Amino Acid Sequence Of Protein Research Articles

Identifying virulence factors using graph transformer autoencoder with ESMFold-predicted structures

With the increasing resistance of bacterial pathogens to conventional antibiotics, antivirulence strategies targeting virulence factors (VFs) have become an effective new therapy for the treatment of pathogenic bacterial infections. Therefore, the identification and prediction of VFs can provide ideal candidate targets for the implementation of antivirulence strategies in treating infections caused by pathogenic bacteria. Currently, the existing computational models predominantly rely on the amino acid sequences of virulence proteins while overlooking structural information. Here, we propose a novel graph transformer autoencoder for VF identification (GTAE-VF), which utilizes ESMFold-predicted 3D structures and converts the VF identification problem into a graph-level prediction task. In an encoder-decoder framework, GTAE-VF adaptively learns both local and global information by integrating a graph convolutional network and a transformer to implement all-pair message passing, which can better capture long-range correlations and potential relationships. Extensive experiments on an independent test dataset demonstrate that GTAE-VF achieves reliable and robust prediction accuracy with an AUC of 0.963, which is consistently better than that of other structure-based and sequence-based approaches. We believe that GTAE-VF has the potential to emerge as a valuable tool for assessing VFs and devising antivirulence strategies.

Transfer learning to leverage larger datasets for improved prediction of protein stability changes

Amino acid mutations that lower a protein's thermodynamic stability are implicated in numerous diseases, and engineered proteins with enhanced stability can be important in research and medicine. Computational methods for predicting how mutations perturb protein stability are, therefore, of great interest. Despite recent advancements in protein design using deep learning, in silico prediction of stability changes has remained challenging, in part due to a lack of large, high-quality training datasets for model development. Here, we describe ThermoMPNN, a deep neural network trained to predict stability changes for protein point mutations given an initial structure. In doing so, we demonstrate the utility of a recently released megascale stability dataset for training a robust stability model. We also employ transfer learning to leverage a second, larger dataset by using learned features extracted from ProteinMPNN, a deep neural network trained to predict a protein's amino acid sequence given its three-dimensional structure. We show that our method achieves state-of-the-art performance on established benchmark datasets using a lightweight model architecture that allows for rapid, scalable predictions. Finally, we make ThermoMPNN readily available as a tool for stability prediction and design.

Identification of ARF genes in Cucurbita pepo L and analysis of expression patterns, and functional analysis of CpARF22 under drought, salt stress

BackgroundAuxin transcription factor (ARF) is an important transcription factor that transmits auxin signals and is involved in plant growth and development as well as stress response. However, genome-wide identification and responses to abiotic and pathogen stresses of the ARF gene family in Cucurbita pepo L, especially pathogen stresses, have not been reported.ResultsFinally, 33 ARF genes (CpARF01 to CpARF33) were identified in C.pepo from the Cucurbitaceae genome database using bioinformatics methods. The putative protein contains 438 to 1071 amino acids, the isoelectric point is 4.99 to 8.54, and the molecular weight is 47759.36 to 117813.27 Da, the instability index ranged from 40.74 to 68.94, and the liposoluble index ranged from 62.56 to 76.18. The 33 genes were mainly localized in the nucleus and cytoplasm, and distributed on 16 chromosomes unevenly. Phylogenetic analysis showed that 33 CpARF proteins were divided into 6 groups. According to the amino acid sequence of CpARF proteins, 10 motifs were identified, and 1,3,6,8,10 motifs were highly conserved in most of the CpARF proteins. At the same time, it was found that genes in the same subfamily have similar gene structures. Cis-elements and protein interaction networks predicted that CpARF may be involved in abiotic factors related to the stress response. QRT-PCR analysis showed that most of the CpARF genes were upregulated under NaCl, PEG, and pathogen treatment compared to the control. Subcellular localization showed that CpARF22 was localized in the nucleus. The transgenic Arabidopsis thaliana lines with the CpARF22 gene enhanced their tolerance to salt and drought stress.ConclusionIn this study, we systematically analyzed the CpARF gene family and its expression patterns under drought, salt, and pathogen stress, which improved our understanding of the ARF protein of zucchini, and laid a solid foundation for functional analysis of the CpARF gene.

