Review
Biochemistry & Molecular Biology
Yi Liu, Qian Yang, Fangzhou Zhao
Summary: Codon usage bias affects gene expression levels and protein structures, influencing translation speed, accuracy, and the protein folding process. It also plays a key role in determining the proteome landscape. Furthermore, codon usage impacts mRNA levels through various effects on transcription and translation processes.
ANNUAL REVIEW OF BIOCHEMISTRY, VOL 90, 2021
(2021)
Article
Biochemistry & Molecular Biology
Gabriel Wright, Anabel Rodriguez, Jun Li, Tijana Milenkovic, Scott J. Emrich, Patricia L. Clark
Summary: Research has shown that synonymous codon usage can affect various mechanisms related to protein production, especially in co-translational protein folding. Conservation of synonymous codon usage patterns across evolution highlights the potential benefits of matching codon usage patterns from the original organism in heterologous gene expression.
Article
Evolutionary Biology
Rosa M. Pinto, Albert Bosch
Summary: Codon bias, a common phenomenon in all organisms, is influenced by mutation, drift, and selection. While selection for translation efficiency and accuracy is well recognized, fewer studies have explored the impact of translation rate control on codon usage. Experimental molecular evolution using RNA virus populations is a powerful tool in understanding the mechanisms behind codon bias. Furthermore, experimental studies are encouraged to define the role of selection in codon evolution, as most studies on virus codon usage rely on computational analyses.
GENOME BIOLOGY AND EVOLUTION
(2021)
Article
Multidisciplinary Sciences
Shakibur Rahman, Sergei L. Kosakovsky Pond, Andrew Webb, Jody Hey
Summary: Research shows that synonymous codon substitutions are not always selectively neutral, and conventional dN/dS analyses may overestimate the frequency of positive diversifying selection and underestimate the strength of purifying selection. By analyzing two classes of synonymous substitutions, it was found that this discrepancy can be adequately explained by very weak selection with a mean product of population size and selection coefficient, Ns = 0.8.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Biochemistry & Molecular Biology
Maria Kompatscher, Karolina Bartosik, Kevin Erharter, Raphael Plangger, Fabian Sebastian Juen, Christoph Kreutz, Ronald Micura, Eric Westhof, Matthias D. Erlacher
Summary: tRNA superwobbling is an intriguing decoding concept where a single tRNA isoacceptor is used to decode multiple synonymous codons. By studying tRNA(Gly) in Escherichia coli and Mycoplasma mycoides, we found that the tRNA sequence and modifications play important roles in decoding.
NUCLEIC ACIDS RESEARCH
(2023)
Article
Multidisciplinary Sciences
Amila Zuko, Moushami Mallik, Robin Thompson, Emily L. Spaulding, Anne R. Wienand, Marije Been, Abigail L. D. Tadenev, Nick van Bakel, Celine Sijlmans, Leonardo A. Santos, Julia Bussmann, Marica Catinozzi, Sarada Das, Divita Kulshrestha, Robert W. Burgess, Zoya Ignatova, Erik Storkebaum
Summary: Heterozygous mutations in six transfer RNA synthetase genes cause Charcot-Marie-Tooth peripheral neuropathy. Mutant tRNA synthetases in CMT inhibit protein synthesis by binding tRNA(Gly) but failing to release it, leading to tRNAGly sequestration and ribosome stalling. Overexpression of tRNA(Gly) rescues protein synthesis, peripheral neuropathy, and ISR activation, suggesting therapeutic potential in CMT2D.
Article
Biochemistry & Molecular Biology
Michal Perach, Zohar Zafrir, Tamir Tuller, Oded Lewinson
Summary: Due to genetic code redundancy, most amino acids are encoded by multiple synonymous codons, which are used unevenly in different organisms. Slow codons play important roles in regulating protein folding and function.
