PFASUM: a substitution matrix from Pfam structural alignments

RESP2: An uncertainty aware multi-target multi-property optimization AI pipeline for antibody discovery

Discovery of therapeutic antibodies against infectious disease pathogens presents distinct challenges. Ideal candidates must possess not only the properties required for any therapeutic antibody (e.g. specificity, low immunogenicity) but also high affinity to many mutants of the target antigen. Here we present RESP2, an enhanced version of our RESP pipeline, designed for the discovery of antibodies against one or multiple antigens with simultaneously optimized developability properties. We first evaluate this pipeline in silico using the Absolut! database of scores for antibodies docked to target antigens. We show that RESP2 consistently identifies sequences that bind more tightly to a group of target antigens than any sequence present in the training set with success rates >= 85%. Popular generative AI techniques evaluated on the same datasets achieve success rates of 1.5% or less by comparison. Next we use the receptor binding domain (RBD) of the COVID-19 spike protein as a case study, and discover a highly human antibody with broad (mid to high-affinity) binding to at least 8 different variants of the RBD. These results illustrate the advantages of this pipeline for antibody discovery against a challenging target. A Python package that enables users to utilize the RESP pipeline on their own targets is available at https://github.com/Wang-lab-UCSD/RESP2, together with code needed to reproduce the experiments in this paper.

Computational scoring and experimental evaluation of enzymes generated by neural networks

In recent years, generative protein sequence models have been developed to sample novel sequences. However, predicting whether generated proteins will fold and function remains challenging. We evaluate a set of 20 diverse computational metrics to assess the quality of enzyme sequences produced by three contrasting generative models: ancestral sequence reconstruction, a generative adversarial network and a protein language model. Focusing on two enzyme families, we expressed and purified over 500 natural and generated sequences with 70-90% identity to the most similar natural sequences to benchmark computational metrics for predicting in vitro enzyme activity. Over three rounds of experiments, we developed a computational filter that improved the rate of experimental success by 50-150%. The proposed metrics and models will drive protein engineering research by serving as a benchmark for generative protein sequence models and helping to select active variants for experimental testing.

Tandem Repeats Provide Evidence for Convergent Evolution to Similar Protein Structures

Homology is a key concept underpinning the comparison of sequences across organisms. Sequence-level homology is based on a statistical framework optimized over decades of work. Recently, computational protein structure prediction has enabled large-scale homology inference beyond the limits of accurate sequence alignment. In this regime, it is possible to observe nearly identical protein structures lacking detectable sequence similarity. In the absence of a robust statistical framework for structure comparison, it is largely assumed similar structures are homologous. However, it is conceivable that matching structures could arise through convergent evolution, resulting in analogous proteins without shared ancestry. Large databases of predicted structures offer a means of determining whether analogs are present among structure matches. Here, I find that a small subset (∼2.6%) of Foldseek clusters lack sequence-level support for homology, including ∼1% of strong structure matches with template modeling score ≥ 0.5. This result by itself does not imply these structure pairs are nonhomologous, since their sequences could have diverged beyond the limits of recognition. Yet, strong matches without sequence-level support for homology are enriched in structures with predicted repeats that could induce spurious matches. Some of these structural repeats are underpinned by sequence-level tandem repeats in both matching structures. I show that many of these tandem repeat units have genealogies inconsistent with their corresponding structures sharing a common ancestor, implying these highly similar structure pairs are analogous rather than homologous. This result suggests caution is warranted when inferring homology from structural resemblance alone in the absence of sequence-level support for homology.

A BLAST from the past: revisiting blastp's E-value

MOTIVATION

The Basic Local Alignment Search Tool, BLAST, is an indispensable tool for genomic research. BLAST has established itself as the canonical tool for sequence similarity search in large part thanks to its meaningful statistical analysis. Specifically, BLAST reports the E-value of each reported alignment, which is defined as the expected number of optimal local alignments that will score at least as high as the observed alignment score, assuming that the query and the database sequences are randomly generated.

