Set Of Protein Sequences Research Articles

Exploring the potential of structure-based deep learning approaches for T cell receptor design.

Deep learning methods, trained on the increasing set of available protein 3D structures and sequences, have substantially impacted the protein modeling and design field. These advancements have facilitated the creation of novel proteins, or the optimization of existing ones designed for specific functions, such as binding a target protein. Despite the demonstrated potential of such approaches in designing general protein binders, their application in designing immunotherapeutics remains relatively underexplored. A relevant application is the design of T cell receptors (TCRs). Given the crucial role of T cells in mediating immune responses, redirecting these cells to tumor or infected target cells through the engineering of TCRs has shown promising results in treating diseases, especially cancer. However, the computational design of TCR interactions presents challenges for current physics-based methods, particularly due to the unique natural characteristics of these interfaces, such as low affinity and cross-reactivity. For this reason, in this study, we explored the potential of two structure-based deep learning protein design methods, ProteinMPNN and ESM-IF1, in designing fixed-backbone TCRs for binding target antigenic peptides presented by the MHC through different design scenarios. To evaluate TCR designs, we employed a comprehensive set of sequence- and structure-based metrics, highlighting the benefits of these methods in comparison to classical physics-based design methods and identifying deficiencies for improvement.

Identification and characterization of specific motifs in effector proteins of plant parasites using MOnSTER

Plant pathogens cause billions of dollars of crop loss every year and are a major threat to global food security. Identifying and characterizing pathogens effectors is crucial towards their improved control. Because of their poor sequence conservation, effector identification is challenging, and current methods generate too many candidates without indication for prioritizing experimental studies. In most phyla, effectors contain specific sequence motifs which influence their localization and targets in the plant. Therefore, there is an urgent need to develop bioinformatics tools tailored for pathogen effectors. To circumvent these limitations, we have developed MOnSTER a specific tool that identifies clusters of motifs of protein sequences (CLUMPs). MOnSTER can be fed with motifs identified by de novo tools or from databases such as Pfam and InterProScan. The advantage of MOnSTER is the reduction of motif redundancy by clustering them and associating a score. This score encompasses the physicochemical properties of AAs and the motif occurrences. We built up our method to identify discriminant CLUMPs in oomycetes effectors. Consequently, we applied MOnSTER on plant parasitic nematodes and identified six CLUMPs in about 60% of the known nematode candidate parasitism proteins. Furthermore, we found co-occurrences of CLUMPs with protein domains important for invasion and pathogenicity. The potentiality of this tool goes beyond the effector characterization and can be used to easily cluster motifs and calculate the CLUMP-score on any set of protein sequences.

Exploring gene content with pangene graphs.

The gene content regulates the biology of an organism. It varies between species and between individuals of the same species. Although tools have been developed to identify gene content changes in bacterial genomes, none is applicable to collections of large eukaryotic genomes such as the human pangenome. We developed pangene, a computational tool to identify gene orientation, gene order and gene copy-number changes in a collection of genomes. Pangene aligns a set of input protein sequences to the genomes, resolves redundancies between protein sequences and constructs a gene graph with each genome represented as a walk in the graph. It additionally finds subgraphs, which we call bibubbles, that capture gene content changes. Applied to the human pangenome, pangene identifies known gene-level variations and reveals complex haplotypes that are not well studied before. Pangene also works with high-quality bacterial pangenome and reports similar numbers of core and accessory genes in comparison to existing tools. Source code at https://github.com/lh3/pangene; pre-built pangene graphs can be downloaded from https://zenodo.org/records/8118576 and visualized at https://pangene.bioinweb.org.

Efficient Motif Discovery in Protein Sequences Using a Branch and Bound Algorithm.

