Sequence-Similar Protein Domain Pairs With Structural or Topological Dissimilarity.

For a variety of applications, protein structures are clustered by sequence similarity, and sequence-redundant structures are disregarded. Sequence-similar chains are likely to have similar structures, but significant structural variation, as measured with RMSD, has been documented for sequence-similar chains and found usually to have a functional explanation. Moving two neighboring stretches of backbone through each other may change the chain topology and alter possible folding paths. The size of this motion is compatible to a variation in a flexible loop. We search and find domains with alternate chain topology in CATH4.2 sequence families relatively independent of sequence identity and of structural similarity as measured by RMSD. Structural, topological, and functional representative sets should therefore keep sequence-similar domains not just with structural variation but also with topological variation. We present BCAlign that finds Alignment and superposition of protein Backbone Curves by optimizing a user chosen convex combination of structural derivation and derivation between the structure-based sequence alignment and an input sequence alignment. Steric and topological obstructions from deforming a curve into an aligned curve are then found by a previously developed algorithm. For highly sequence-similar domains, sequence-based structural alignment better represents the chains motion and generally reveals larger structural and topological variation than structure-based does. Fold-switching protein pairs have been reported to be most frequent between X-ray and NMR structures and estimated to be underrepresented in the PDB as the alternate configuration is harder to resolve. Here we similarly find chain topology most frequently altered between X-ray and NMR structures.

Assessing the role of evolutionary information for enhancing protein language model embeddings.

Embeddings from protein Language Models (pLMs) are replacing evolutionary information from multiple sequence alignments (MSAs) as the most successful input for protein prediction. Is this because embeddings capture evolutionary information? We tested various approaches to explicitly incorporate evolutionary information into embeddings on various protein prediction tasks. While older pLMs (SeqVec, ProtBert) significantly improved through MSAs, the more recent pLM ProtT5 did not benefit. For most tasks, pLM-based outperformed MSA-based methods, and the combination of both even decreased performance for some (intrinsic disorder). We highlight the effectiveness of pLM-based methods and find limited benefits from integrating MSAs.

Human lncRNAs harbor conserved modules embedded in different sequence contexts.

We analyzed the structure of human long non-coding RNA (lncRNAs) genes to investigate whether the non-coding transcriptome is organized in modular domains, as is the case for protein-coding genes. To this aim, we compared all known human lncRNA exons and identified 340 pairs of exons with high sequence and/or secondary structure similarity but embedded in a dissimilar sequence context. We grouped these pairs in 106 clusters based on their reciprocal similarities. These shared modules are highly conserved between humans and the four great ape species, display evidence of purifying selection and likely arose as a result of recent segmental duplications. Our analysis contributes to the understanding of the mechanisms driving the evolution of the non-coding genome and suggests additional strategies towards deciphering the functional complexity of this class of molecules.

Functional effects of mutations in proteins can be predicted and interpreted by guided selection of sequence covariation information.

Predicting the effects of one or more mutations to the in vivo or in vitro properties of a wild-type protein is a major computational challenge, due to the presence of epistasis, that is, of interactions between amino acids in the sequence. We introduce a computationally efficient procedure to build minimal epistatic models to predict mutational effects by combining evolutionary (homologous sequence) and few mutational-scan data. Mutagenesis measurements guide the selection of links in a sparse graphical model, while the parameters on the nodes and the edges are inferred from sequence data. We show, on 10 mutational scans, that our pipeline exhibits performances comparable to state-of-the-art deep networks trained on many more data, while requiring much less parameters and being hence more interpretable. In particular, the identified interactions adapt to the wild-type protein and to the fitness or biochemical property experimentally measured, mostly focus on key functional sites, and are not necessarily related to structural contacts. Therefore, our method is able to extract information relevant for one mutational experiment from homologous sequence data reflecting the multitude of structural and functional constraints acting on proteins throughout evolution.

Molecular, in silico and expression studies on lipoxygenases (LOXs) in potato (Solanum tuberosum L.).