Porcine epidemic diarrhea virus strain CH/HLJ/18 isolated in China: characterization and phylogenetic analysis

BackgroundPorcine epidemic diarrhea (PED) is an infectious disease of the digestive tract caused by the porcine epidemic diarrhea virus (PEDV), characterized by vomiting, severe diarrhea, and high mortality rates in piglets. In recent years, the distribution of this disease in China has remarkably increased, and its pathogenicity has also increased. PEDV has been identified as the main cause of viral diarrhea in piglets. This study aimed to understand the genetic evolution and diversity of PEDV to provide a theoretical basis for the development of new vaccines and the prevention and treatment of PED.MethodsA PEDV strain was isolated from the small intestine of a diarrheal piglet using Vero cells. The virus was identified using reverse transcription-polymerase chain reaction (RT-PCR), indirect immunofluorescence assay (IFA), and transmission electron microscopy. The whole genome sequence was sequenced, phylogenetic analysis was conducted using MEGA (version 7.0), and recombination analysis was performed using RDP4 and SimPlot. The S protein amino acid sequence was aligned using Cluster X (version 2.0), and the S protein was modeled using SWISS-MODEL to compare differences in structure and antigenicity. Finally, the piglets were inoculated with PEDV to evaluate its pathogenicity in newborn piglets.ResultPEDV strain CH/HLJ/18 was isolated. CH/HLJ/18 shared 89.4–99.2% homology with 52 reference strains of PEDV belonging to the GII-a subgroup. It was a recombinant strain of PEDV BJ-2011-1 and PEDV CH_hubei_2016 with a breakpoint located in ORF1b. Unique amino acid deletions and mutations were observed in the CH/HLJ/18 S protein. The piglets then developed severe watery diarrhea and died within 7 d of inoculation with CH/HLJ/18, suggesting that CH/HLJ/18 was highly pathogenic to newborn piglets.ConclusionA highly pathogenic recombinant PEDV GII-a strain, CH/HLJ/18, was identified in China, with unique deletion and mutation of amino acids in the S protein that may lead to changes in protein structure and antigenicity. These results will be crucial for understanding the prevalence and variation of PEDV and for preventing and controlling PED.

Analysis of the application of bioinformatics in the medicine

As humans develop, science is also rapidly evolving, and biological science and medical support are vital for humans need. The development of biology has been particularly important. It stretches from the initial understanding of plants and animals to human biology and micro molecular biology, from macro understanding of life to micro molecular biology. Although it has accumulated a lot of knowledge about biology, it still has many unknowns about this giant and precise system. However, in recent years, a new interdisciplinary and emerging discipline has greatly opened up peoples understanding about biological information-bioinformatics. The most fundamental research target of bioinformatics is different sequences. The most important is the amino acid sequence of proteins and the base sequence of DNA. To study sequences and their constituent components, bioinformatics has another major research target - the biological database. This involves analysing the structure, connotation, and function of biological information through basic protein and nucleic acid sequences. Therefore, bioinformatics has made tremendous applications in the medical field. This research mainly analyzes and discusses the advantages and disadvantages of the application of bioinformatics in the medical field.

RNA-seq reveals multifaceted gene expression response to Fab production in Escherichia coli fed-batch processes with particular focus on ribosome stalling

BackgroundEscherichia coli is a cost-effective expression system for production of antibody fragments like Fabs. Various yield improvement strategies have been applied, however, Fabs remain challenging to produce. This study aimed to characterize the gene expression response of commonly used E. coli strains BL21(DE3) and HMS174(DE3) to periplasmic Fab expression using RNA sequencing (RNA-seq). Two Fabs, Fabx and FTN2, fused to a post-translational translocation signal sequence, were produced in carbon-limited fed-batch cultivations.ResultsProduction of Fabx impeded cell growth substantially stronger than FTN2 and yields of both Fabs differed considerably. The most noticeable, common changes in Fab-producing cells suggested by our RNA-seq data concern the cell envelope. The Cpx and Psp stress responses, both connected to inner membrane integrity, were activated, presumably by recombinant protein aggregation and impairment of the Sec translocon. The data additionally suggest changes in lipopolysaccharide synthesis, adjustment of membrane permeability, and peptidoglycan maturation and remodeling. Moreover, all Fab-producing strains showed depletion of Mg2+, indicated by activation of the PhoQP two-component signal transduction system during the early stage and sulfur and phosphate starvation during the later stage of the process. Furthermore, our data revealed ribosome stalling, caused by the Fabx amino acid sequence, as a contributor to low Fabx yields. Increased Fabx yields were obtained by a site-specific amino acid exchange replacing the stalling sequence. Contrary to expectations, cell growth was not impacted by presence or removal of the stalling sequence. Considering ribosome rescue is a conserved mechanism, the substantial differences observed in gene expression between BL21(DE3) and HMS174(DE3) in response to ribosome stalling on the recombinant mRNA were surprising.ConclusionsThrough characterization of the gene expression response to Fab production under industrially relevant cultivation conditions, we identified potential cell engineering targets. Thereby, we hope to enable rational approaches to improve cell fitness and Fab yields. Furthermore, we highlight ribosome stalling caused by the amino acid sequence of the recombinant protein as a possible challenge during recombinant protein production.

