HNSPPI: a hybrid computational model combing network and sequence information for predicting protein–protein interaction DOI

Shijie Xie,

Xiaojun Xie, Xin Zhao

и другие.

Briefings in Bioinformatics, Год журнала: 2023, Номер 24(5)

Опубликована: Июнь 28, 2023

Most life activities in organisms are regulated through protein complexes, which mainly controlled via Protein-Protein Interactions (PPIs). Discovering new interactions between proteins and revealing their biological functions of great significance for understanding the molecular mechanisms processes identifying potential targets drug discovery. Current experimental methods only capture stable interactions, lead to limited coverage. In addition, expensive cost time consuming also obvious shortcomings. recent years, various computational have been successfully developed predicting PPIs based on homology, primary sequences or gene ontology information. Computational efficiency data complexity still main bottlenecks algorithm generalization. this study, we proposed a novel framework, HNSPPI, predict PPIs. As hybrid supervised learning model, HNSPPI comprehensively characterizes intrinsic relationship two by integrating amino acid sequence information connection properties PPI network. The results show that works very well six benchmark datasets. Moreover, comparison analysis proved our model significantly outperforms other five existing algorithms. Finally, used explore SARS-CoV-2-Human interaction system found several regulations. summary, is promising from known data.

Язык: Английский

From machine learning to deep learning: Advances of the recent data-driven paradigm shift in medicine and healthcare DOI Creative Commons
Chiranjib Chakraborty, Manojit Bhattacharya, Soumen Pal

и другие.

Current Research in Biotechnology, Год журнала: 2023, Номер 7, С. 100164 - 100164

Опубликована: Ноя. 22, 2023

The medicine and healthcare sector has been evolving advancing very fast. advancement initiated shaped by the applications of data-driven, robust, efficient machine learning (ML) to deep (DL) technologies. ML in medical is developing quickly, causing rapid progress, reshaping medicine, improving clinician patient experiences. technologies evolved into data-hungry DL approaches, which are more robust dealing with data. This article reviews some critical data-driven aspects intelligence field. In this direction, illustrated recent progress science using two categories: firstly, development data uses and, secondly, Chabot particularly on ChatGPT. Here, we discuss ML, DL, transition requirements from DL. To science, illustrate prospective studies image data, newly interpretation EMR or EHR, big personalized dataset shifts artificial (AI). Simultaneously, recently developed DL-enabled ChatGPT technology. Finally, summarize broad role significant challenges for implementing healthcare. overview paradigm shift will benefit researchers immensely.

Язык: Английский

Процитировано

73

Machine Learning Empowering Drug Discovery: Applications, Opportunities and Challenges DOI Creative Commons
Xin Qi,

Yuanchun Zhao,

Zhuang Qi

и другие.

Molecules, Год журнала: 2024, Номер 29(4), С. 903 - 903

Опубликована: Фев. 18, 2024

Drug discovery plays a critical role in advancing human health by developing new medications and treatments to combat diseases. How accelerate the pace reduce costs of drug has long been key concern for pharmaceutical industry. Fortunately, leveraging advanced algorithms, computational power biological big data, artificial intelligence (AI) technology, especially machine learning (ML), holds promise making hunt drugs more efficient. Recently, Transformer-based models that have achieved revolutionary breakthroughs natural language processing sparked era their applications discovery. Herein, we introduce latest ML discovery, highlight potential models, discuss future prospects challenges field.

Язык: Английский

Процитировано

26

Leveraging machine learning models for peptide–protein interaction prediction DOI Creative Commons
Yin Song, Xuenan Mi, Diwakar Shukla

и другие.

RSC Chemical Biology, Год журнала: 2024, Номер 5(5), С. 401 - 417

Опубликована: Янв. 1, 2024

A timeline showcasing the progress of machine learning and deep methods for peptide–protein interaction predictions.

Язык: Английский

Процитировано

21

The role of machine learning in discovering biomarkers and predicting treatment strategies for neurodegenerative diseases: A narrative review DOI Creative Commons
Abdullahi Tunde Aborode,

Ogunware Adedayo Emmanuel,

Isreal Ayobami Onifade

и другие.

