Search

The accurate prediction of binding between T cell receptors (TCR) and their cognate epitopes is key to understanding the adaptive immune response and developing immunotherapies. Current methods face two significant limitations: the shortage of comprehensive high-quality data and the bias introduced by the selection of the negative training data commonly used in the supervised learning approaches. We propose a method, Transformer-based Unsupervised Language model for Interacting Peptides and T cell receptors (TULIP), that addresses both limitations by leveraging incomplete data and unsupervised learning and using the transformer architecture of language models. Our model is flexible and integrates all possible data sources, regardless of their quality or completeness. We demonstrate the existence of a bias introduced by the sampling procedure used in previous supervised approaches, emphasizing the need for an unsupervised approach. TULIP recognizes the specific TCRs binding an epitope, performing well on unseen epitopes. Our model outperforms state-of-the-art models and offers a promising direction for the development of more accurate TCR epitope recognition models., Competing Interests: Competing interests statement:The authors declare no competing interest.

The emergence of new variants of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a major concern given their potential impact on the transmissibility and pathogenicity of the virus as well as the efficacy of therapeutic interventions. Here, we predict the mutability of all positions in SARS-CoV-2 protein domains to forecast the appearance of unseen variants. Using sequence data from other coronaviruses, preexisting to SARS-CoV-2, we build statistical models that not only capture amino acid conservation but also more complex patterns resulting from epistasis. We show that these models are notably superior to conservation profiles in estimating the already observable SARS-CoV-2 variability. In the receptor binding domain of the spike protein, we observe that the predicted mutability correlates well with experimental measures of protein stability and that both are reliable mutability predictors (receiver operating characteristic areas under the curve ∼0.8). Most interestingly, we observe an increasing agreement between our model and the observed variability as more data become available over time, proving the anticipatory capacity of our model. When combined with data concerning the immune response, our approach identifies positions where current variants of concern are highly overrepresented. These results could assist studies on viral evolution and future viral outbreaks and, in particular, guide the exploration and anticipation of potentially harmful future SARS-CoV-2 variants., Competing Interests: The authors declare no competing interest., (Copyright © 2022 the Author(s). Published by PNAS.)

Understanding the extreme variation among bacterial genomes remains an unsolved challenge in evolutionary biology, despite long-standing debate about the relative importance of natural selection, mutation, and random drift. A potentially important confounding factor is the variation in mutation rates between lineages and over evolutionary history, which has been documented in several species. Mutation accumulation experiments have shown that hypermutability can erode genomes over short timescales. These results, however, were obtained under conditions of extremely weak selection, casting doubt on their general relevance. Here, we circumvent this limitation by analyzing genomes from mutator populations that arose during a long-term experiment with Escherichia coli , in which populations have been adaptively evolving for >50,000 generations. We develop an analytical framework to quantify the relative contributions of mutation and selection in shaping genomic characteristics, and we validate it using genomes evolved under regimes of high mutation rates with weak selection (mutation accumulation experiments) and low mutation rates with strong selection (natural isolates). Our results show that, despite sustained adaptive evolution in the long-term experiment, the signature of selection is much weaker than that of mutational biases in mutator genomes. This finding suggests that relatively brief periods of hypermutability can play an outsized role in shaping extant bacterial genomes. Overall, these results highlight the importance of genomic draft, in which strong linkage limits the ability of selection to purge deleterious mutations. These insights are also relevant to other biological systems evolving under strong linkage and high mutation rates, including viruses and cancer cells., Competing Interests: The authors declare no conflict of interest., (Published under the PNAS license.)

Proteins have evolved to perform diverse cellular functions, from serving as reaction catalysts to coordinating cellular propagation and development. Frequently, proteins do not exert their full potential as monomers but rather undergo concerted interactions as either homo-oligomers or with other proteins as hetero-oligomers. The experimental study of such protein complexes and interactions has been arduous. Theoretical structure prediction methods are an attractive alternative. Here, we investigate homo-oligomeric interfaces by tracing residue coevolution via the global statistical direct coupling analysis (DCA). DCA can accurately infer spatial adjacencies between residues. These adjacencies can be included as constraints in structure prediction techniques to predict high-resolution models. By taking advantage of the ongoing exponential growth of sequence databases, we go significantly beyond anecdotal cases of a few protein families and apply DCA to a systematic large-scale study of nearly 2,000 Pfam protein families with sufficient sequence information and structurally resolved homo-oligomeric interfaces. We find that large interfaces are commonly identified by DCA. We further demonstrate that DCA can differentiate between subfamilies with different binding modes within one large Pfam family. Sequence-derived contact information for the subfamilies proves sufficient to assemble accurate structural models of the diverse protein-oligomers. Thus, we provide an approach to investigate oligomerization for arbitrary protein families leading to structural models complementary to often-difficult experimental methods. Combined with ever more abundant sequential data, we anticipate that this study will be instrumental to allow the structural description of many heteroprotein complexes in the future.

Understanding protein-protein interactions is central to our understanding of almost all complex biological processes. Computational tools exploiting rapidly growing genomic databases to characterize protein-protein interactions are urgently needed. Such methods should connect multiple scales from evolutionary conserved interactions between families of homologous proteins, over the identification of specifically interacting proteins in the case of multiple paralogs inside a species, down to the prediction of residues being in physical contact across interaction interfaces. Statistical inference methods detecting residue-residue coevolution have recently triggered considerable progress in using sequence data for quaternary protein structure prediction; they require, however, large joint alignments of homologous protein pairs known to interact. The generation of such alignments is a complex computational task on its own; application of coevolutionary modeling has, in turn, been restricted to proteins without paralogs, or to bacterial systems with the corresponding coding genes being colocalized in operons. Here we show that the direct coupling analysis of residue coevolution can be extended to connect the different scales, and simultaneously to match interacting paralogs, to identify interprotein residue-residue contacts and to discriminate interacting from noninteracting families in a multiprotein system. Our results extend the potential applications of coevolutionary analysis far beyond cases treatable so far., Competing Interests: The authors declare no conflict of interest.