Molecular identification and functional characterization of a transcription factor GeRAV1 from Gelsemium elegans

BackgroundGelsemium elegans is a traditional Chinese medicinal plant and temperature is one of the key factors affecting its growth. RAV (related to ABI3/VP1) transcription factor plays multiple roles in higher plants, including the regulation of plant growth, development, and stress response. However, RAV transcription factor in G. elegans has not been reported.ResultsIn this study, three novel GeRAV genes (GeRAV1-GeRAV3) were identified from the transcriptome of G. elegans under low temperature stress. Phylogenetic analysis showed that GeRAV1-GeRAV3 proteins were clustered into groups II, IV, and V, respectively. RNA-sequencing (RNA-seq) and real-time quantitative PCR (qRT-PCR) analyses indicated that the expression of GeRAV1 and GeRAV2 was increased in response to cold stress. Furthermore, the GeRAV1 gene was successfully cloned from G. elegans leaf. It encoded a hydrophilic, unstable, and non-secretory protein that contained both AP2 and B3 domains. The amino acid sequence of GeRAV1 protein shared a high similarity of 81.97% with Camptotheca acuminata CaRAV. Subcellular localization and transcriptional self-activation experiments demonstrated that GeRAV1 was a nucleoprotein without self-activating activity. The GeRAV1 gene was constitutively expressed in the leaves, stems, and roots of the G. elegans, with the highest expression levels in roots. In addition, the expression of the GeRAV1 gene was rapidly up-regulated under abscisic acid (ABA), salicylic acid (SA), and methyl jasmonate (MeJA) stresses, suggesting that it may be involved in hormonal signaling pathways. Moreover, GeRAV1 conferred improved cold and sodium chloride tolerance in Escherichia coli Rosetta cells.ConclusionsThese findings provided a foundation for further understanding on the function and regulatory mechanism of the GeRAV1 gene in response to low-temperature stress in G. elegans.

Protein aggregation: An overview.

In order for an ordered protein to perform its specific function, it must have a specific molecular structure. Information about this structure is encoded in the protein's amino acid sequence. The unique functional state is achieved as a result of a specific process, known as protein folding. However, as a result of partial or complete unfolding of the polypeptide chain, proteins may misfold and aggregate, leading to the formation of various aggregated structures, such as like amyloid aggregates with the cross-β structure. A variety of cellular biological processes can be affected by protein aggregates that consume essential factors necessary for maintaining proteostasis, which leads to the proteostasis imbalance and further accumulation of protein aggregates, often resulting in age-related neurodegenerative disease progression and aging. However, in addition to their well-established pathological effects, amyloids also play various physiological roles, and many important biological processes involve such 'functional amyloids'. This chapter represents a brief overview of the protein aggregation phenomenon outlines a timeline provides of some key discoveries in this exciting field.