RESULTS

Here, we critically evaluate the E-values provided by the standard protein BLAST (blastp), showing that they can be at times significantly conservative while at others too liberal. We offer an alternative approach based on generating a small sample from the null distribution of random optimal alignments, and testing whether the observed alignment score is consistent with it. In contrast with blastp, our significance analysis seems valid, in the sense that it did not deliver inflated significance estimates in any of our extensive experiments. Moreover, although our method is slightly conservative, it is often significantly less so than the blastp E-value. Indeed, in cases where blastp's analysis is valid (i.e., not too liberal), our approach seems to deliver a greater number of correct alignments. One advantage of our approach is that it works with any reasonable choice of substitution matrix and gap penalties, avoiding blastp's limited options of matrices and penalties. In addition, we can formulate the problem using a canonical family-wise error rate control setup, thereby dispensing with E-values, which can at times be difficult to interpret.

AVAILABILITY AND IMPLEMENTATION

The Apache licensed source code is available at https://github.com/batmen-lab/SGPvalue.

tcrBLOSUM: an amino acid substitution matrix for sensitive alignment of distant epitope-specific TCRs

Deciphering the specificity of T-cell receptor (TCR) repertoires is crucial for monitoring adaptive immune responses and developing targeted immunotherapies and vaccines. To elucidate the specificity of previously unseen TCRs, many methods employ the BLOSUM62 matrix to find TCRs with similar amino acid (AA) sequences. However, while BLOSUM62 reflects the AA substitutions within conserved regions of proteins with similar functions, the remarkable diversity of TCRs means that both TCRs with similar and dissimilar sequences can bind the same epitope. Therefore, reliance on BLOSUM62 may bias detection towards epitope-specific TCRs with similar biochemical properties, overlooking those with more diverse AA compositions. In this study, we introduce tcrBLOSUMa and tcrBLOSUMb, specialized AA substitution matrices for CDR3 alpha and CDR3 beta TCR chains, respectively. The matrices reflect AA frequencies and variations occurring within TCRs that bind the same epitope, revealing that both CDR3 alpha and CDR3 beta display tolerance to a wide range of AA substitutions and differ noticeably from the standard BLOSUM62. By accurately aligning distant TCRs employing tcrBLOSUMb, we were able to improve clustering performance and capture a large number of epitope-specific TCRs with diverse AA compositions and physicochemical profiles overlooked by BLOSUM62. Utilizing both the general BLOSUM62 and specialized tcrBLOSUM matrices in existing computational tools will broaden the range of TCRs that can be associated with their cognate epitopes, thereby enhancing TCR repertoire analysis.

SHARK enables sensitive detection of evolutionary homologs and functional analogs in unalignable and disordered sequences

Intrinsically disordered regions (IDRs) are structurally flexible protein segments with regulatory functions in multiple contexts, such as in the assembly of biomolecular condensates. Since IDRs undergo more rapid evolution than ordered regions, identifying homology of such poorly conserved regions remains challenging for state-of-the-art alignment-based methods that rely on position-specific conservation of residues. Thus, systematic functional annotation and evolutionary analysis of IDRs have been limited, despite them comprising ~21% of proteins. To accurately assess homology between unalignable sequences, we developed an alignment-free sequence comparison algorithm, SHARK (Similarity/Homology Assessment by Relating K-mers). We trained SHARK-dive, a machine learning homology classifier, which achieved superior performance to standard alignment-based approaches in assessing evolutionary homology in unalignable sequences. Furthermore, it correctly identified dissimilar but functionally analogous IDRs in IDR-replacement experiments reported in the literature, whereas alignment-based tools were incapable of detecting such functional relationships. SHARK-dive not only predicts functionally similar IDRs at a proteome-wide scale but also identifies cryptic sequence properties and motifs that drive remote homology and analogy, thereby providing interpretable and experimentally verifiable hypotheses of the sequence determinants that underlie such relationships. SHARK-dive acts as an alternative to alignment to facilitate systematic analysis and functional annotation of the unalignable protein universe.