Identifying motifs within sets of protein sequences constitutes a pivotal challenge in proteomics, imparting insights into protein evolution, function prediction, and structural attributes. Motifs hold the potential to unveil crucial protein aspects like transcription factor binding sites and protein-protein interaction regions. However, prevailing techniques for identifying motif sequences in extensive protein collections often entail significant time investments. Furthermore, ensuring the accuracy of obtained results remains a persistent motif discovery challenge. This paper introduces an innovative approach-a branch and bound algorithm-for exact motif identification across diverse lengths. This algorithm exhibits superior performance in terms of reduced runtime and enhanced result accuracy, as compared to existing methods. To achieve this objective, the study constructs a comprehensive tree structure encompassing potential motif evolution pathways. Subsequently, the tree is pruned based on motif length and targeted similarity thresholds. The proposed algorithm efficiently identifies all potential motif subsequences, characterized by maximal similarity, within expansive protein sequence datasets. Experimental findings affirm the algorithm's efficacy, highlighting its superior performance in terms of runtime, motif count, and accuracy, in comparison to prevalent practical techniques.

Refined Evolutionary Trees Through an Exceptionally Compatible Alignment-Substitution Model

A phylogenetic tree commonly represents evolutionary relationships within a set of protein sequences. Various methods and strategies have been used to improve the accuracy of phylogenetic trees, but their capacity to derive a biologically credible relationship appears to be overestimated. Although the quality of the protein sequence alignment and the choice of substitution matrix are preliminary constraints to define the biological accuracy of the overlapped residues, the alignment is not iteratively optimized through the statistical testing of residue-substitution models. The exact alignment protocol and substitution model information are by default used for every sequence set by a server to construct an often-irrelevant phylogenetic tree, and no sequence-based tailoring of phylogenetic strategy is implemented by any server. Rigorously constructing 270 evolutionary trees, constructed using IQ-TREE based on 13 different alignments (Clustal-Omega, Kalign, MAFFT, MUSCLE, TCoffee, and Promals3D, as well as their HHPred-based hidden Markov model [HMM] alignments using HHPred) and nine substitution models (Dayhoff, JJT, block substitution matrix62, WAG, probability matrix from blocks [PMB], direct computation with mutability [DCMUT], JTTDCmut, LG, and variable time), the present study highlights the failure of the current methods and emphasizes the need for a more accurate scrutiny of the entire phylogenetic methodology. MUSCLE alignment and the LG and Dayhoff matrices yield more accurate phylogenetic results for sequences shorter than 500 residues for the log-likelihood measure. Moreover, Kalign 1 HMM alignment yields the top-ranked tree with the lowest tree length score with only the PMB matrix, making this substitution model more accurate in terms of total tree length score. The suggested strategy would be beneficial for understanding the potential pitfalls of phylogenetic inference and would aid us in deriving a more accurate evolutionary relationship for a sequence dataset.

PEPMatch: a tool to identify short peptide sequence matches in large sets of proteins

BackgroundNumerous tools exist for biological sequence comparisons and search. One case of particular interest for immunologists is finding matches for linear peptide T cell epitopes, typically between 8 and 15 residues in length, in a large set of protein sequences. Both to find exact matches or matches that account for residue substitutions. The utility of such tools is critical in applications ranging from identifying conservation across viral epitopes, identifying putative epitope targets for allergens, and finding matches for cancer-associated neoepitopes to examine the role of tolerance in tumor recognition.ResultsWe defined a set of benchmarks that reflect the different practical applications of short peptide sequence matching. We evaluated a suite of existing methods for speed and recall and developed a new tool, PEPMatch. The tool uses a deterministic k-mer mapping algorithm that preprocesses proteomes before searching, achieving a 50-fold increase in speed over methods such as the Basic Local Alignment Search Tool (BLAST) without compromising recall. PEPMatch’s code and benchmark datasets are publicly available.ConclusionsPEPMatch offers significant speed and recall advantages for peptide sequence matching. While it is of immediate utility for immunologists, the developed benchmarking framework also provides a standard against which future tools can be evaluated for improvements. The tool is available at https://nextgen-tools.iedb.org, and the source code can be found at https://github.com/IEDB/PEPMatch.