Algorithm for in silico search of immunodominant epitopes for immunodiagnosis of vaccine-targeted infections

This publication presents an algorithm for selecting immunodominant epitopes that are regions of influenza virus proteins used in vaccination in 2023-2024 for use in immunodiagnostics. Using various remote access programmes, an algorithm for searching for immunodominant epitopes was implemented based on the analysis of the amino acid sequence of virus proteins included in vaccines marketed in the Republic of Belarus. This is a stage before synthesis of peptides for laboratory diagnostics of cellular response of immunity system to virus vaccine. That in the future will allow to use only suitable peptides for diagnostics of postvaccinal immune response. Thus, a different set of peptides allows to detect the response of the immunity system to the surface proteins of viruses from vaccines "Grippol", "Vaxigripp Tetra", "Influvak Tetra" or to other proteins of influenza viruses in case of disease. Which is an additional test after evaluation of the humoral response in the form of neutralising antibodies to haemagglutinin.

Decoding the relationship between cow's milk proteins and development of type 1 diabetes mellitus.

To analyze in silico the evidence of molecular mimicry between human beta-cell autoantigens and cow's milk proteins as a potential type 1 diabetes mellitus (T1DM) trigger. The in silico analysis was performed using bioinformatics tools to compare the amino acid sequences of cow's milk proteins (bovine serum albumin [BSA] and beta-lactoglobulin [BLG]) and human beta-cell autoantigens (glutamic acid decarboxylase-65 [GAD-65], insulin, and zinc transporter 8 [ZnT8]). The structural and functional characteristics of the proteins were analyzed to identify potential molecular mimicry mechanisms. The results of the in silico analysis showed significant sequence similarity between BSA/BLG and GAD-65/human insulin/ZnT8, ranging from 19.64% to 27.27%. The cow's milk proteins evaluated shared structural features with the beta-cell antigens selected for comparison, indicating a potential for molecular mimicry between these proteins. The findings of this study provide further evidence for a potential role of cow's milk proteins in triggering T1DM. The in silico analysis suggests that molecular mimicry mechanisms between cow's milk proteins and human beta-cell antigens may contribute to the autoimmune response leading to T1DM.

Boosting Prediction of Protein-Protein Interactions using Word Embedding Techniques

Understanding protein-protein interactions (PPIs) helps to identify protein functions and develop other important applications such as drug preparation, protein-disease relationship identification. Machine learning methods have been developed for the PPI prediction task in order to reduce the cost and time of previous experimental methods. In this paper, we study a method for determining PPIs using deep learning and protein sequence representation learning. In our method, an word embedding technique is utilized for protein sequence representation learning. This technique captures the semantic relationship between amino acids in protein sequences. The semantic relationship is then used as the input information, which is fed into a neural network to help recognize the interaction signature of the input protein pair. Different from previous studies, we integrate the protein sequence embedding mechanism into a neural network model. Thereby, the protein sequence embedding is better controlled for PPI prediction by our neural network model. We evaluate our method on benchmark datasets including Yeast, Human, and eight different independent sets. In addition, we also conduct an extensive comparison with the other existing methods. Our results show that the proposed method is superior to other existing methods and achieves high efficiency in predicting cross-species PPIs. The dataset and our source code are available at https://github.com/thnhub/BoostPPIP.git.

Advancing drug–target interaction prediction: a comprehensive graph-based approach integrating knowledge graph embedding and ProtBert pretraining

BackgroundThe pharmaceutical field faces a significant challenge in validating drug target interactions (DTIs) due to the time and cost involved, leading to only a fraction being experimentally verified. To expedite drug discovery, accurate computational methods are essential for predicting potential interactions. Recently, machine learning techniques, particularly graph-based methods, have gained prominence. These methods utilize networks of drugs and targets, employing knowledge graph embedding (KGE) to represent structured information from knowledge graphs in a continuous vector space. This phenomenon highlights the growing inclination to utilize graph topologies as a means to improve the precision of predicting DTIs, hence addressing the pressing requirement for effective computational methodologies in the field of drug discovery.ResultsThe present study presents a novel approach called DTIOG for the prediction of DTIs. The methodology employed in this study involves the utilization of a KGE strategy, together with the incorporation of contextual information obtained from protein sequences. More specifically, the study makes use of Protein Bidirectional Encoder Representations from Transformers (ProtBERT) for this purpose. DTIOG utilizes a two-step process to compute embedding vectors using KGE techniques. Additionally, it employs ProtBERT to determine target–target similarity. Different similarity measures, such as Cosine similarity or Euclidean distance, are utilized in the prediction procedure. In addition to the contextual embedding, the proposed unique approach incorporates local representations obtained from the Simplified Molecular Input Line Entry Specification (SMILES) of drugs and the amino acid sequences of protein targets.ConclusionsThe effectiveness of the proposed approach was assessed through extensive experimentation on datasets pertaining to Enzymes, Ion Channels, and G-protein-coupled Receptors. The remarkable efficacy of DTIOG was showcased through the utilization of diverse similarity measures in order to calculate the similarities between drugs and targets. The combination of these factors, along with the incorporation of various classifiers, enabled the model to outperform existing algorithms in its ability to predict DTIs. The consistent observation of this advantage across all datasets underlines the robustness and accuracy of DTIOG in the domain of DTIs. Additionally, our case study suggests that the DTIOG can serve as a valuable tool for discovering new DTIs.