Accurately clustering biological sequences in linear time by relatedness sorting

Clustering biological sequences into similar groups is an increasingly important task as the number of available sequences continues to grow exponentially. Search-based approaches to clustering scale super-linearly with the number of input sequences, making it impractical to cluster very large sets of sequences. Approaches to clustering sequences in linear time currently lack the accuracy of super-linear approaches. Here, I set out to develop and characterize a strategy for clustering with linear time complexity that retains the accuracy of less scalable approaches. The resulting algorithm, named Clusterize, sorts sequences by relatedness to linearize the clustering problem. Clusterize produces clusters with accuracy rivaling popular programs (CD-HIT, MMseqs2, and UCLUST) but exhibits linear asymptotic scalability. Clusterize generates higher accuracy and oftentimes much larger clusters than Linclust, a fast linear time clustering algorithm. I demonstrate the utility of Clusterize by accurately solving different clustering problems involving millions of nucleotide or protein sequences.

New alignment method for remote protein sequences by the direct use of pairwise sequence correlations and substitutions

Understanding protein sequences and how they relate to the functions of proteins is extremely important. One of the most basic operations in bioinformatics is sequence alignment and usually the first things learned from these are which positions are the most conserved and often these are critical parts of the structure, such as enzyme active site residues. In addition, the contact pairs in a protein usually correspond closely to the correlations between residue positions in the multiple sequence alignment, and these usually change in a systematic and coordinated way, if one position changes then the other member of the pair also changes to compensate. In the present work, these correlated pairs are taken as anchor points for a new type of sequence alignment. The main advantage of the method here is its combining the remote homolog detection from our method PROST with pairwise sequence substitutions in the rigorous method from Kleinjung et al. We show a few examples of some resulting sequence alignments, and how they can lead to improvements in alignments for function, even for a disordered protein.

Mutation Space of Spatially Conserved Amino Acid Sites in Proteins

The mutation space of spatially conserved (MSSC) amino acid residues is a protein structural quantity developed and described in this work. The MSSC quantifies how many mutations and which different mutations, i.e., the mutation space, occur in each amino acid site in a protein. The MSSC calculates the mutation space of amino acids in a target protein from the spatially conserved residues in a group of multiple protein structures. Spatially conserved amino acid residues are identified based on their relative positions in the protein structure. The MSSC examines each residue in a target protein, compares it to the residues present in the same relative position in other protein structures, and uses physicochemical criteria of mutations found in each conserved spatial site to quantify the mutation space of each amino acid in the target protein. The MSSC is analogous to scoring each site in a multiple sequence alignment but in three-dimensional space considering the spatial location of residues instead of solely the order in which they appear in a protein sequence. MSSC analysis was performed on example cases, and it reproduces the well-known observation that, regardless of secondary structure, solvent-exposed residues are more likely to be mutated than internal ones. The MSSC code is available on GitHub: "https://github.com/Cantu-Research-Group/Mutation_Space".

Deep embedding and alignment of protein sequences

Protein sequence alignment is a key component of most bioinformatics pipelines to study the structures and functions of proteins. Aligning highly divergent sequences remains, however, a difficult task that current algorithms often fail to perform accurately, leaving many proteins or open reading frames poorly annotated. Here we leverage recent advances in deep learning for language modeling and differentiable programming to propose DEDAL (deep embedding and differentiable alignment), a flexible model to align protein sequences and detect homologs. DEDAL is a machine learning-based model that learns to align sequences by observing large datasets of raw protein sequences and of correct alignments. Once trained, we show that DEDAL improves by up to two- or threefold the alignment correctness over existing methods on remote homologs and better discriminates remote homologs from evolutionarily unrelated sequences, paving the way to improvements on many downstream tasks relying on sequence alignment in structural and functional genomics.