Generative Modeling, Design, and Analysis of Spider Silk Protein Sequences for Enhanced Mechanical Properties

AbstractSpider silks are remarkable materials characterized by superb mechanical properties such as strength, extensibility, and lightweightedness. Yet, to date, limited models are available to fully explore sequence‐property relationships for analysis and design. Here a custom generative large‐language model is proposed to enable the design of novel spider silk protein sequences to meet complex combinations of target mechanical properties. The model, pretrained on a large set of protein sequences, is fine‐tuned on ≈1,000 major ampullate spidroin (MaSp) sequences for which associated fiber‐level mechanical properties exist, to yield an end‐to‐end forward and inverse generative approach that is aplied in a multi‐agent strategy. Performance is assessed through: 1) a novelty analysis and protein type classification for generated spidroin sequences through Basic Local Alignment Search Tool (BLAST) searches, 2) property evaluation and comparison with similar sequences, 3) comparison of resulting molecular structures, and 4) a detailed sequence motif analyses. This work generates silk sequences with property combinations that do not exist in nature and develops a deeper understanding of the mechanistic roles of sequence patterns in achieving overarching key mechanical properties (elastic modulus (E), strength, toughness, failure strain). The model provides an efficient approach to expand the silkome dataset, facilitating further sequence‐structure analyses of silks, and establishes a foundation for synthetic silk design and optimization.

Protein language models can capture protein quaternary state

BackgroundDetermining a protein’s quaternary state, i.e. the number of monomers in a functional unit, is a critical step in protein characterization. Many proteins form multimers for their activity, and over 50% are estimated to naturally form homomultimers. Experimental quaternary state determination can be challenging and require extensive work. To complement these efforts, a number of computational tools have been developed for quaternary state prediction, often utilizing experimentally validated structural information. Recently, dramatic advances have been made in the field of deep learning for predicting protein structure and other characteristics. Protein language models, such as ESM-2, that apply computational natural-language models to proteins successfully capture secondary structure, protein cell localization and other characteristics, from a single sequence. Here we hypothesize that information about the protein quaternary state may be contained within protein sequences as well, allowing us to benefit from these novel approaches in the context of quaternary state prediction.ResultsWe generated ESM-2 embeddings for a large dataset of proteins with quaternary state labels from the curated QSbio dataset. We trained a model for quaternary state classification and assessed it on a non-overlapping set of distinct folds (ECOD family level). Our model, named QUEEN (QUaternary state prediction using dEEp learNing), performs worse than approaches that include information from solved crystal structures. However, it successfully learned to distinguish multimers from monomers, and predicts the specific quaternary state with moderate success, better than simple sequence similarity-based annotation transfer. Our results demonstrate that complex, quaternary state related information is included in such embeddings.ConclusionsQUEEN is the first to investigate the power of embeddings for the prediction of the quaternary state of proteins. As such, it lays out strengths as well as limitations of a sequence-based protein language model approach, compared to structure-based approaches. Since it does not require any structural information and is fast, we anticipate that it will be of wide use both for in-depth investigation of specific systems, as well as for studies of large sets of protein sequences. A simple colab implementation is available at: https://colab.research.google.com/github/Furman-Lab/QUEEN/blob/main/QUEEN_prediction_notebook.ipynb.

Improving ISOMAP Efficiency with RKS: A Comparative Study with t-Distributed Stochastic Neighbor Embedding on Protein Sequences

Data visualization plays a crucial role in gaining insights from high-dimensional datasets. ISOMAP is a popular algorithm that maps high-dimensional data into a lower-dimensional space while preserving the underlying geometric structure. However, ISOMAP can be computationally expensive, especially for large datasets, due to the computation of the pairwise distances between data points. The motivation behind this study is to improve efficiency by leveraging an approximate method, which is based on random kitchen sinks (RKS). This approach provides a faster way to compute the kernel matrix. Using RKS significantly reduces the computational complexity of ISOMAP while still obtaining a meaningful low-dimensional representation of the data. We compare the performance of the approximate ISOMAP approach using RKS with the traditional t-SNE algorithm. The comparison involves computing the distance matrix using the original high-dimensional data and the low-dimensional data computed from both t-SNE and ISOMAP. The quality of the low-dimensional embeddings is measured using several metrics, including mean squared error (MSE), mean absolute error (MAE), and explained variance score (EVS). Additionally, the runtime of each algorithm is recorded to assess its computational efficiency. The comparison is conducted on a set of protein sequences, used in many bioinformatics tasks. We use three different embedding methods based on k-mers, minimizers, and position weight matrix (PWM) to capture various aspects of the underlying structure and the relationships between the protein sequences. By comparing different embeddings and by evaluating the effectiveness of the approximate ISOMAP approach using RKS and comparing it against t-SNE, we provide insights on the efficacy of our proposed approach. Our goal is to retain the quality of the low-dimensional embeddings while improving the computational performance.