Identification and characterization of two closely related virga-like viruses latently infecting rubber trees (Hevea brasiliensis).

A novel virga-like virus, provisionally named Rubber tree latent virus 2 (RTLV2), was identified from rubber tree (Hevea brasiliensis). It is a close relative of the previously reported Rubber tree latent virus 1 (RTLV1). The complete genomes of RTLV1 and RTLV2 were sequenced and comparatively analyzed in terms of genome organization, putative gene products and phylogenetic relationship. Both RTLV1 and RTLV2 have positive-sense single-stranded RNA genomes that encode seven open reading frames (ORFs), forming a similar genomic layout. In phylogenetic analyses based on replicase and coat protein amino acid sequences, RTLV1 and RTLV2 were clustered with unclassified virga-like viruses. They are distinct from currently recognized plant virus families. RTLV1 and RTLV2 can be distinguished from members of Virgaviridae by the presence of a putative coat protein duplex and a poly(A) tail at the 3'-terminus. The authenticity of RTLV1 and RTLV2 as infectious viruses was confirmed through field investigations and transmissibility assays. In conclusion, RTLV1 and RTLV2 represent a novel plant virus group that does not readily fit into current virus families.

Acetone-induced structural variant of insulin amyloid fibrils

Self-propagating polymorphism of amyloid fibrils is a distinct manifestation of non-equilibrium conditions under which protein aggregation typically occurs. Structural variants of fibrils can often be accessed through physicochemical perturbations of the de novo aggregation process. On the other hand, tiny changes in the amino acid sequence of the parent protein may also result in structurally distinguishable amyloid fibrils. Here, we show that in the presence of acetone, the low-pH fibrillization pathway of bovine insulin (BI) leads to a new type of amyloid with the infrared features (split amide I′ band with the maximum at 1623 cm−1) bearing a striking resemblance to those of the previously reported fibrils from recombinant LysB31-ArgB32 human insulin analog formed in the absence of the co-solvent. Insulin fibrils formed in the presence ([BI-ace]) and absence ([BI]) of acetone cross-seed each other and pass their infrared features to the daughter generations of fibrils. We have used dimethyl sulfoxide (DMSO) coupled to in situ infrared spectroscopy measurements to probe the stability of fibrils against chemical denaturation. While both types of fibrils eventually undergo DMSO-induced disassembly coupled to a β-sheet→coil transition, in the case of [BI-ace] amyloid, the denaturation is preceded by the fibrils transiently acquiring the [BI]-like infrared characteristics. We argue that this effect is caused by DMSO-induced dehydration of [BI-ace]. In support to this hypothesis, we show that, even in the absence of DMSO, the infrared features of [BI-ace] disappear upon drying. We discuss this very peculiar aspect of [BI-ace] fibrils in the context of recently accessed in silico models of plausible structural variants of insulin protofilaments.

MEnTaT: A machine-learning approach for the identification of mutations to increase protein stability.

Enhancing protein thermal stability is important for biomedical and industrial applications as well as in the research laboratory. Here, we describe a simple machine-learning method which identifies amino acid substitutions that contribute to thermal stability based on comparison of the amino acid sequences of homologous proteins derived from bacteria that grow at different temperatures. A key feature of the method is that it compares the sequences based not simply on the amino acid identity, but rather on the structural and physicochemical properties of the side chain. The method accurately identified stabilizing substitutions in three well-studied systems and was validated prospectively by experimentally testing predicted stabilizing substitutions in a polyamine oxidase. In each case, the method outperformed the widely used bioinformatic consensus approach. The method can also provide insight into fundamental aspects of protein structure, for example, by identifying how many sequence positions in a given protein are relevant to temperature adaptation.

