Biology subjects

Steinegger, M.

Publications and source records attributed to Steinegger, M..

At least 37 records · Page 2Linked to original sources

bioRxiv2025

Structural motif search across the protein-universe with Folddisco

Detecting similar protein structural motifs, functionally crucial short 3D patterns, in large structure collections is computationally prohibitive. Therefore, we developed Folddisco, which overcomes this through an index of position-independent geometric features, including side-chain orientation, combined with a rarity-based scoring system. Folddiscos compact 1.45-terabyte index of 53 million AFDB50 structures enables querying this database within seconds. It is 20-fold faster in querying and 4-fold more storage-efficient than existing methods. Folddisco is more accurate and storage-efficient than state-of-the-art methods, while being an order of magnitude faster. Folddisco is free software available at folddisco.foldseek.com and a webserver at search.foldseek.com/folddisco.

bioinformatics↗

bioRxiv2025

Ancient convergence with prokaryote defense and recent adaptations to lentiviruses in primates characterize the ancestral immune factors SAMD9s

Human SAMD9 and SAMD9L are duplicated genes that encode innate immune proteins restricting poxviruses and lentiviruses, such as HIV, and implicated in life-threatening genetic diseases and cancer. Here, we combined structural similarity searches, phylogenetics and population genomics with experimental assays of SAMD9/9L functions to resolve the evolutionary and functional dynamics of these immune proteins, spanning from prokaryotes to primates. We discovered structural analogs of SAMD9/9L in the anti-bacteriophage defense system Avs, resulting from convergent evolution. Further, the predicted nuclease active site was conserved in bacterial analogs and was essential for cell death functions, suggesting a fundamental role in defense across different life kingdoms. Despite this ancestral immunity, we identified genomic signatures of evolutionary arms-races in mammals, with remarkable gene copy number variations targeted by natural selection. We further unveiled that the absence of SAMD9 in bonobos corresponds to a recent gene loss still segregating in the population. Finally, we found that chimp and bonobo SAMD9Ls have enhanced anti-HIV-1 functions, and that bonobo-specific SAMD9L polymorphisms confer increased anti-HIV-1 activity to human SAMD9L without compromising its effect on cell translation. These SAMD9/9L adaptations likely resulted from strong viral selective pressures, including by primate lentiviruses, and could contribute to lentiviral resistance in bonobos. Altogether, this study elucidates the interplay between ancient immune convergence across kingdoms and species-specific adaptations within the Avs9 and SAMD9/9L antiviral shared immunity. Significance statementThe SAMD9 gene family encodes antiviral factors of poxviruses and lentiviruses/HIV and is implicated in genetic diseases. Here, we found strong structural similarity with proteins from the Avs anti-bacteriophage system and uncovered ancient functional convergence in immune strategies between prokaryotes and metazoans. Within mammals, and more importantly in primates, we describe a highly dynamic evolutionary history of the SAMD9 gene family that underwent adaptive episodic gene losses. Unlike humans and chimps, some bonobos lack the SAMD9 gene entirely. Bonobos and chimps also possess unique variants of SAMD9L enhancing anti-HIV-1 activity without compromising cell functions, suggesting super-restrictors. This could also participate in shaping SIVcpz evolution and contribute to the absence of lentivirus-infected bonobos. Overall, the seeming dichotomy between the ancient evolutionary convergence in different kingdoms and recent functional adaptation within primates highlights the arms-races between key immune defense systems and viruses. This study paves the way for evolutionary medicine, where evolutionary-based discoveries may have application to human health, providing a deeper understanding of how the immune system adapts to fight viral infections over billion years of evolution.