Determining epitope specificity of T-cell receptors with transformers.

T-cell receptors (TCRs) on T cells recognize and bind to epitopes presented by the major histocompatibility complex in case of an infection or cancer. However, the high diversity of TCRs, as well as their unique and complex binding mechanisms underlying epitope recognition, make it difficult to predict the binding between TCRs and epitopes. Here, we present the utility of transformers, a deep learning strategy that incorporates an attention mechanism that learns the informative features, and show that these models pre-trained on a large set of protein sequences outperform current strategies. We compared three pre-trained auto-encoder transformer models (ProtBERT, ProtAlbert, and ProtElectra) and one pre-trained auto-regressive transformer model (ProtXLNet) to predict the binding specificity of TCRs to 25 epitopes from the VDJdb database (human and murine). Two additional modifications were performed to incorporate gene usage of the TCRs in the four transformer models. Of all 12 transformer implementations (four models with three different modifications), a modified version of the ProtXLNet model could predict TCR-epitope pairs with the highest accuracy (weighted F1 score 0.55 simultaneously considering all 25 epitopes). The modification included additional features representing the gene names for the TCRs. We also showed that the basic implementation of transformers outperformed the previously available methods, i.e. TCRGP, TCRdist, and DeepTCR, developed for the same biological problem, especially for the hard-to-classify labels. We show that the proficiency of transformers in attention learning can be made operational in a complex biological setting like TCR binding prediction. Further ingenuity in utilizing the full potential of transformers, either through attention head visualization or introducing additional features, can extend T-cell research avenues. Data and code are available on https://github.com/InduKhatri/tcrformer.

AlexandrusPS: A User-Friendly Pipeline for the Automated Detection of Orthologous Gene Clusters and Subsequent Positive Selection Analysis.

The detection of adaptive selection in a system approach considering all protein-coding genes allows for the identification of mechanisms and pathways that enabled adaptation to different environments. Currently, available programs for the estimation of positive selection signals can be divided into two groups. They are either easy to apply but can analyze only one gene family at a time, restricting system analysis; or they can handle larger cohorts of gene families, but require considerable prerequisite data such as orthology associations, codon alignments, phylogenetic trees, and proper configuration files. All these steps require extensive computational expertise, restricting this endeavor to specialists. Here, we introduce AlexandrusPS, a high-throughput pipeline that overcomes technical challenges when conducting transcriptome-wide positive selection analyses on large sets of nucleotide and protein sequences. The pipeline streamlines 1) the execution of an accurate orthology prediction as a precondition for positive selection analysis, 2) preparing and organizing configuration files for CodeML, 3) performing positive selection analysis using CodeML, and 4) generating an output that is easy to interpret, including all maximum likelihood and log-likelihood test results. The only input needed from the user is the CDS and peptide FASTA files of proteins of interest. The pipeline is provided in a Docker image, requiring no program or module installation, enabling the application of the pipeline in any computing environment. AlexandrusPS and its documentation are available via GitHub (https://github.com/alejocn5/AlexandrusPS).

MC-NN: An End-to-End Multi-Channel Neural Network Approach for Predicting Influenza A Virus Hosts and Antigenic Types

Influenza poses a significant threat to public health, particularly among the elderly, young children, and people with underlying diseases. The manifestation of severe conditions, such as pneumonia, highlights the importance of preventing the spread of influenza. An accurate and cost-effective prediction of the host and antigenic subtypes of influenza A viruses is essential to addressing this issue, particularly in resource-constrained regions. In this study, we propose a multi-channel neural network model to predict the host and antigenic subtypes of influenza A viruses from hemagglutinin and neuraminidase protein sequences. Our model was trained on a comprehensive data set of complete protein sequences and evaluated on various test data sets of complete and incomplete sequences. The results demonstrate the potential and practicality of using multi-channel neural networks in predicting the host and antigenic subtypes of influenza A viruses from both full and partial protein sequences.