Article
Oncology
Qun Li, Jian Li, Chun-peng Yu, Shuai Chang, Ling-ling Xie, Song Wang
Summary: Synonymous mutations, which do not change protein sequences, can alter codon optimality and translational velocity, playing a significant role in liver cancer development. These mutations should no longer be ignored in genome-wide studies.
Article
Plant Sciences
Maxime Fages-Lartaud, Kristoffer Hundvin, Martin Frank Hohmann-Marriott
Summary: This article provides a comprehensive perspective on the mechanisms driving codon selection in Chlamydomonas reinhardtii chloroplasts and the functional consequences for protein expression. It is found that highly expressed genes prefer translationally optimal codons, while genes with lower functional importance are affected by directional mutational bias. Codon optimality can be deduced from codon-anticodon pairing affinity and tRNA concentrations for certain amino acids. The impact of codon usage on protein yield, mRNA secondary structure, translation initiation and termination, amino acid composition, and cotranslational protein folding is also discussed.
Article
Virology
Zhen He, Shiwen Ding, Jiyuan Guo, Lang Qin, Xiaowei Xu
Summary: This study analyzed the phylogeny and codon usage pattern of narcissus viruses using the coat protein. The results showed that codon usage bias in these viruses is mainly influenced by natural selection, indicating the importance of evolutionary-based design for controlling these viruses.
Article
Virology
Elisson N. Lopes, Vagner Fonseca, Diego Frias, Stephane Tosta, Alvaro Salgado, Ricardo Assuncao Vialle, Toscano S. Paulo Eduardo, Fernanda K. Barreto, Vasco Ariston de Azevedo, Michele Guarino, Silvia Angeletti, Massimo Ciccozzi, Luiz C. Junior Alcantara, Marta Giovanetti
Summary: The COVID-19 pandemic caused by SARS-CoV-2 has become a major global health issue, with vaccine rollouts bringing hope of herd immunity. However, concerns arise with the emergence of virus variants that may possess enhanced virulence. Codon usage analysis helps elucidate viral evolution and potential variants, highlighting the importance of genomic monitoring to prevent and mitigate the pandemic.
JOURNAL OF MEDICAL VIROLOGY
(2021)
Article
Materials Science, Biomaterials
Yang Wu, Mengtong Tang, Zhaoguan Wang, Youhui Yang, Zhong Li, Shurui Liang, Peng Yin, Hao Qi
Summary: This study demonstrates that the E. coli Pth and ArfB proteins can efficiently terminate translation without codon preference in the absence of class-I release factors. By degrading the target protein, both essential and alternative termination machinery types are removed, disabling codon-dependent termination in cell extract. The researchers also screened 153 engineered tRNAs to construct a codon-dependent termination defect in vitro protein synthesis system, called iPSSC, which efficiently decodes all stop codons. The full sense genetic code achieved significant improvement in incorporating unnatural amino acids and synthesizing proteins with consecutive NNN codons, suggesting great potential for building artificial protein synthesis beyond the cell.
Article
Biotechnology & Applied Microbiology
Jayanta Kumar Das, Swarup Roy
Summary: This study provides a comprehensive quantitative analysis of the protein-coding sequences of seven human coronaviruses, revealing the nucleotide sequence variability and codon usage patterns. Different categories of codons exhibit diverse adaptability, with notable variability in GC content at the third position among the seven coronaviruses. Phylogenetic analysis depicted the evolutionary relationships among these viruses and their positions in the tree.
Article
Parasitology
Dawei Wang, Baoling Yang
Summary: The study analyzed the codon usage bias of 32 thioredoxin coding sequences from 11 apicomplexan protozoa. The results indicated a certain degree of codon preference among these protozoa Trxs. Additionally, differences in codon base composition and relative synonymous codon usage were observed among different species of apicomplexan protozoa.
