Related references
Note: Only part of the references are listed.
Article
Biochemistry & Molecular Biology
Hansol Lee et al.
Summary: Antimicrobial resistance is a growing health concern, and antimicrobial peptides (AMPs) are considered promising alternatives to traditional antimicrobial drugs because of their nonspecific mechanisms that make it difficult for microbes to develop resistance. In this study, we developed a deep learning model, AMP-BERT, which utilizes the BERT architecture to extract information from input peptides and accurately classify them as AMP or non-AMP. AMP-BERT outperformed other machine/deep learning methods and provided interpretable feature analysis to identify specific residues contributing to peptide structure and antimicrobial function. AMP-BERT is expected to aid in the identification and validation of candidate AMPs for drug development.
Article
Biochemical Research Methods
Qiushi Cao et al.
Summary: With the rise of multidrug-resistant bacteria, antimicrobial peptides (AMPs) have emerged as potential alternatives to traditional antibiotics for treating bacterial infections. However, traditional methods of discovering and designing AMPs are time-consuming and costly. This study utilized deep learning techniques, including sequence generative adversarial nets, bidirectional encoder representations from transformers, and multilayer perceptron, to design and identify AMPs. Six candidate AMPs were then screened and one of them, A-222, showed inhibition against both gram-positive and gram-negative bacteria. Structural analysis and subsequent structure-activity relationship studies led to the design of peptide analogs with increased activity against specific bacteria. Overall, deep learning holds great promise in accelerating the discovery of novel AMPs and could have significant implications in developing new antimicrobial treatments.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Biochemistry & Molecular Biology
Francesca Grisoni
Summary: Generative deep learning is revolutionizing de novo drug design by enabling the generation of molecules with specific properties. Chemical language models, which use deep learning to generate new molecules as strings, have been remarkably successful in this endeavor. With advances in natural language processing and interdisciplinary collaborations, chemical language models are expected to play a key role in the future of drug discovery.
CURRENT OPINION IN STRUCTURAL BIOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Yiping Liu et al.
Summary: This paper presents an evolutionary multi-objective approach for designing antimicrobial peptides (AMPs) that optimizes both antimicrobial activity and diversity. The approach utilizes a deep learning model to predict antimicrobial activity and a niche sharing method to estimate peptide density. An evolutionary multi-objective algorithm is used to simultaneously optimize antimicrobial activity and diversity, and a local search strategy is applied to improve the quality of the identified AMPs. Experimental results demonstrate the superiority of the proposed approach in searching for diverse AMPs with high antimicrobial activities.
IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE
(2023)
Article
Chemistry, Medicinal
Hamid Teimouri et al.
Summary: There are different types of short peptide molecules called antimicrobial peptides (AMPs) that are produced by living organisms as a response to infections. Machine-learning methods have been used to predict the activities of AMPs against bacteria, but they typically do not consider specific features of the bacteria. To address this, a new computational approach was developed to train supervised machine-learning models using data associated with peptides targeting E. coli bacteria. The study provides recommendations for designing more efficient antibacterial drug therapies.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Chemistry, Medicinal
Mehrad Ansari et al.
Summary: We propose three deep learning sequence-based prediction models for peptide properties, achieving comparable results to state-of-the-art models. Our sequence-based solubility predictor, MahLooL, outperforms current methods for short peptides. These models are implemented as a static website, providing accessible and reproducible predictions without the need for third-party servers or cloud computing.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Chemistry, Medicinal
Xuetong Yang et al.
Summary: Anticancer peptides (ACPs) have gained attention in cancer therapy for their low consumption, few side effects, and easy accessibility. However, identifying anticancer peptides through experimental approaches is challenging and time-consuming. Traditional machine-learning-based methods for ACP prediction have low performance. In this study, a deep learning framework called CACPP is proposed, based on a convolutional neural network and contrastive learning, to accurately predict anticancer peptides. The comparative results on benchmark datasets demonstrate that CACPP outperforms state-of-the-art methods in predicting anticancer peptides. Visualization of feature dimension reduction and exploration of the relationship between ACP sequences and anticancer functions are also discussed.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Chemistry, Medicinal
Andrejs Tucs et al.
Summary: Increasing the variety of antimicrobial peptides is crucial to tackle multi-drug-resistant bacterial pathogens. A multi-objective peptide design pipeline based on a discrete latent space and D-Wave quantum annealer was developed to solve the local minima problem. Four peptides designed by the pipeline were validated in wet-lab experiments, with three of them showing high antimicrobial activity and two being non-hemolytic.
ACS MEDICINAL CHEMISTRY LETTERS
(2023)
Article
Multidisciplinary Sciences
Paulina Szymczak et al.