Bridging the gaps in statistical models of protein alignment

Summary

Sequences of proteins evolve by accumulating substitutions together with insertions and deletions (indels) of amino acids. However, it remains a common practice to disconnect substitutions and indels, and infer approximate models for each of them separately, to quantify sequence relationships. Although this approach brings with it computational convenience (which remains its primary motivation), there is a dearth of attempts to unify and model them systematically and together. To overcome this gap, this article demonstrates how a complete statistical model quantifying the evolution of pairs of aligned proteins can be constructed using a time-parameterized substitution matrix and a time-parameterized alignment state machine. Methods to derive all parameters of such a model from any benchmark collection of aligned protein sequences are described here. This has not only allowed us to generate a unified statistical model for each of the nine widely used substitution matrices (PAM, JTT, BLOSUM, JO, WAG, VTML, LG, MIQS and PFASUM), but also resulted in a new unified model, MMLSUM. Our underlying methodology measures the Shannon information content using each model to explain losslessly any given collection of alignments, which has allowed us to quantify the performance of all the above models on six comprehensive alignment benchmarks. Our results show that MMLSUM results in a new and clear overall best performance, followed by PFASUM, VTML, BLOSUM and MIQS, respectively, amongst the top five. We further analyze the statistical properties of MMLSUM model and contrast it with others.

Supplementary information

Supplementary data are available at Bioinformatics online.

RPfam: A refiner towards curated-like multiple sequence alignments of the Pfam protein families

High-quality multiple sequence alignments can provide insights into the architecture and function of protein families. The existing MSA tools often generate results inconsistent with biological distribution of conserved regions because of positioning amino acid residues and gaps only by symbols. We propose RPfam, a refiner towards curated-like MSAs for modeling the protein families in the Pfam database. RPfam refines the automatic alignments via scoring alignments based on the PFASUM matrix, restricting realignments within badly aligned blocks, optimizing the block scores by dynamic programming, and running refinements iteratively using the Simulated Annealing algorithm. Experiments show RPfam effectively refined the alignments produced by the MSA tools ClustalO and Muscle with reference to the curated seed alignments of the Pfam protein families. Especially RPfam improved the quality of the ClustalO alignments by 4.4% and the Muscle alignments by 2.8% on the gp32 DNA binding protein-like family. Supplementary Table is available at http://www.worldscinet.com/jbcb/ .

Remote homology clustering identifies lowly conserved families of effector proteins in plant-pathogenic fungi

Plant diseases caused by fungal pathogens are typically initiated by molecular interactions between 'effector' molecules released by a pathogen and receptor molecules on or within the plant host cell. In many cases these effector-receptor interactions directly determine host resistance or susceptibility. The search for fungal effector proteins is a developing area in fungal-plant pathology, with more than 165 distinct confirmed fungal effector proteins in the public domain. For a small number of these, novel effectors can be rapidly discovered across multiple fungal species through the identification of known effector homologues. However, many have no detectable homology by standard sequence-based search methods. This study employs a novel comparison method (RemEff) that is capable of identifying protein families with greater sensitivity than traditional homology-inference methods, leveraging a growing pool of confirmed fungal effector data to enable the prediction of novel fungal effector candidates by protein family association. Resources relating to the RemEff method and data used in this study are available from https://figshare.com/projects/Effector_protein_remote_homology/87965.

Substitution scoring matrices for proteins - An overview

Sequence analysis is the primary and simplest approach to discover structural, functional and evolutionary details of related proteins. All the alignment based approaches of sequence analysis make use of amino acid substitution matrices, and the accuracy of the results largely depends on the type of scoring matrices used to perform alignment tasks. An amino acid substitution matrix is a 20 × 20 matrix in which the individual elements encapsulate the rates at which each of the 20 amino acid residues in proteins are substituted by other amino acid residues over time. In contrast to most globular/ordered proteins whose amino acids composition is considered as standard, there are several classes of proteins (e.g., transmembrane proteins) in which certain types of amino acid (e.g., hydrophobic residues) are enriched. These compositional differences among various classes of proteins are manifested in their underlying residue substitution frequencies. Therefore, each of the compositionally distinct class of proteins or protein segments should be studied using specific scoring matrices that reflect their distinct residue substitution pattern. In this review, we describe the development and application of various substitution scoring matrices peculiar to proteins with standard and biased compositions. Along with most commonly used standard matrices (PAM, BLOSUM, MD and VTML) that act as default parameters in various homologs search and alignment tools, different substitution scoring matrices specific to compositionally distinct class of proteins are discussed in detail.