The roles of 6K protein on Getah virus replication and pathogenicity.

Alphavirus is a type of arbovirus that can infect both humans and animals. The amino acid sequence of the 6K protein, being one of the structural proteins of the alphavirus, is not conserved. Deletion of this protein will result in varying effects on different alphaviruses. Our study focuses on the function of the Getah virus (GETV) 6K protein in infected cells and mice. We successfully constructed infectious clone plasmids and created resulting viruses (rGETV and rGETV-Δ6K). Our comprehensive microscopic analysis revealed that the 6 K protein mainly stays in the endoplasmic reticulum. In addition, rGETV-Δ6K has lower thermal stability and sensitivity to temperature than GETV. Although the deletion of the 6K protein does not reduce virion production in ST cells, it affects the release of virions from host cells by inhibiting the process of E2 protein transportation to the plasma membrane. Subsequent in vivo testing demonstrated that neonatal mice infected with rGETV-Δ6K had a lower virus content, less significant pathological changes in tissue slices, and milder disease than those infected with the wild-type virus. Our results indicate that the 6K protein effectively reduces the viral titer by influencing the release of viral particles. Furthermore, the 6K protein play a role in the clinical manifestation of GETV disease.

Abstract A014: Deep learning-driven drug discovery: A breakthrough algorithm and its implication in lung cancer therapy development

Abstract We have developed an algorithm and implemented it in a software platform for the purpose of developing new anti-tumor drugs in the form of small molecules. In this study, we focused on generating molecules specifically for the treatment of lung cancer patients. To begin with, we employed deep learning (DL) techniques to evaluate the genes associated with poor clinical outcomes in lung cancer patients. By utilizing generative adversarial neural networks (GAN), we acquired additional patient data. The results of each experiment were presented as a list of genes ordered by their impact on the desired effect. We then intersected the lists of genes obtained from experiments with overall survival (OS) and progression-free interval (PFI) data. This allowed us to identify a set of genes whose expression was correlated with poor prognosis. In order to enhance the precision, we trained another DL model to distinguish between normal and tumor tissue based on gene expression. By doing so, we were able to identify the smaller set of genes that could be targeted. Subsequently, we developed a module that predicts the interactions between inhibitors and proteins. This involved representing protein amino acid sequences and chemical compound formulas in vector form, and a virtual screening of the Pubchem database. The implementation of the Drug-protein interactions module resulted in a dataset of 118,379 pairs, including 19,250 pairs describing compounds bound to proteins, and 99,129 precedents describing non-bound ones. DLwas applied, yielding a ROC-AUC of 0.86. Following the search for candidate molecules, we obtained 160,000 pairs with a predicted interaction probability above 0.99, as well as 2,921 pairs with probability of 1.0. Additionally, we created a DL-based module to predict the IC50 values in cell line experiments. Virtual pre-clinical trials were conducted using the selected inhibitors to identify relevant cell lines for subsequent laboratory experiments. Through this process, we obtained formulas for several molecules that demonstrated predicted binding to specific proteins. During the cell experiment emulation, our feature importance algorithm selected 129 genes. For the cell experiment emulation stage, we specifically chose interactions with a probability of at least 0.9. We prioritized molecules that acted on the minimum number of cell lines with a higher probability, thus ensuring higher specificity. Ultimately, we selected 5 small molecules as potential candidates, as well as certain cell lines for their validation. The NLP technologies utilized in this study demonstrated their effectiveness in processing tens of thousands of articles. The pipeline of methods presented in this paper lays the groundwork for automated AI-driven drug discovery. We have showcased the application of modern machine learning methods, particularly DL, as well as the methods used to prepare the initial data for the learning algorithms. The performance of these methods has been validated through cross-validation using data from publicly available sources. Citation Format: Dmitrii K Chebanov, Vsevolod A Misyurin, Nadezhda S Tatevosova. Deep learning-driven drug discovery: A breakthrough algorithm and its implication in lung cancer therapy development [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr A014.