Lipoxygenases (LOXs) namely 9-LOXs and 13-LOXs catalyse the oxygenation of polyunsaturated fatty acids to produce fatty acid hydroperoxides which are crucial in growth, development and stress responses in plants. Here, we isolated and characterized a 2723-bp cDNA encoding a distinct 861-aa 9-LOX form, designated StKCLX-1, using tuber total RNA from an Indian potato cultivar, Kufri Chipsona-1 through RT-PCR. A total of 17 LOX genes distributed in different chromosomes were identified and characterized in the potato genome. Multiple sequence alignment revealed highly conserved amino acids in the crucial domains, motifs and variable N-terminal regions between the LOX classes. A total of 36 LOXs from potato, tomato and Arabidopsis were used in phylogenetic analysis. A 3-D structure of StKCLX-1 was predicted by AlphaFold tool, validated through the predicted local-distance difference test (pLDDT) and Ramachandran Plot. Molecular docking predicted the nature of receptor-ligand interactions. STRING database was used to predict the protein-protein interactions. Expression patterns of the LOXs in the potato organs were examined by Expression Atlas and semi-quantitative RT-PCR. 9-LOX activity was noticed at early stages of tuberization, and significantly increased in the freshly-harvested mature tubers. This report would be useful in gaining insights into the structure-function relationships of the LOXs and corresponding multigene family-prerequisites for understanding tuber development in potato.

Bioinformatics Analysis of Global Diversity in Meningococcal Vaccine Antigens over the Past 10 Years: Vaccine Efficacy Prognosis.

Neisseria meningitidis (N. meningitidis) serogroup B (MenB) is the leading cause of invasive meningococcal disease worldwide. The pathogen has a wide range of virulence factors, which are potential vaccine components. Studying the genetic variability of antigens within a population, especially their long-term persistence, is necessary to develop new vaccines and predict the effectiveness of existing ones. The multicomponent 4CMenB vaccine (Bexsero), used since 2014, contains three major genome-derived recombinant proteins: factor H-binding protein (fHbp), Neisserial Heparin-Binding Antigen (NHBA) and Neisserial adhesin A (NadA). Here, we assessed the prevalence and sequence variations of these vaccine antigens in a panel of 5667 meningococcal isolates collected worldwide over the past 10 years and deposited in the PubMLST database. Using multiple amino acid sequence alignments and Random Forest Classifier machine learning methods, we estimated the potential strain coverage of fHbp and NHBA vaccine variants (51 and about 25%, respectively); the NadA antigen sequence was found in only 18% of MenB genomes analyzed, but cross-reactive variants were present in less than 1% of isolates. Based on our findings, we proposed various strategies to improve the 4CMenB vaccine and broaden the coverage of N. meningitidis strains.

SlimShadey: A manipulation tool for multiple sequence alignments.

The visualization of sequence alignments with the addition of meaningful shading and annotation is critical to convey the importance of structural elements, domains, motifs, and individual residues. Hence, we have developed a JavaFX based software package (SlimShadey) with an intuitive graphical user interface that allows for the creation and visualization of features on sequence alignments, as well as trimming and editing of subsequences. SlimShadey will run without modification on Windows 7 (or higher) and will also run on OS X / macOS, most Linux distributions, and servers. SlimShadey features real-time shading and comparison of residues based on user-defined measures of conservation such as frequency, user-selected substitution matrices, composition-based consensus sequence, regular expressions, and hidden Markov models. The software also allows users to generate custom sequence logos, configurable publication quality images of alignments with shading and annotation, and shareable self-contained project files for collaboration. SlimShadey is an open source freely available Java program. Compiled .jar executables, source code, supplementary materials including the user manual, links to video tutorials, and all sample data are available through the URLS at availability.

ETLD: an encoder-transformation layer-decoder architecture for protein contact and mutation effects prediction.

The latent features extracted from the multiple sequence alignments (MSAs) of homologous protein families are useful for identifying residue-residue contacts, predicting mutation effects, shaping protein evolution, etc. Over the past three decades, a growing body of supervised and unsupervised machine learning methods have been applied to this field, yielding fruitful results. Here, we propose a novel self-supervised model, called encoder-transformation layer-decoder (ETLD) architecture, capable of capturing protein sequence latent features directly from MSAs. Compared to the typical autoencoder model, ETLD introduces a transformation layer with the ability to learn inter-site couplings, which can be used to parse out the two-dimensional residue-residue contacts map after a simple mathematical derivation or an additional supervised neural network. ETLD retains the process of encoding and decoding sequences, and the predicted probabilities of amino acids at each site can be further used to construct the mutation landscapes for mutation effects prediction, outperforming advanced models such as GEMME, DeepSequence and EVmutation in general. Overall, ETLD is a highly interpretable unsupervised model with great potential for improvement and can be further combined with supervised methods for more extensive and accurate predictions.

Venom Proteomics of Trimeresurus gracilis, a Taiwan-Endemic Pitviper, and Comparison of Its Venom Proteome and VEGF and CRISP Sequences with Those of the Most Related Species.