The ranging of amino acids substitution matrices of various types in accordance with the alignment accuracy criterion

Background

The alignment of character sequences is important in bioinformatics. The quality of this procedure is determined by the substitution matrix and parameters of the insertion-deletion penalty function. These matrices are derived from sequence alignment and thus reflect the evolutionary process. Currently, in addition to evolutionary matrices, a large number of different background matrices have been obtained. To make an optimal choice of the substitution matrix and the penalty parameters, we conducted a numerical experiment using a representative sample of existing matrices of various types and origins.

Results

We tested both the classical evolutionary matrix series (PAM, Blosum, VTML, Pfasum); structural alignment based matrices, contact energy matrix, and matrix based on the properties of the genetic code. This study presents results for two test set types: first, we simulated sequences that reflect the divergent evolution; second, we performed tests on Balibase sequences. In both cases, we obtained the dependences of the alignment quality (Accuracy, Confidence) on the evolutionary distance between sequences and the evolutionary distance to which the substitution matrices correspond. Optimization of a combination of matrices and the penalty parameters was carried out for local and global alignment on the values of penalty function parameters.

Consequently, we found that the best alignment quality is achieved with matrices corresponding to the largest evolutionary distance. These matrices prove to be universal, i.e. suitable for aligning sequences separated by both large and small evolutionary distances. We analysed the correspondence of the correlation coefficients of matrices to the alignment quality. It was found that matrices showing high quality alignment have an above average correlation value, but the converse is not true.

Conclusions

This study showed that the best alignment quality is achieved with evolutionary matrices designed for long distances: Gonnet, VTML250, PAM250, MIQS, and Pfasum050. The same property is inherent in matrices not only of evolutionary origin, but also of another background corresponding to a large evolutionary distance. Therefore, matrices based on structural data show alignment quality close enough to its value for evolutionary matrices. This agrees with the idea that the spatial structure is more conservative than the protein sequence.

Precise phylogenetic analysis of microbial isolates and genomes from metagenomes using PhyloPhlAn 3.0

Microbial genomes are available at an ever-increasing pace, as cultivation and sequencing become cheaper and obtaining metagenome-assembled genomes (MAGs) becomes more effective. Phylogenetic placement methods to contextualize hundreds of thousands of genomes must thus be efficiently scalable and sensitive from closely related strains to divergent phyla. We present PhyloPhlAn 3.0, an accurate, rapid, and easy-to-use method for large-scale microbial genome characterization and phylogenetic analysis at multiple levels of resolution. PhyloPhlAn 3.0 can assign genomes from isolate sequencing or MAGs to species-level genome bins built from >230,000 publically available sequences. For individual clades of interest, it reconstructs strain-level phylogenies from among the closest species using clade-specific maximally informative markers. At the other extreme of resolution, it scales to large phylogenies comprising >17,000 microbial species. Examples including Staphylococcus aureus isolates, gut metagenomes, and meta-analyses demonstrate the ability of PhyloPhlAn 3.0 to support genomic and metagenomic analyses.

Phylogenomics of 10,575 genomes reveals evolutionary proximity between domains Bacteria and Archaea

Rapid growth of genome data provides opportunities for updating microbial evolutionary relationships, but this is challenged by the discordant evolution of individual genes. Here we build a reference phylogeny of 10,575 evenly-sampled bacterial and archaeal genomes, based on a comprehensive set of 381 markers, using multiple strategies. Our trees indicate remarkably closer evolutionary proximity between Archaea and Bacteria than previous estimates that were limited to fewer "core" genes, such as the ribosomal proteins. The robustness of the results was tested with respect to several variables, including taxon and site sampling, amino acid substitution heterogeneity and saturation, non-vertical evolution, and the impact of exclusion of candidate phyla radiation (CPR) taxa. Our results provide an updated view of domain-level relationships.