NeuroMarkers., Год журнала: 2025, Номер 2(1), С. 100034 - 100034

Опубликована: Янв. 6, 2025

Язык: Английский

Процитировано

3

Accelerating Biocatalysis Discovery with Machine Learning: A Paradigm Shift in Enzyme Engineering, Discovery, and Design DOI Creative Commons

Braun Markus,

Gruber Christian C,

Krassnigg Andreas

и другие.

ACS Catalysis, Год журнала: 2023, Номер 13(21), С. 14454 - 14469

Опубликована: Окт. 26, 2023

Emerging computational tools promise to revolutionize protein engineering for biocatalytic applications and accelerate the development timelines previously needed optimize an enzyme its more efficient variant. For over a decade, benefits of predictive algorithms have helped scientists engineers navigate complexity functional sequence space. More recently, spurred by dramatic advances in underlying tools, faster, cheaper, accurate identification, characterization, has catapulted terms such as artificial intelligence machine learning must-have vocabulary field. This Perspective aims showcase current status pharmaceutical industry also discuss celebrate innovative approaches science highlighting their potential selected recent developments offering thoughts on future opportunities biocatalysis. It critically assesses technology's limitations, unanswered questions, unmet challenges.

Язык: Английский

Процитировано

39

Plant-based proteins: advanced extraction technologies, interactions, physicochemical and functional properties, food and related applications, and health benefits DOI Creative Commons
Ahmed K. Rashwan, Ahmed I. Osman, Asem Mahmoud Abdelshafy

и другие.

Critical Reviews in Food Science and Nutrition, Год журнала: 2023, Номер 65(4), С. 667 - 694

Опубликована: Ноя. 15, 2023

Even though plant proteins are more plentiful and affordable than animal in comparison, direct usage of plant-based (PBPs) is still limited because PBPs fed to animals as feed produce animal-based proteins. Thus, this work has comprehensively reviewed the effects various factors such pH, temperature, pressure, ionic strength on PBP properties, well describes protein interactions, extraction methods know optimal conditions for preparing PBP-based products with high functional properties health benefits. According cited studies current work, environmental factors, particularly pH significantly affected physicochemical PBPs, especially solubility was 76.0% 83.9% at = 2, while 5.0 reduced from 5.3% 9.6%, emulsifying ability lowest 5.8 highest 8.0, foaming capacity 7.0. Electrostatic interactions main way which can be used create protein/polysaccharide complexes food industrial purposes. The yield reached up 86-95% using sustainable efficient routes, including enzymatic, ultrasound-, microwave-, pulsed electric field-, high-pressure-assisted extraction. Nondairy alternative products, yogurt, 3D printing meat analogs, synthesis nanoparticles, bioplastics packaging films best available PBPs-based products. Moreover, those that contain pigments their showed good bioactivities, antioxidants, antidiabetic, antimicrobial.

Язык: Английский

Процитировано

31

Open-Source Machine Learning in Computational Chemistry DOI Creative Commons
Alexander Hagg, Karl N. Kirschner

Journal of Chemical Information and Modeling, Год журнала: 2023, Номер 63(15), С. 4505 - 4532

Опубликована: Июль 19, 2023

The field of computational chemistry has seen a significant increase in the integration machine learning concepts and algorithms. In this Perspective, we surveyed 179 open-source software projects, with corresponding peer-reviewed papers published within last 5 years, to better understand topics being investigated by approaches. For each project, provide short description, link code, accompanying license type, whether training data resulting models are made publicly available. Based on those deposited GitHub repositories, most popular employed Python libraries identified. We hope that survey will serve as resource learn about or specific architectures thereof identifying accessible codes topic basis. To end, also include for generating fundamental learning. our observations considering three pillars collaborative work, open data, source (code), models, some suggestions community.

Язык: Английский

Процитировано

28

Computational prediction of disordered binding regions DOI Creative Commons
Sushmita Basu, Daisuke Kihara, Lukasz Kurgan

и другие.