Trimeresurus gracilis is an endemic alpine pitviper in Taiwan with controversial phylogeny, and its venom proteome remains unknown. In this study, we conducted a proteomic analysis of T. gracilis venom using high-performance liquid chromatography-tandem mass spectrometry and identified 155 toxin proteoforms that belong to 13 viperid venom toxin families. By searching the sequences of trypsin-digested peptides of the separated HPLC fractions against the NCBI database, T. gracilis venom was found to contain 40.3% metalloproteases (SVMPs), 15.3% serine proteases, 6.6% phospholipases A2, 5.0% L-amino acid oxidase, 4.6% Cys-rich secretory proteins (CRISPs), 3.2% disintegrins, 2.9% vascular endothelial growth factors (VEGFs), 1.9% C-type lectin-like proteins, and 20.2% of minor toxins, nontoxins, and unidentified peptides or compounds. Sixteen of these proteoforms matched the toxins whose full amino-acid sequences have been deduced from T. gracilis venom gland cDNA sequences. The hemorrhagic venom of T. gracilis appears to be especially rich in PI-class SVMPs and lacks basic phospholipase A2. We also cloned and sequenced the cDNAs encoding two CRISP and three VEGF variants from T. gracilis venom glands. Sequence alignments and comparison revealed that the PI-SVMP, kallikrein-like proteases, CRISPs, and VEGF-F of T. gracilis and Ovophis okinavensis are structurally most similar, consistent with their close phylogenetic relationship. However, the expression levels of some of their toxins were rather different, possibly due to their distinct ecological and prey conditions.

Creating De Novo Overlapped Genes.

Future applications of synthetic biology will rely on deploying engineered cells outside of lab environments for long periods of time. Currently, a significant roadblock to this application is the potential for deactivating mutations in engineered genes. A recently developed method to protect engineered coding sequences from mutation is called Constraining Adaptive Mutations using Engineered Overlapping Sequences (CAMEOS). In this chapter we provide a workflow for utilizing CAMEOS to create synthetic overlaps between two genes, one essential (infA) and one non-essential (aroB), to protect the non-essential gene from mutation and loss of protein function. In this workflow we detail the methods to collect large numbers of related protein sequences, produce multiple sequence alignments (MSAs), use the MSAs to generate hidden Markov models and Markov random field models, and finally generate a library of overlapping coding sequences through CAMEOS scripts. To assist practitioners with basic coding skills to try out the CAMEOS method, we have created a virtual machine containing all the required packages already installed that can be downloaded and run locally.

Evolution of γ chain cytokines: Mechanisms, methods and applications.

The common γ chain family of cytokines and their receptors play fundamental roles in the immune system. Evolutionary studies of γ chain cytokines have elegantly illustrated how the immune system adapts to ever-changing environmental conditions. Indeed, these studies have revealed the uniqueness of cytokine evolution, which exhibits strong positive selection pressure needed to adapt to rapidly evolving threats whilst still conserving their receptor binding capabilities. In this review, we summarise the evolutionary mechanisms that gave rise to the characteristically diverse family of γ chain cytokines. We also speculate on the benefits of studying cytokine evolution, which may provide alternative ways to design novel cytokine therapeutic strategies. Additionally, we discuss current evolutionary models that elucidate the emergence of distinct cytokines (IL-4 and IL-13) and cytokine receptors (IL-2Rα and IL-15Rα). Finally, we address and reflect on the difficulties associated with evolutionary studies of rapidly evolving genes and describe a variety of computational methods that have revealed numerous aspects of cytokine evolution.

3D structures inferred from cDNA clones identify the CD1D-Restricted γδ T cell receptor in dromedaries.

The Camelidae species occupy an important immunological niche within the humoral as well as cell mediated immune response. Although recent studies have highlighted that the somatic hypermutation (SHM) shapes the T cell receptor gamma (TRG) and delta (TRD) repertoire in Camelus dromedarius, it is still unclear how γδ T cells use the TRG/TRD receptors and their respective variable V-GAMMA and V-DELTA domains to recognize antigen in an antibody-like fashion. Here we report about 3D structural analyses of the human and dromedary γδ T cell receptor. First, we have estimated the interaction energies at the interface within the human crystallized paired TRG/TRD chains and quantified interaction energies within the same human TRG/TRD chains in complex with the CD1D, an RPI-MH1-LIKE antigen presenting glycoprotein. Then, we used the human TRG/TRD-CD1D complex as template for the 3D structure of the dromedary TRG/TRD-CD1D complex and for guiding the 3D human/dromedary comparative analysis. The choice of mutated TRG alternatively combined with mutated TRD cDNA clones originating from the spleen of one single dromedary was crucial to quantify the strength of the interactions at the protein-protein interface between the paired C. dromedarius TRG and TRD V-domains and between the C. dromedarius TRG/TRD V-domains and CD1D G-domains. Interacting amino acids located in the V-domain Complementarity Determining Regions (CDR) and Framework Regions (FR) according to the IMGT unique numbering for V-domains were identified. The resulting 3D dromedary TRG V-GAMMA combined with TRD V-DELTA protein complexes allowed to deduce the most stable gamma/delta chains pairings and to propose a candidate CD1D-restricted γδ T cell receptor complex.