Summary: Antimicrobial peptides have the potential to address the global health threat of antimicrobial resistance. The authors introduce HydrAMP, an extended conditional variational autoencoder, which successfully generates potent antimicrobial peptides against bacteria, including multidrug-resistant strains.
NATURE COMMUNICATIONS
(2023)
Article
Biochemical Research Methods
Sen Yang et al.
Summary: This paper proposes a cascaded computational model called AMPFinder to identify antimicrobial peptides (AMPs) and their functional types. AMPFinder outperforms other methods in AMP identification and function prediction. It achieves higher F1-score (1.45%-6.13%), MCC (2.92%-12.86%), AUC (5.13%-8.56%), AP (9.20%-21.07%) on an independent test dataset, and lower R2 bias (18.82%-19.46%) on a public dataset through 10-fold cross-validation.
ANALYTICAL BIOCHEMISTRY
(2023)
Review
Mathematical & Computational Biology
Fabiano C. Fernandes et al.
Summary: This review explores the latest developments in the use of geometric deep learning (GDL) techniques for designing and predicting antimicrobial peptides (AMPs), and discusses the current research gaps and future directions in this field.
FRONTIERS IN BIOINFORMATICS
(2023)
Article
Biochemistry & Molecular Biology
Alexey Strokach et al.
Summary: Deep learning approaches have made significant contributions in protein design and generative models can create millions of novel proteins similar to native ones. These models can learn protein representations that are more informative than hand-engineered features, and the design process can be guided by discriminative models.
CURRENT OPINION IN STRUCTURAL BIOLOGY
(2022)
Article
Medicine, General & Internal
Christopher J. L. Murray et al.
Summary: Antimicrobial resistance (AMR) poses a significant threat to global human health, and this study provides the most comprehensive estimates of AMR burden to date. By estimating deaths and disability-adjusted life-years (DALYs) associated with bacterial AMR in 204 countries and territories in 2019, the study highlights the impact of resistance and the leading pathogen-drug combinations contributing to it.
Review
Chemistry, Medicinal
Yiyun Zhu et al.
Summary: Antimicrobial peptides (AMPs), also known as host defense peptides (HDPs), are important immune defense molecules in multicellular organisms. AMPs have great potential as new generation antimicrobial agents and can be used in combination with conventional antibiotics to kill drug-resistant pathogens and improve therapeutic effects.
MEDICINAL RESEARCH REVIEWS
(2022)
Review
Biochemical Research Methods
Charles H. Chen et al.
Summary: With the advancement in understanding the relationship between peptide structure and activity, computational tools for peptide design and optimization have become essential in saving time and cost. Improvements in peptide synthesis and experimental platforms have also increased the efficiency of peptide-drug screening.
CURRENT OPINION IN BIOTECHNOLOGY
(2022)
Editorial Material
Biotechnology & Applied Microbiology
Yue Ma et al.
NATURE BIOTECHNOLOGY
(2022)
Review
Chemistry, Multidisciplinary
Camille Bilodeau et al.
Summary: This article reviews the recent advances in generative molecular design and discusses the considerations for integrating these models into real molecular discovery campaigns. It covers the model design choices, problem statements for molecular discovery applications, benchmarks used to evaluate models, and important factors for integrating generative models into experimental workflows.
WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL MOLECULAR SCIENCE
(2022)
Article
Microbiology
Aaron B. Benjamin et al.
Summary: Staphylococcus aureus resistance is a major concern in healthcare, and antimicrobial peptoids show promise as an alternative to conventional antibiotics for treating S. aureus wound infections, with efficient killing of bacteria and prevention of biofilm formation and detachment.
MICROBIOLOGY SPECTRUM
(2022)
Article
Biochemical Research Methods
Boris Vishnepolsky et al.
Summary: The evolution of drug-resistant microbial species is a global health concern. Antimicrobial peptides (AMPs) are potential candidates for addressing antibiotic resistance, and computational methods have been developed to predict their activity. However, most methods are not strain-specific, and only a few models have been developed due to limited data. To improve predictions, a novel approach based on AMP sequences and target microbial genome characteristics was developed. The new models, using machine learning algorithms, showed improved performance compared to sequence-based models. The MSS AMP predictor is freely accessible as part of the DBAASP database.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Milad Salem et al.
Summary: This study utilizes deep learning models for the classification of hemolysis activity in antimicrobial peptides, leveraging transfer learning to overcome the challenge of small data. AMPDeep demonstrates superior performance on three datasets by fine-tuning large transformer models and leveraging patterns learned from other protein and peptide databases.
BMC BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Katarzyna Sidorczuk et al.