Learning to Read and Write in the Language of Proteins

Learning to Read and Write in the Language of Proteins

UniProt: the Universal Protein Knowledgebase in 2023.

The aim of the UniProt Knowledgebase is to provide users with a comprehensive, high-quality and freely accessible set of protein sequences annotated with functional information. In this publication we describe enhancements made to our data processing pipeline and to our website to adapt to an ever-increasing information content. The number of sequences in UniProtKB has risen to over 227 million and we are working towards including a reference proteome for each taxonomic group. We continue to extract detailed annotations from the literature to update or create reviewed entries, while unreviewed entries are supplemented with annotations provided by automated systems using a variety of machine-learning techniques. In addition, the scientific community continues their contributions of publications and annotations to UniProt entries of their interest. Finally, we describe our new website (https://www.uniprot.org/), designed to enhance our users' experience and make our data easily accessible to the research community. This interface includes access to AlphaFold structures for more than 85% of all entries as well as improved visualisations for subcellular localisation of proteins.

MLO Proteins from Tomato (Solanum lycopersicum L.) and Related Species in the Broad Phylogenetic Context.

MLO proteins are a family of transmembrane proteins in land plants that play an important role in plant immunity and host–pathogen interactions, as well as a wide range of development processes. Understanding the evolutionary history of MLO proteins is important for understanding plant physiology and health. In the present work, we conducted a phylogenetic analysis on a large set of MLO protein sequences from publicly available databases, specifically emphasising MLOs from the tomato plant and related species. As a result, 4886 protein sequences were identified and used to construct a phylogenetic tree. In comparison to previous findings, we identified nine phylogenetic clades, revealed the internal structure of clades I and II as additional clades and showed the presence of monocotyledon species in all MLO clades. We identified a set of 19 protein motifs that allowed for the identification of particular clades. Sixteen SlMLO proteins from tomato were located in the phylogenetic tree and identified in relation to homologous sequences from other Solanaceae species. The obtained results could be useful for further work on the use of MLO proteins in the study of mildew resistance in Solanaceae and other plant families.

A Greedy Clustering Algorithm for Multiple Sequence Alignment

This paper presents a strategy to tackle the Multiple Sequence Alignment (MSA) problem, which is one of the most important tasks in the biological sequence analysis. Its role is to align the sequences in their entirety to derive relationships and common characteristics between a set of protein or nucleotide sequences. The MSA problem was proved to be an NP-Hard problem. The proposed strategy incorporates a new idea based on the well-known divide and conquer paradigm. This paper presents a novel method of clustering sequences as a preliminary step to improve the final alignment; this decomposition can be used as an optimization procedure with any MSA aligner to explore promising alignments of the search space. In their solution, authors proposed to align the clusters in a parallel and distributed way in order to benefit from parallel architectures. The strategy was tested using classical benchmarks like BAliBASE, Sabre, Prefab4 and Oxm, and the experimental results show that it gives good results by comparing to the other aligners.

A Greedy Clustering Algorithm for Multiple Sequence Alignment

This paper presents a strategy to tackle the Multiple Sequence Alignment (MSA) problem, which is one of the most important tasks in the biological sequence analysis. Its role is to align the sequences in their entirety to derive relationships and common characteristics between a set of protein or nucleotide sequences. The MSA problem was proved to be an NP-Hard problem. The proposed strategy incorporates a new idea based on the well-known divide and conquer paradigm. This paper presents a novel method of clustering sequences as a preliminary step to improve the final alignment; this decomposition can be used as an optimization procedure with any MSA aligner to explore promising alignments of the search space. In their solution, authors proposed to align the clusters in a parallel and distributed way in order to benefit from parallel architectures. The strategy was tested using classical benchmarks like BAliBASE, Sabre, Prefab4 and Oxm, and the experimental results show that it gives good results by comparing to the other aligners.