PARASITES & VECTORS
(2023)
Article
Genetics & Heredity
Wenjing Xu, Yingchun Li, Yajing Li, Chun Liu, Yanxia Wang, Guangmin Xia, Mengcheng Wang
Summary: Asymmetric somatic hybridization affects synonymous codon usage bias (SCUB) and promotes a bias towards A- and T-ending synonymous codons, with a stronger effect in indel-flanking sequences. This shift is attributed to whole genomic shock, and DNA methylation may play a role in SCUB shift during asymmetric somatic hybridization. Exogenous chromatin fragments introduced through this strategy do not have a local chromosomal effect on SCUB frequencies.
FRONTIERS IN GENETICS
(2021)
Article
Biochemical Research Methods
Dominik Schwarz, Guy Georges, Sebastian Kelm, Jiye Shi, Anna Vangone, Charlotte M. Deane
Summary: Co-evolution analysis and machine-learning techniques have improved the accuracy of predicting residue-residue contacts and distances in protein structures. The shape of predicted distance distributions contains information on the rigidity or flexibility of a protein and its constituent residues.
Article
Biochemical Research Methods
Constantin Schneider, Andrew Buchanan, Bruck Taddese, Charlotte M. Deane
Summary: DLAB can be used to improve antibody-antigen docking and structure-based virtual screening, enhancing pose ranking for antibody docking experiments and selection of accurate and correctly ranked antibody-antigen pairings.
Article
Chemistry, Medicinal
Fergus Boyles, Charlotte M. Deane, Garrett M. Morris
Summary: Machine learning scoring functions for protein-ligand binding affinity perform better on crystal structures than on docked poses, but a hybrid scoring function combining structure-based and ligand-based features shows comparable performance on docked poses to purely structure-based scoring functions trained on crystal poses. However, the hybrid scoring function may not always generalize well to protein targets not represented in the training set, indicating the need for improved scoring functions and additional validation benchmarks.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Biology
Oliver M. Crook, Chun-wa Chung, Charlotte M. Deane
Summary: Hydrogen deuterium exchange mass spectrometry (HDX-MS) is a technique to explore differential protein structure. We developed a new statistical testing procedure that incorporates a hydrogen deuterium exchange model and an empirical Bayes method to improve statistical power and interpretation. The flexibility of our method allows it to be applied in more experimental scenarios, reducing the burden on experimentalists to conduct excessive replicates.
COMMUNICATIONS BIOLOGY
(2022)
Article
Economics
Sita J. Saunders, Jody L. Grisamore, Tess Wong, Rafael Torrejon Torres, Rhodri Saunders, Brett Einerson
Summary: Utilizing the synthetic hygroscopic cervical dilator (SHCD) to introduce a low-acuity care room (ripening room) for cervical ripening as a part of labor induction can reduce the burden of care provision without compromising safety.
JOURNAL OF MEDICAL ECONOMICS
(2022)
Article
Chemistry, Medicinal
Stephanie Wills, Ruben Sanchez-Garcia, Tim Dudgeon, Stephen D. Roughley, Andy Merritt, Roderick E. Hubbard, James Davidson, Frank von Delft, Charlotte M. Deane
Summary: Fragment merging is a promising approach to advancing fragments directly to on-scale potency. Searching commercial catalogues allows for quick and cost-effective identification of these merges, circumventing the challenge of synthetic accessibility.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Biology
Brennan Abanades, Wing Ki Wong, Fergus Boyles, Guy Georges, Alexander Bujotzek, Charlotte M. Deane
Summary: Immune receptor proteins are important in the immune system and have potential as biotherapeutics. A deep learning model called ImmuneBuilder is presented, which accurately predicts the structure of antibodies, nanobodies, and T-cell receptors. ImmuneBuilder outperforms AlphaFold2 in terms of accuracy and speed.