Protein language-model embeddings for fast, accurate, and alignment-free protein structure prediction.

Advanced protein structure prediction requires evolutionary information from multiple sequence alignments (MSAs) from evolutionary couplings that are not always available. Artificial intelligence (AI)-based predictions inputting only single sequences are faster but so inaccurate as to render speed irrelevant. Here, we described a competitive prediction of inter-residue distances (2D structure) exclusively inputting embeddings from pre-trained protein language models (pLMs), namely ProtT5, from single sequences into a convolutional neural network (CNN) with relatively few layers. The major advance used the ProtT5 attention heads. Our new method, EMBER2, which never requires any MSAs, performed similarly to other methods that fully rely on co-evolution. Although clearly not reaching AlphaFold2, our leaner solution came somehow close at substantially lower costs. By generating protein-specific rather than family-averaged predictions, EMBER2 might better capture some features of particular protein structures. Results from using protein engineering and deep mutational scanning (DMS) experiments provided at least a proof of principle for such a speculation.

Boosting the analysis of protein interfaces with multiple interface string alignments: Illustration on the spikes of coronaviruses.

We introduce multiple interface string alignment (MISA), a visualization tool to display coherently various sequence and structure based statistics at protein-protein interfaces (SSE elements, buried surface area, ΔASA , B factor values, etc). The amino acids supporting these annotations are obtained from Voronoi interface models. The benefit of MISA is to collate annotated sequences of (homologous) chains found in different biological contexts, that is, bound with different partners or unbound. The aggregated views MISA/SSE, MISA/BSA, MISA/ΔASA, and so forth, make it trivial to identify commonalities and differences between chains, to infer key interface residues, and to understand where conformational changes occur upon binding. As such, they should prove of key relevance for knowledge-based annotations of protein databases such as the Protein Data Bank. Illustrations are provided on the receptor binding domain of coronaviruses, in complex with their cognate partner or (neutralizing) antibodies. MISA computed with a minimal number of structures complement and enrich findings previously reported. The corresponding package is available from the Structural Bioinformatics Library (http://sbl.inria.frand https://sbl.inria.fr/doc/Multiple_interface_string_alignment-user-manual.html).

Decoding the link of microbiome niches with homologous sequences enables accurately targeted protein structure prediction.

Information derived from metagenome sequences through deep-learning techniques has significantly improved the accuracy of template free protein structure modeling. However, most of the deep learning-based modeling studies are based on blind sequence database searches and suffer from low efficiency in computational resource utilization and model construction, especially when the sequence library becomes prohibitively large. We proposed a MetaSource model built on 4.25 billion microbiome sequences from four major biomes (Gut, Lake, Soil, and Fermentor) to decode the inherent linkage of microbial niches with protein homologous families. Large-scale protein family folding experiments on 8,700 unknown Pfam families showed that a microbiome targeted approach with multiple sequence alignment constructed from individual MetaSource biomes requires more than threefold less computer memory and CPU (central processing unit) time but generates contact-map and three-dimensional structure models with a significantly higher accuracy, compared with that using combined metagenome datasets. These results demonstrate an avenue to bridge the gap between the rapidly increasing metagenome databases and the limited computing resources for efficient genome-wide database mining, which provides a useful bluebook to guide future microbiome sequence database and modeling development for high-accuracy protein structure and function prediction.

Evolutionary coupling-inspired engineering of alcohol dehydrogenase reveals the influence of distant sites on its catalytic efficiency for stereospecific synthesis of chiral alcohols.