Summary: Antimicrobial peptides (AMPs) are short polypeptides that effectively target microorganisms, viruses, and cancer cells. Machine learning plays an important role in AMP discovery and prediction due to their low resistance selection. The impact of negative data sampling on model performance was investigated, with the finding that similar training and benchmark data sets positively affect model performance. However, existing benchmark analyses for AMP prediction models are biased and the most accurate model remains unknown.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Samuel C. Hoffman et al.
Summary: Machine learning-based method QMO, utilizing latent embeddings from a molecule autoencoder, outperforms existing methods in optimizing small organic molecules for drug-likeness and solubility. It also shows substantial property improvement in tasks such as optimizing potential SARS-CoV-2 main protease inhibitors and lowering toxicity of antimicrobial peptides. Zeroth-order optimization is used to address discreteness of molecule sequences and incorporate external guidance from property evaluations and design constraints.
NATURE MACHINE INTELLIGENCE
(2022)
Review
Biology
Marcelo C. R. Melo et al.
Summary: Antibiotics insert themselves into the ancient struggle of host-pathogen evolution, driving urgent interest in computational methods for candidate discovery. Advances in artificial intelligence have been applied to antibiotic discovery, emphasizing antimicrobial activity prediction, drug-likeness traits, resistance, and de novo molecular design. Best practices such as open science and reproducibility are crucial in accelerating preclinical research in the face of antimicrobial resistance crisis.
COMMUNICATIONS BIOLOGY
(2021)
Article
Engineering, Biomedical
Payel Das et al.
Summary: The combination of deep learning and molecular dynamics simulations enables the rapid discovery of antimicrobial peptides with low toxicity and high potency against a variety of Gram-positive and Gram-negative pathogens.
NATURE BIOMEDICAL ENGINEERING
(2021)
Article
Microbiology
Scott N. Dean et al.
Summary: The study introduces the PepVAE framework for designing novel AMPs using VAE and antimicrobial activity prediction models. By sampling from different regions of the learned latent space, new AMP sequences can be generated with minimal input parameters for controlled production of AMPs with predicted antimicrobial activity. This modular design framework shows promise for development of novel AMPs with experimental validation.
FRONTIERS IN MICROBIOLOGY
(2021)
Article
Chemistry, Medicinal
Colin M. Van Oort et al.
Summary: Antibiotic resistance poses a critical public health problem, with AMPs showing promise as a treatment for resistant infections. However, existing AMPs face issues in development, production, and shelf-life. AMPGAN v2 is a bidirectional conditional generative adversarial network (BiCGAN) approach that can efficiently design AMPs with novel and tailored candidates.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2021)
Article
Biochemical Research Methods
Xiang Wu et al.
Summary: This article introduces a novel deep learning architecture named EPSOL for predicting protein solubility in an E.coli expression system, which achieves high accuracy and reliability in predicting the solubility of new recombinant proteins.
Article
Biophysics
William F. Porto et al.
BIOCHIMICA ET BIOPHYSICA ACTA-GENERAL SUBJECTS
(2021)
Review
Chemistry, Medicinal
Pietro G. A. Aronica et al.
Summary: The evolution of antibiotic-resistant bacteria has led to an increase in untreatable diseases, highlighting the urgent need for alternative treatment methods. Antimicrobial peptides (AMPs) have attracted significant interest, with their design and development aided by molecular models and computational approaches can aid experimental studies in this area.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2021)
Article
Multidisciplinary Sciences
John Jumper et al.
Summary: Proteins are essential for life, and accurate prediction of their structures is a crucial research problem. Current experimental methods are time-consuming, highlighting the need for accurate computational approaches to address the gap in structural coverage. Despite recent progress, existing methods fall short of atomic accuracy in protein structure prediction.
Article
Multidisciplinary Sciences
Maria Rosa Loffredo et al.
Summary: The activity of many antibiotics depends on the initial cell density used in bacterial growth inhibition assays, with a significant inoculum effect observed for higher cell densities. Antimicrobial peptides perturb cell membranes and the inoculum effect can impact their bacterial killing ability. Routine evaluation of the dependence of antimicrobial peptides and peptidomimetics on inoculum density should be considered.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Biochemistry & Molecular Biology
Malak Pirtskhalava et al.
Summary: DBAASP is a comprehensive database containing information on peptides with antimicrobial properties, continuously updated with tools for developing and optimizing new peptides; while there is an increased interest in antimicrobial peptides, their clinical application is still limited.
NUCLEIC ACIDS RESEARCH
(2021)
Article
Chemistry, Multidisciplinary
Scott N. Dean et al.
Review
Infectious Diseases
Maria Magana et al.
LANCET INFECTIOUS DISEASES
(2020)