Computational and Structural Biotechnology Journal, Год журнала: 2023, Номер 21, С. 1487 - 1497

Опубликована: Янв. 1, 2023

One of the key features intrinsically disordered regions (IDRs) is their ability to interact with a broad range partner molecules. Multiple types interacting IDRs were identified including molecular recognition fragments (MoRFs), short linear sequence motifs (SLiMs), and protein-, nucleic acids- lipid-binding regions. Prediction binding in protein sequences gaining momentum recent years. We survey 38 predictors that target interactions diverse set partners, such as peptides, proteins, RNA, DNA lipids. offer historical perspective highlight events fueled efforts develop these methods. These tools rely on predictive architectures include scoring functions, regular expressions, traditional deep machine learning meta-models. Recent focus development neural network-based extending coverage IDRs. analyze availability methods show providing implementations webservers results much higher rates citations/use. also make several recommendations take advantage modern network architectures, bundle predictions multiple different IDRs, work algorithms model structures resulting complexes.

Язык: Английский

Процитировано

27

ProtInteract: A deep learning framework for predicting protein–protein interactions DOI Creative Commons
Farzan Soleymani, Eric Paquet, Herna L. Viktor

и другие.

Computational and Structural Biotechnology Journal, Год журнала: 2023, Номер 21, С. 1324 - 1348

Опубликована: Янв. 1, 2023

Proteins mainly perform their functions by interacting with other proteins. Protein–protein interactions underpin various biological activities such as metabolic cycles, signal transduction, and immune response. However, due to the sheer number of proteins, experimental methods for finding non-interacting protein pairs are time-consuming costly. We therefore developed ProtInteract framework predict protein–protein interaction. comprises two components: first, a novel autoencoder architecture that encodes each protein's primary structure lower-dimensional vector while preserving its underlying sequence attributes. This leads faster training second network, deep convolutional neural network (CNN) receives encoded proteins predicts interaction under three different scenarios. In scenario, CNN class given pair. Each indicates ranges confidence scores corresponding probability whether predicted occurs or not. The proposed features significantly low computational complexity relatively fast contributions this work twofold. First, assimilates into pseudo-time series. Therefore, we leverage nature time series physicochemical properties encode amino acid space. approach enables extracting highly informative attributes reducing complexity. Second, utilises information identify based on configuration. Our results suggest performs high accuracy efficiency in predicting protein-protein interactions.

Язык: Английский

Процитировано

25

SpatialPPI: Three-dimensional space protein-protein interaction prediction with AlphaFold Multimer DOI Creative Commons
Wenxing Hu, Masahito Ohue

Computational and Structural Biotechnology Journal, Год журнала: 2024, Номер 23, С. 1214 - 1225

Опубликована: Март 15, 2024

Rapid advancements in protein sequencing technology have resulted gaps between proteins with identified sequences and those mapped structures. Although sequence-based predictions offer insights, they can be incomplete due to the absence of structural details. Conversely, structure-based methods face challenges respect newly sequenced proteins. The AlphaFold Multimer has remarkable accuracy predicting structure complexes. However, it cannot distinguish whether input interact. Nonetheless, by analyzing information models predicted Multimer, we propose a highly accurate method for interactions. This study focuses on use deep neural networks, specifically analyze complex structures Multimer. By transforming atomic coordinates utilizing sophisticated image-processing techniques, vital 3D details were extracted from Recognizing significance evaluating residue distances interactions, this leveraged image recognition approaches integrating Densely Connected Convolutional Networks (DenseNet) Deep Residual Network (ResNet) within convolutional networks analysis. When benchmarked against leading protein-protein interaction prediction methods, such as SpeedPPI, D-script, DeepTrio, PEPPI, our proposed method, named SpatialPPI, exhibited notable efficacy, emphasizing promising role spatial processing advancing realm biology. SpatialPPI code is available at: https://github.com/ohuelab/SpatialPPI.

Язык: Английский

Процитировано

12