Alcohol dehydrogenase (ADH) has attracted much attention due to its ability to catalyze the synthesis of important chiral alcohol pharmaceutical intermediates with high stereoselectivity. ADH protein engineering efforts have generally focused on reshaping the substrate-binding pocket. However, distant sites outside the pocket may also affect its activity, although the underlying molecular mechanism remains unclear. The current study aimed to apply evolutionary coupling-inspired engineering to the ADH CpRCR and to identify potential mutation sites. Through conservative analysis, phylogenic analysis and residues distribution analysis, the co-evolution hotspots Leu34 and Leu137 were confirmed to be highly evolved under the pressure of natural selection and to be possibly related to the catalytic function of the protein. Hence, Leu34 and Leu137, far away from the active center, were selected for mutation. The generated CpRCR-L34A and CpRCR-L137V variants showed high stereoselectivity and 1.24-7.81 fold increase in k cat /K m value compared with that of the wild type, when reacted with 8 aromatic ketones or ß-ketoesters. Corresponding computational study implied that L34 and L137 may extend allosteric fluctuation in the protein structure from the distal mutational site to the active site. Moreover, the L34 and L137 mutations modified the pre-reaction state in multiple ways, in terms of position of the hydride with respect to the target carbonyl. These findings provide insights into the catalytic mechanism of the enzyme and facilitate its regulation from the perspective of the site interaction network.

DyP-Type Peroxidases: Recent Advances and Perspectives.

In this review, we chart the major milestones in the research progress on the DyP-type peroxidase family over the past decade. Though mainly distributed among bacteria and fungi, this family actually exhibits more widespread diversity. Advanced tertiary structural analyses have revealed common and different features among members of this family. Notably, the catalytic cycle for the peroxidase activity of DyP-type peroxidases appears to be different from that of other ubiquitous heme peroxidases. DyP-type peroxidases have also been reported to possess activities in addition to peroxidase function, including hydrolase or oxidase activity. They also show various cellular distributions, functioning not only inside cells but also outside of cells. Some are also cargo proteins of encapsulin. Unique, noteworthy functions include a key role in life-cycle switching in Streptomyces and the operation of an iron transport system in Staphylococcus aureus, Bacillus subtilis and Escherichia coli. We also present several probable physiological roles of DyP-type peroxidases that reflect the widespread distribution and function of these enzymes. Lignin degradation is the most common function attributed to DyP-type peroxidases, but their activity is not high compared with that of standard lignin-degrading enzymes. From an environmental standpoint, degradation of natural antifungal anthraquinone compounds is a specific focus of DyP-type peroxidase research. Considered in its totality, the DyP-type peroxidase family offers a rich source of diverse and attractive materials for research scientists.

The T Cell Receptor (TRB) Locus in Tursiops truncatus: From Sequence to Structure of the Alpha/Beta Heterodimer in the Human/Dolphin Comparison.

The bottlenose dolphin (Tursiops truncatus) belongs to the Cetartiodactyla and, similarly to other cetaceans, represents the most successful mammalian colonization of the aquatic environment. Here we report a genomic, evolutionary, and expression study of T. truncatus T cell receptor beta (TRB) genes. Although the organization of the dolphin TRB locus is similar to that of the other artiodactyl species, with three in tandem D-J-C clusters located at its 3' end, its uniqueness is given by the reduction of the total length due essentially to the absence of duplications and to the deletions that have drastically reduced the number of the germline TRBV genes. We have analyzed the relevant mature transcripts from two subjects. The simultaneous availability of rearranged T cell receptor α (TRA) and TRB cDNA from the peripheral blood of one of the two specimens, and the human/dolphin amino acids multi-sequence alignments, allowed us to calculate the most likely interactions at the protein interface between the alpha/beta heterodimer in complex with major histocompatibility class I (MH1) protein. Interacting amino acids located in the complementarity-determining region according to IMGT numbering (CDR-IMGT) of the dolphin variable V-alpha and beta domains were identified. According to comparative modelization, the atom pair contact sites analysis between the human MH1 grove (G) domains and the T cell receptor (TR) V domains confirms conservation of the structure of the dolphin TR/pMH.

Advances in integrative structural biology: Towards understanding protein complexes in their cellular context.

Microorganisms rely on protein interactions to transmit signals, react to stimuli, and grow. One of the best ways to understand these protein interactions is through structural characterization. However, in the past, structural knowledge was limited to stable, high-affinity complexes that could be crystallized. Recent developments in structural biology have revolutionized how protein interactions are characterized. The combination of multiple techniques, known as integrative structural biology, has provided insight into how large protein complexes interact in their native environment. In this mini-review, we describe the past, present, and potential future of integrative structural biology as a tool for characterizing protein interactions in their cellular context.