evolutionary biology↗

bioRxiv2025

SoftAlign: End-to-end protein structures alignment

With the recent breakthrough of highly accurate structure prediction methods, there has been a rapid growth of available protein structures. Efficient methods are needed to infer structural similarity within these datasets. We present an end-to-end alignment method, called SoftAlign, which takes as input the 3D coordinates of a protein pair and outputs a structural alignment. In addition to the traditional Smith-Waterman alignment method, we introduce a modified softmax alignment that shows very promising results for structure similarity detection. We demonstrate that the SoftAlign model is able to recapitulate TM-align alignments while running faster, and it is more accurate than Foldseek on alignment and classification tasks. Although SoftAlign is not the fastest method available, it is highly precise and can be used effectively with other prefilters. In addition to developing an end-to-end structural aligner, our main contribution is the introduction and analysis of a pseudo-alignment method based on softmax, which can be used with other architectures, even those not based on structural information. The code for SoftAlign is available at https://github.com/jtrinquier/SoftAlign.

bioinformatics↗

bioRxiv2025

Metagenomic-scale analysis of the predicted protein structure universe

Protein structure prediction breakthroughs, AlphaFold2 and ESMfold, have led to an unprecedented influx of computationally derived structures. The AlphaFold Protein Structure Database provides over 200 million predictions, while the ESM Metagenomic Atlas (ESMatlas) includes over 600 million predictions from uncultured microbes. We combine these into AFESM, an 820-million-entry dataset, and cluster them using a scalable pipeline based on sequence and structure similarity, yielding 5.12 million non-singleton structural clusters. We identify common ancestors and biomes for these clusters to explore their environmental diversity and specificity, and we investigate their structural novelties. From non-singleton clusters unique to ESMatlas, we identified 12 novel domain folds, repredicting a subset ([~]45%) of low-quality domains with ColabFold yielded 33 additional novel folds. This underscores the importance of prediction quality in structural novelty discovery. We also identified 11,941 previously unseen domain combinations, highlighting the untapped structural diversity and importance of metagenomic data for illuminating underexplored regions of the protein structural universe. An interactive webserver and data are available at afesm.foldseek.com.

bioinformatics↗

bioRxiv2025

An nf-core framework for the systematic comparison of alternative modeling tools: the multiple sequence alignment case study

The computational complexity of many key bioinformatics problems has resulted in numerous alternative heuristic solutions, where no single approach consistently outperforms all others. This creates difficulties for users trying to identify the most suitable tool for their dataset and for developers managing and evaluating alternative methods. As data volumes grow, deploying these methods becomes increasingly difficult, highlighting the need for standardized frameworks for seamless tool deployment and comparison in HPC environments. Multiple sequence aligners (MSAs) rank among the most commonly employed modeling techniques in bioinformatics, playing a crucial role in applications such as protein structure prediction, phylogenetic reconstruction, and variant effect prediction. The NP-hardness of MSAs makes them a major example of problems where heuristics stand central, as no optimal solution can be currently obtained, within the limits of operational computational requirements. Here, we present a pilot design of an nf-core framework for streamlined tool deployment and rigorous performance evaluation focusing on the MSAs software ecosystem. By integrating the most popular MSA tools and focusing on a modular, and extensible architecture, we aspire to provide a key platform supporting MSA deployment, evaluation, and algorithmics development to the MSA community, and a proof-of-principle to the wider bioinformatics community.

bioinformatics↗

bioRxiv2025

Easy and interactive taxonomic profiling with Metabuli App

SummaryAccurate metagenomic taxonomic profiling is critical for understanding microbial communities. However, computational analysis often requires command-line proficiency and high-performance computing resources. To lower these barriers, we developed Metabuli App, an all-in-one desktop application that efficiently runs taxonomic profiling locally on a consumer-grade computer. It features user-friendly graphical interfaces for custom database curation, raw read quality control (QC), taxonomic profiling, and interactive result visualization. Availability and implementationGPLv3-licensed source code and prebuilt apps for Windows, macOS, and Linux are available at https://github.com/steineggerlab/Metabuli-App and are archived at https://doi.org/10.5281/zenodo.15876171. Analysis scripts are available at https://github.com/jaebeom-kim/metabuli-app-analysis. The Sankey-based taxonomy visualization component is available at https://github.com/steineggerlab/taxoview for easy integration into other web projects. Contactmartin.steinegger@snu.ac.kr

bioinformatics↗

bioRxiv2024

Unicore enables scalable and accurate phylogenetic reconstruction with structural core genes