Prediction of folding patterns for intrinsic disordered protein

The conformation flexibility of natural protein causes both complexity and difficulty to understand the relationship between structure and function. The prediction of intrinsically disordered protein primarily is focusing on to disclose the regions with structural flexibility involving relevant biological functions and various diseases. The order of amino acids in protein sequence determines possible conformations, folding flexibility and biological function. Although many methods provided the information of intrinsically disordered protein (IDP), but the results are mainly limited to determine the locations of regions without knowledge of possible folding conformations. Here, the developed protein folding fingerprint adopted the protein folding variation matrix (PFVM) to reveal all possible folding patterns for the intrinsically disordered protein along its sequence. The PFVM integrally exhibited the intrinsically disordered protein with disordering regions, degree of disorder as well as folding pattern. The advantage of PFVM will not only provide rich information for IDP, but also may promote the study of protein folding problem.

Image-centric compression of protein structures improves space savings

BackgroundBecause of the rapid generation of data, the study of compression algorithms to reduce storage and transmission costs is important to bioinformaticians. Much of the focus has been on sequence data, including both genomes and protein amino acid sequences stored in FASTA files. Current standard practice is to use an ordinary lossless compressor such as gzip on a sequential list of atomic coordinates, but this approach expends bits on saving an arbitrary ordering of atoms, and it also prevents reordering the atoms for compressibility. The standard MMTF and BCIF file formats extend this approach with custom encoding of the coordinates. However, the brand new Foldcomp tool introduces a new paradigm of compressing local angles, to great effect. In this article, we explore a different paradigm, showing for the first time that image-based compression using global angles can also significantly improve compression ratios. To this end, we implement a prototype compressor ‘PIC’, specialized for point clouds of atom coordinates contained in PDB and mmCIF files. PIC maps the 3D data to a 2D 8-bit greyscale image and leverages the well developed PNG image compressor to minimize the size of the resulting image, forming the compressed file.ResultsPIC outperforms gzip in terms of compression ratio on proteins over 20,000 atoms in size, with a savings over gzip of up to 37.4% on the proteins compressed. In addition, PIC’s compression ratio increases with protein size.ConclusionImage-centric compression as demonstrated by our prototype PIC provides a potential means of constructing 3D structure-aware protein compression software, though future work would be necessary to make this practical.

Immune targets to stop future SARS-CoV-2 variants

ABSTRACT The SARS-COV-2 variants of concern (VOCs) accumulated mutations that confer to the viral particle a higher infectivity as well as a capacity to escape neutralizing antibodies (NAbs) elicited by vaccines. In this study, we aimed to identify immune determinants that can be targeted to control future SARS-CoV-2 VOC. Using an immunoinformatics pipeline, we constructed a first data set consisting of 215 spike (S) protein amino acid sequences of the wild-type SARS-CoV-2, as well as of the Alpha, Beta, Delta, Gamma, and Omicron variants/subvariants (BA.1, BA.2, BA.4, BA.5, XBB, and BQ.1). A second data set was composed of epitope amino acid sequences for NAb as well as for T-cells involved in anti-viral activity, retrieved from the Immune Epitope Database (IEDB). Epitope conservation and population coverage analyses were carried out using the data sets. The localization of fully conserved epitopes in the S protein was performed using PyMOL. As main results, (i) fully conserved epitopes were identified: 28 for NAbs and 53 and 99 for class-I and class-II human leukocyte antigen, respectively; (ii) the fully conserved epitopes were shown to have high coverage in the world population (99.77% class combined); and (iii) the receptor-binding domain (RBD) of the S protein better balanced numbers of epitopes for NAb and T-cells, while the subunit two concentrated the highest numbers of T-cell epitopes. These results indicate that the RBD region holds different kinds of immune targets that could be used in an escape mutation-proof vaccine antigen to control future SARS-CoV-2 variants. IMPORTANCE The emergence of SARS-CoV-2 had a major impact across the world. It is true that the collaboration of scientists from all over the world resulted in a rapid response against COVID-19, mainly with the development of vaccines against the disease. However, many viral genetic variants that threaten vaccines have emerged. Our study reveals highly conserved antigenic regions in the vaccines have emerged. Our study reveals highly conserved antigenic regions in the spike protein in all variants of concern (Alpha, Beta, Gamma, Delta, and Omicron) as well as in the wild-type virus. Such immune targets can be used to fight future SARS-CoV-2 variants.