Signal transduction proteins such as bacterial sensor histidine kinases, designed to transition between multiple conformations, are often ruled by unstable transient interactions making structural characterization of all functional states difficult. This study explored the inactive and signal-activated conformational states of the two catalytic domains of sensor histidine kinases, HisKA and HATPase. Direct coupling analyses, a global statistical inference approach, was applied to >13,000 such domains from protein databases to identify residue contacts between the two domains. These contacts guided structural assembly of the domains using MAGMA, an advanced molecular dynamics docking method. The active conformation structure generated by MAGMA simultaneously accommodated the sequence derived residue contacts and the ATP-catalytic histidine contact. The validity of this structure was confirmed biologically by mutation of contact positions in the Bacillus subtilis sensor histidine kinase KinA and by restoration of activity in an inactive KinA(HisKA):KinD(HATPase) hybrid protein. These data indicate that signals binding to sensor domains activate sensor histidine kinases by causing localized strain and unwinding at the end of the C-terminal helix of the HisKA domain. This destabilizes the contact positions of the inactive conformation of the two domains, identified by previous crystal structure analyses and by the sequence analysis described here, inducing the formation of the active conformation. This study reveals that structures of unstable transient complexes of interacting proteins and of protein domains are accessible by applying this combination of cross-validating technologies.

We introduce a theoretical framework that exploits the ever-increasing genomic sequence information for protein structure prediction. Structure-based models are modified to incorporate constraints by a large number of non-local contacts estimated from direct coupling analysis (DCA) of co-evolving genomic sequences. A simple hybrid method, called DCA-fold, integrating DCA contacts with an accurate knowledge of local information (e.g., the local secondary structure) is sufficient to fold proteins in the range of 1-3 Å resolution.

The similarity in the three-dimensional structures of homologous proteins imposes strong constraints on their sequence variability. It has long been suggested that the resulting correlations among amino acid compositions at different sequence positions can be exploited to infer spatial contacts within the tertiary protein structure. Crucial to this inference is the ability to disentangle direct and indirect correlations, as accomplished by the recently introduced direct-coupling analysis (DCA). Here we develop a computationally efficient implementation of DCA, which allows us to evaluate the accuracy of contact prediction by DCA for a large number of protein domains, based purely on sequence information. DCA is shown to yield a large number of correctly predicted contacts, recapitulating the global structure of the contact map for the majority of the protein domains examined. Furthermore, our analysis captures clear signals beyond intradomain residue contacts, arising, e.g., from alternative protein conformations, ligand-mediated residue couplings, and interdomain interactions in protein oligomers. Our findings suggest that contacts predicted by DCA can be used as a reliable guide to facilitate computational predictions of alternative protein conformations, protein complex formation, and even the de novo prediction of protein domain structures, contingent on the existence of a large number of homologous sequences which are being rapidly made available due to advances in genome sequencing.

Bacteria use two-component signal transduction systems (TCS) extensively to sense and react to external stimuli. In these, a membrane-bound sensor histidine kinase (SK) autophosphorylates in response to an environmental stimulus and transfers the phosphoryl group to a transcription factor/response regulator (RR) that mediates the cellular response. The complex between these two proteins is ruled by transient interactions, which provides a challenge to experimental structure determination techniques. The functional and structural homolog of an SK/RR pair Spo0B/Spo0F, however, has been structurally resolved. Here, we describe a method capable of generating structural models of such transient protein complexes. By using existing structures of the individual proteins, our method combines bioinformatically derived contact residue information with molecular dynamics simulations. We find crystal resolution accuracy with existing crystallographic data when reconstituting the known system Spo0B/Spo0F. Using this approach, we introduce a complex structure of TM0853/TM0468 as an exemplary SK/RR TCS, consistent with all experimentally available data.

Understanding the molecular determinants of specificity in protein-protein interaction is an outstanding challenge of postgenome biology. The availability of large protein databases generated from sequences of hundreds of bacterial genomes enables various statistical approaches to this problem. In this context covariance-based methods have been used to identify correlation between amino acid positions in interacting proteins. However, these methods have an important shortcoming, in that they cannot distinguish between directly and indirectly correlated residues. We developed a method that combines covariance analysis with global inference analysis, adopted from use in statistical physics. Applied to a set of >2,500 representatives of the bacterial two-component signal transduction system, the combination of covariance with global inference successfully and robustly identified residue pairs that are proximal in space without resorting to ad hoc tuning parameters, both for heterointeractions between sensor kinase (SK) and response regulator (RR) proteins and for homointeractions between RR proteins. The spectacular success of this approach illustrates the effectiveness of the global inference approach in identifying direct interaction based on sequence information alone. We expect this method to be applicable soon to interaction surfaces between proteins present in only 1 copy per genome as the number of sequenced genomes continues to expand. Use of this method could significantly increase the potential targets for therapeutic intervention, shed light on the mechanism of protein-protein interaction, and establish the foundation for the accurate prediction of interacting protein partners.