Article
Biochemical Research Methods
Kaitao Wu, Lexiang Wang, Bo Liu, Yang Liu, Yadong Wang, Junyi Li
Summary: Using computational methods to predict protein function effectively remains a challenge. Methods based on single species or single data source have limitations: different species require different models, and single perspective methods such as using Protein-Protein Interaction network only consider the protein environment and ignore intrinsic characteristics of protein sequences. To solve these problems, we propose PSPGO, a cross-species heterogeneous network propagation method based on graph attention mechanism, which can predict gene ontology terms by propagating feature and label information on sequence similarity and PPI networks. Our model is evaluated on a large multi-species dataset and compared with state-of-the-art methods, showing good performance.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2023)
Article
Biochemical Research Methods
Weixia Xu, Yangyun Gao, Yang Wang, Jihong Guan
Summary: This study presents a deep learning method (OR-RCNN) for predicting protein-protein interactions from the perspective of confidence scores. By constructing an encoder and predictor, it achieves high accuracy in PPI prediction.
BMC BIOINFORMATICS
(2021)
Article
Computer Science, Information Systems
Haojun Huang, Li Li, Geyong Min, Wang Miao, Yingying Zhu, Yangming Zhao
Summary: This article proposes a tensor-based network distance prediction (TNDP) approach to represent network distance with confidence intervals. Experimental results demonstrate that this approach outperforms other methods in terms of accuracy for network distance prediction.
IEEE TRANSACTIONS ON SERVICES COMPUTING
(2022)
Article
Biochemical Research Methods
Qianmu Yuan, Junjie Xie, Jiancong Xie, Huiying Zhao, Yuedong Yang
Summary: Protein function prediction is crucial in bioinformatics and has implications for disease mechanism elucidation and drug target discovery. However, accurately predicting protein functions solely from sequences remains challenging. This study introduces SPROF-GO, a sequence-based alignment-free predictor that utilizes a pretrained language model to extract informative sequence embeddings and implements self-attention pooling to focus on important residues. SPROF-GO outperforms state-of-the-art approaches in precision-recall curves and demonstrates generalization capabilities.
BRIEFINGS IN BIOINFORMATICS
(2023)
Review
Plant Sciences
Peipei Wang, Ally M. Schumacher, Shin-Han Shiu
Summary: Predicting plant metabolic pathways is crucial for metabolic engineering and the production of plant metabolite-derived medicine. Recent progress has been made in using multi-omics data and computational approaches to predict the pathways, complementing traditional genetic and biochemical approaches.
CURRENT OPINION IN PLANT BIOLOGY
(2022)
Article
Chemistry, Multidisciplinary
Beihong Ji, Xibing He, Yuzhao Zhang, Jingchen Zhai, Viet Hoang Man, Shuhan Liu, Junmei Wang
Summary: The novel algorithm improves screening performance by recalibrating docking scores based on structure similarity, achieving significant enhancements especially with CSE = S-4 and FP2 fingerprints. The method increases predictive index values for drug receptors without additional computational cost, demonstrating superior virtual screening performance.
JOURNAL OF CHEMINFORMATICS
(2021)
Article
Genetics & Heredity
Jiaogen Zhou, Wei Xiong, Yang Wang, Jihong Guan
Summary: The study found that edge enrichment in PPI networks works better than network reconstruction and original networks, with sequence similarity outperforming both local and global similarity. These results are helpful for biologists to choose suitable pre-processing schemes and achieve more accurate protein function prediction.
FRONTIERS IN GENETICS
(2021)
Review
Biotechnology & Applied Microbiology
Ilaria Porello, Francesco Cellesi
Summary: Achieving effective delivery of therapeutic proteins to intracellular targets is crucial for advancing human health. Current methods, such as chemical modification and nanocarrier-based approaches, have limitations in efficiency and safety. Developing more versatile delivery tools that trigger endocytosis or directly deliver proteins to the cytosol is essential for successful therapeutic effects. This article provides an overview of current methods for intracellular protein delivery, highlighting challenges and opportunities for future research.
FRONTIERS IN BIOENGINEERING AND BIOTECHNOLOGY
(2023)
Article
Biology
Richa Dhanuka, Jyoti Prakash Singh
Summary: With the advancement of high throughput sequencing technologies, protein sequence generation has become fast and cost-effective, leading to a significant increase in the number of known proteins. Identifying functions of newly discovered proteins remains a challenge. This study introduces a machine learning-based approach that leverages inter-relationships between functions to improve predictability and uses statistical methods to reduce redundant functions.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2021)
Article
Biochemical Research Methods
Jia-Ming Chang, Evan W. Floden, Javier Herrero, Olivier Gascuel, Paolo Di Tommaso, Cedric Notredame
Summary: This study demonstrates that taking into account the uncertainty introduced by multiple sequence alignment in reconstructing phylogenies can significantly increase the correlation between Glade correctness and its corresponding bootstrap value. The new Weighted Partial Super Bootstrap method can improve the predictive power of bootstrap values and enhance the discrimination capacity between correct and incorrect trees.
