifan Wu, Yaohang Li, Min Li
Summary: This paper proposes EP-EDL, an ensemble deep learning model that uses only protein sequence information to predict human essential proteins. EP-EDL outperforms state-of-the-art sequence-based methods and provides a practical and flexible way for accurate prediction of essential proteins.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Review
Genetics & Heredity
Cornille Amandine, Dieter Ebert, Eva Stukenbrock, Ricardo C. Rodriguez de la Vega, Peter Tiff, Daniel Croll, Aurelien Tellier
Summary: Coevolutionary interactions are a common driver of adaptation, but little is known about the genomic processes underlying coevolution in an ecological context. This article reviews recent advances in coevolutionary theory and genomics, and proposes a practical guide to understanding the dynamics of coevolution using an ecological genomics approach.
TRENDS IN GENETICS
(2022)
Article
Genetics & Heredity
Zixuan Meng, Linai Kuang, Zhiping Chen, Zhen Zhang, Yihong Tan, Xueyong Li, Lei Wang
Summary: A prediction model called WPDINM is proposed in this study to detect key proteins based on a novel weighted protein-domain interaction network. Experimental results show that WPDINM achieves significantly higher predictive accuracy for key protein identification compared to traditional competing measures.
FRONTIERS IN GENETICS
(2021)
Article
Multidisciplinary Sciences
Yang Li, Zheng Wang, Li-Ping Li, Zhu-Hong You, Wen-Zhun Huang, Xin-Ke Zhan, Yan-Bin Wang
Summary: This study introduces a computational method for predicting PPIs based on protein sequence information, utilizing a combination of OLPP and RoF models to identify non-interacting and interacting protein pairs with high accuracy on Yeast and Human datasets. The proposed method serves as a valuable tool in accelerating the resolution of key problems in proteomics.
SCIENTIFIC REPORTS
(2021)
Article
Genetics & Heredity
Kaustav Sengupta, Sovan Saha, Anup Kumar Halder, Piyali Chatterjee, Mita Nasipuri, Subhadip Basu, Dariusz Plewczynski
Summary: Protein function prediction is an important field in biology and computer science. A new method called PFP-GO is proposed for predicting protein function using information from multiple sources. Performance analysis shows that PFP-GO outperforms other existing methods. Additionally, the predicted top-ranked GO terms are checked through multilayer network propagation and their impact on the 3D structure of the genome is observed.
FRONTIERS IN GENETICS
(2022)
Article
Mechanics
Sahaj Jain, Y. Sudhakar
Summary: Due to the challenges in accurately predicting interface velocities and computing drag components on rough surfaces, an effective model, called the Transpiration-Resistance model, has been developed. This model introduces shear and pressure correction factors as constitutive parameters to accurately predict interface velocities and partition the total drag into viscous and pressure components.
Article
Multidisciplinary Sciences
Adnan Khan, Jamal Uddin, Farman Ali, Ashfaq Ahmad, Omar Alghushairy, Ameen Banjar, Ali Daud
Summary: In this research, a novel computational method called AFP-LXGB has been proposed for more precise prediction of antifreeze proteins (AFPs). By exploring information through various feature sets and selecting the best feature set, the method has shown significant improvements in prediction accuracy. This approach has important applications in fields such as medicine, agriculture, industry, and biotechnology.
SCIENTIFIC REPORTS
(2022)
Article
Biochemical Research Methods
Gui-Jun Zhang, Teng-Yu Xie, Xiao-Gen Zhou, Liu-Jing Wang, Jun Hu
Summary: This study proposed a population-based algorithm guided by information entropy (PAIE) for protein structure prediction, which improved performance by exploring and exploiting in two stages.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2021)
Article
Biochemistry & Molecular Biology
Seung Hwan Hong, InSuk Joung, Jose C. Flores-Canales, Balachandran Manavalan, Qianyi Cheng, Seungryong Heo, Jong Yun Kim, Sun Young Lee, Mikyung Nam, Keehyoung Joo, In-Ho Lee, Sung Jong Lee, Jooyoung Lee
PROTEINS-STRUCTURE FUNCTION AND BIOINFORMATICS
(2018)
Article
Biochemistry & Molecular Biology
Keehyoung Joo, Seungryong Heo, InSuk Joung, Seung Hwan Hong, Sung Jong Lee, Jooyoung Lee
PROTEINS-STRUCTURE FUNCTION AND BIOINFORMATICS
(2018)
Article
Physics, Multidisciplinary
Seung Hwan Hong, Jin Mo Bok, Wentao Zhang, Junfeng He, X. J. Zhou, C. M. Varma, Han-Yong Choi
PHYSICAL REVIEW LETTERS
(2014)
Article
Chemistry, Medicinal
Seung Hwan Hong, Seongok Ryu, Jaechang Lim, Woo Youn Kim
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2020)