Proteomic Identification and Meta-Analysis in Salvia hispanica RNA-Seq de novo Assemblies.

While proteomics has demonstrated its value for model organisms and for organisms with mature genome sequence annotations, proteomics has been of less value in nonmodel organisms that are unaccompanied by genome sequence annotations. This project sought to determine the value of RNA-Seq experiments as a basis for establishing a set of protein sequences to represent a nonmodel organism, in this case, the pseudocereal chia. Assembling four publicly available chia RNA-Seq datasets produced transcript sequence sets with a high BUSCO completeness, though the number of transcript sequences and Trinity “genes” varied considerably among them. After six-frame translation, ProteinOrtho detected substantial numbers of orthologs among other species within the taxonomic order Lamiales. These protein sequence databases demonstrated a good identification efficiency for three different LC-MS/MS proteomics experiments, though a seed proteome showed considerable variability in the identification of peptides based on seed protein sequence inclusion. If a proteomics experiment emphasizes a particular tissue, an RNA-Seq experiment incorporating that same tissue is more likely to support a database search identification of that proteome.

Extended many-item similarity indices for sets of nucleotide and protein sequences

Quantification of similarities between protein sequences or DNA/RNA strands is a (sub-)task that is ubiquitously present in bioinformatics workflows, and is usually accomplished by pairwise comparisons of sequences, utilizing simple (e.g. percent identity) or more intricate concepts (e.g. substitution scoring matrices). Complex tasks (such as clustering) rely on a large number of pairwise comparisons under the hood, instead of a direct quantification of set similarities. Based on our recently introduced framework that enables multiple comparisons of binary molecular fingerprints (i.e., direct calculation of the similarity of fingerprint sets), here we introduce novel symmetric similarity indices for analogous calculations on sets of character sequences with more than two (t) possible items (e.g. DNA/RNA sequences with t = 4, or protein sequences with t = 20). The features of these new indices are studied in detail with analysis of variance (ANOVA), and demonstrated with three case studies of protein/DNA sequences with varying degrees of similarity (or evolutionary proximity). The Python code for the extended many-item similarity indices is publicly available at: https://github.com/ramirandaq/tn_Comparisons.

DeepPPF: A deep learning framework for predicting protein family

Machine learning pipelines for protein functional family prediction are urgently needed especially now that only 1% of raw protein sequences have been manually annotated. Although existing machine learning algorithms have achieved a decent performance in modeling and predicting the functional families of protein sequences, they still have two drawbacks. First, biological dependencies among nucleotides are not rich enough to describe motifs for these methods. Also, existing algorithms are not accurate enough to predict the functional families of newly discovered proteins. To address the above limitations simultaneously, we propose a novel deep learning framework for predicting protein family, DeepPPF, which employs the word2vec technique in capturing distributional dependencies among nucleotides and discovers rich features from diverse motif lengths to characterize proteins. The novelty of the DeepPPF is in utilizing distributional dependencies among nucleotides. Experimental results on G protein-coupled receptor hierarchical datasets show the effectiveness of DeepPPF in achieving the state-of-the-art performance in items of Mathew’s correlation coefficients (MCC) of 97.62%, 88.45% and, 83.09% for family, sub-family and, sub-subfamily hierarchical levels, respectively. Also, DeepPPF outperformed existing methods in terms of prediction accuracy and Mathew’s correlation coefficients on the cluster of orthologous groups (COG) and phage of orthologous groups (POG) datasets. Furthermore, we analyzed the ability of DeepPPF framework to discover rich motifs for functional classes with the least sets of protein sequences. The experimental results show that rich motif discovery is key to improving the modeling performance of protein families through deep learning techniques. Finally, we investigated the effect of transferring a low-level functional domain level to a high-level functional domain and results show that the target domain prediction can be improved with transfer learning. Therefore, our proposed deep learning framework can be useful in characterizing protein functional families. The codes and datasets are available at https://github.com/CSUBioGroup/DeepPPF.