Article
Biotechnology & Applied Microbiology
Wenqi Chen, Shuang Wang, Tao Song, Xue Li, Peifu Han, Changnan Gao
Summary: In this study, a novel sequence-based computational approach called DCSE was proposed to predict potential protein-protein interactions (PPIs). The method utilized NLP-based encoding and feature extraction using multi-layer neural networks. Comparison with other models demonstrated the superior performance of the proposed method across all evaluation criteria.
Article
Computer Science, Artificial Intelligence
Mei Li, Ye Cao, Xiaoguang Liu, Hua Ji
Summary: This article proposes a structure-aware graph attention diffusion network (SGADN) for efficient spatial structure learning of protein-ligand complexes by incorporating both distance and angle information. The SGADN utilizes line graph attention diffusion layers (LGADLs) to explore long-range bond node interactions and enhance the hierarchical structure learning, and also introduces an attentive pooling layer (APL) to refine the hierarchical structures in complexes.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Biochemical Research Methods
Yunda Si, Chengfei Yan
Summary: A new deep residual learning-based protein contact prediction model was developed in this study, featuring a hybrid residual block combining 1D and 2D convolutions and a new loss function. The model, referred to as DRN-1D2D, was evaluated on various datasets and outperformed other state-of-the-art protein contact prediction models.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Chemistry, Analytical
Bao Li, Jing Xiong, Feng Wan, Changhua Wang, Dongjing Wang
Summary: Traffic flow prediction is a crucial task in Intelligent Transportation Systems (ITSs) and is influenced by various complex factors. Many cities utilize efficient traffic prediction methods for congestion control. However, existing methods often overlook the complexity of multivariate auxiliary information in highways and struggle to explain prediction results based solely on historical traffic flow. To address these challenges, we propose Multi-variate and Multi-horizon prediction based on Long Short-Term Memory (MMLSTM), which effectively incorporates auxiliary information and improves prediction performance through a multi-horizon time spans strategy. MMLSTM combines a bidirectional LSTM model, attention mechanism, and multi-layer perceptron for traffic prediction and provides interpretability by utilizing multivariate information.
Article
Computer Science, Artificial Intelligence
Pengfei Ding, Xianzhen Huang, Chengying Zhao, Huizhen Liu, Xuewei Zhang
Summary: In modern manufacturing, micro-milling technology is crucial for producing high-precision and complex micro-size parts. Understanding the changing rule of time-varying cutting is significant for comprehending the micro-milling mechanism and improving machining efficiency. Additionally, identifying and updating tool wear in advance can enhance the accuracy and sustainability of micromachining. This study proposes a tool wear prediction framework for micro-milling using a temporal convolution network, bi-directional long short-term memory, and a multi-objective arithmetic optimization algorithm. A new integrated model for real-time micro-milling cutting force monitoring is then developed, considering factors such as tool deformation, tool runout, time-varying cutting coefficient, chip separation state, and tool wear estimation results. The accuracy of the proposed tool wear prediction and cutting force model is verified through micro-milling experiments with Al6061 workpiece material. The developed model provides theoretical guidance for statics and dynamics analysis in micro-milling.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Immunology
Robert P. Milius, Michael Heuer, Daniel Valiga, Kathryn J. Doroschak, Caleb J. Kennedy, Yung-Tsi Bolon, Joel Schneider, Jane Pollack, Hwa Ran Kim, Nezih Cereb, Jill A. Hollenbach, Steven J. Mack, Martin Maiers
Article
Otorhinolaryngology
Adrian J. Priesol, Mengfei Cao, Carla E. Brodley, Richard F. Lewis
JAMA OTOLARYNGOLOGY-HEAD & NECK SURGERY
(2015)
Article
Multidisciplinary Sciences
Mengfei Cao, Hao Zhang, Jisoo Park, Noah M. Daniels, Mark E. Crovella, Lenore J. Cowen, Benjamin Hescott
Article
Biotechnology & Applied Microbiology
Nicolas Cardozo, Karen Zhang, Kathryn Doroschak, Aerilynn Nguyen, Zoheb Siddiqui, Nicholas Bogard, Karin Strauss, Luis Ceze, Jeff Nivala
Summary: Detection of specific proteins using nanopores is currently challenging, but the development of NTERs as protein tags has enabled simultaneous detection of up to nine reporter proteins in bacterial or human cells. This multiplexed detection is achieved through the use of protein barcodes measured with nanopores.
NATURE BIOTECHNOLOGY
(2022)
Article
Multidisciplinary Sciences
Kathryn Doroschak, Karen Zhang, Melissa Queen, Aishwarya Mandyam, Karin Strauss, Luis Ceze, Jeff Nivala
NATURE COMMUNICATIONS
(2020)
Proceedings Paper
Biotechnology & Applied Microbiology
Mengfei Cao, Lenore J. Cowen
PACIFIC SYMPOSIUM ON BIOCOMPUTING 2017
(2017)