COMMUNICATIONS BIOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Maranga Mokaya, Fergus Imrie, Willem P. van Hoorn, Aleksandra Kalisz, Anthony R. Bradley, Charlotte M. Deane
Summary: Deep reinforcement learning methods are potentially powerful tools for de novo design in the field of drug discovery. However, when there are few or no examples of desired molecules in the training data, recurrent neural network-based methods may have limitations in generating diverse and controllable molecular sets. To address these issues, a new curriculum-learning-inspired recurrent iterative optimization procedure is proposed, which allows the optimization of generated molecules for both known and unknown molecular profiles and provides control over exploration and exploitation. This method has been shown to generate specific and diverse sets of molecules with significantly more scaffolds compared to standard methods, although the choice of simplified molecular-input line-entry system (SMILES) representation has been found to affect the success of molecular optimization.
NATURE MACHINE INTELLIGENCE
(2023)
Article
Biochemical Research Methods
Oliver M. Crook, Nathan Gittens, Chun-wa Chung, Charlotte M. Deane
Summary: Proteins undergo structural perturbations when binding to others or subjected to environmental changes. HDX-MS can explore conformational changes in proteins by examining differences in deuterium incorporation rate. We propose a flexible Bayesian framework that improves stability, allows uncertainty quantification, and calculates statistical quantities that are inaccessible to other methods. Our findings demonstrate that a Bayesian approach can identify important binding epitopes from HDX data with consistent results.
JOURNAL OF PROTEOME RESEARCH
(2023)
Editorial Material
Biotechnology & Applied Microbiology
Carlos Outeiral, Charlotte M. Deane
Summary: This study presents a general-purpose protein language model that can rapidly improve antibody properties.
NATURE BIOTECHNOLOGY
(2023)
Article
Health Care Sciences & Services
Markus Meissner, Juliane Hafermann, Ubong Silas, Rhodri Saunders
Summary: Operation rooms have a significant environmental impact. Single-use staplers contribute to waste generation, while multi-use staplers can greatly reduce the environmental impact of surgery. Using multi-use staplers with pre-attached buttressing instead of single-use staplers with separate buttressing greatly reduces waste, resource consumption, and greenhouse gas emissions.
RISK MANAGEMENT AND HEALTHCARE POLICY
(2023)
Article
Multidisciplinary Sciences
Tobias H. Olsen, Brennan Abanades, Iain H. Moal, Charlotte M. Deane
Summary: KA-Search is a tool for rapid searching of antibody variable domains by amino acid sequence identity. It can predict antibody properties and find the most similar sequences from billions of antibody sequences within minutes. Examples of insights obtained using KA-Search are provided.
SCIENTIFIC REPORTS
(2023)
Article
Chemistry, Multidisciplinary
Ruben Sanchez-Garcia, David Havasi, Gergely Takacs, Matthew C. Robinson, Alpha Lee, Frank von Delft, Charlotte M. Deane
Summary: Compound availability is crucial in the drug discovery process. Synthetic accessibility scores have been used as proxies for compound availability, but they are not effective in representing compound prices. In this paper, we propose a deep learning model called CoPriNet, which can accurately predict compound prices and has fast execution times, making it suitable for high-throughput experiments.
Article
Biochemistry & Molecular Biology
Brennan Abanades, Tobias H. Olsen, Matthew I. J. Raybould, Broncio Aguilar-Sanjuan, Wing Ki Wong, Guy Georges, Alexander Bujotzek, Charlotte M. Deane
Summary: The study introduces PLAbDab, a self-updating database of antibodies that contains over 150,000 paired antibody sequences and 3D structural models. It allows users to search by sequence, structure, or keyword, and provides functions such as annotating query antibodies, analyzing structural models, and facilitating the compilation of customized datasets.
NUCLEIC ACIDS RESEARCH
(2023)
Article
Biochemical Research Methods
Lewis Chinery, Newton Wahome, Iain Moal, Charlotte M. Deane
Summary: The development of new vaccines and antibody therapeutics is a time-consuming and costly process. In this study, a structure-based paratope prediction tool called Paragraph is introduced, which outperforms existing tools by using simpler feature vectors and no antigen information.