The analysis of single-copy core genes, common to most members of a clade, is important for key tasks in biology including phylogenetic reconstruction and assessing genome quality. Core genes are traditionally identified by the analysis of amino acid similarities among proteomes, but can also be defined using structures, which bear potential in deep clades beyond the twilight zone of amino acids. Despite breakthroughs in accurate AI-driven protein structure prediction, obtaining full 3D structural models on a proteomic scale is still prohibitively slow. Here, we present Unicore, a novel method for identifying structural core genes at a scale suitable for downstream phylogenetic analysis. By applying the ProstT5 protein language model to the input proteomes to obtain their 3Di structural strings, Unicore saves over three orders of magnitude in runtime compared to a full 3D prediction. Using Foldseek clustering, Unicore identifies single-copy structures universally present in the species and aligns them using Foldmason. These structural core gene alignments are projected back to amino acid information for downstream phylogenetic analysis. We demonstrate that this approach defines core genes with linear run-time scaling over the number of species, up to 56 times faster than OrthoFinder, while reconstructing phylogenetic relationships congruent with conventional approaches. Unicore is universally applicable to any given set of taxa, even spanning superkingdoms and overcoming limitations of previous methods requiring orthologs of fixed taxonomic scope, and is available as a free and open source software at https://github.com/steineggerlab/unicore.

bioinformatics↗

bioRxiv2024

GPU-accelerated homology search with MMseqs2

Rapidly growing protein databases demand faster sensitive sequence similarity detection. We present GPU-accelerated search utilizing intra-query parallelization delivering 6x faster single-protein searches compared to state-of-the-art CPU methods on 2x64 cores--speeds previously requiring large protein batches. It is most cost effective, including in large-batches at 0.45x MMseqs2-CPU speed (8 GPUs delivering 2.4x). It accelerates ColabFold structure prediction 31.8x compared to AlphaFold2 and Foldseek search 4-27x. MMseqs2-GPU is open-source at mmseqs.com.

bioinformatics↗

bioRxiv2024

Structural basis of the catalytic and allosteric mechanism of bacterial acetyltransferase PatZ

GCN5-related N-Acetyltransferases (GNATs) play a crucial role in regulating bacterial metabolism by acetylating specific target proteins. Despite their importance in bacterial physiology, the mechanisms underlying GNATs enzymatic and regulatory functions remain poorly understood. In this study, we elucidated the structures of Escherichia coli PatZ, a type I GNAT, and investigated its ligand interactions, catalytic processes, and allosterism. PatZ functions as a homotetramer, with each subunit comprising a catalytic domain and a regulatory domain. Our findings reveal that the regulatory domain is essential for acetyltransferase activity, as it not only induces cooperative conformational changes in the catalytic domain but also directly contributes to the formation of substrate binding pockets. Furthermore, a protein structure-based analysis on the evolution of bacterial GNAT types reveals a distinct pattern of the regulatory domain across phyla, underscoring the regulatory domains critical role in responding to cellular energy status. SIGNIFICANCE STATEMENTPost-translational modifications, particularly acetylation mediated by GCN5-related N-Acetyltransferases (GNATs), play a crucial role in bacterial physiology. Protein acetyltransferase Z (PatZ) is a key GNAT with diverse substrates, essential for understanding the bacterial acetylome. This study employs cryogenic electron microscopy, X-ray crystallography, and biochemical analyses to elucidate the mechanistic regulation of Escherichia coli PatZ. Our high-resolution structures reveal PatZs homo-tetrameric architecture, with each subunit comprising regulatory and GNAT domains. We characterize ligand-PatZ interactions, demonstrating ligand-induced conformational changes that facilitate allosteric regulation of the catalytic domain. Furthermore, our analyses elucidate the regulatory domains contribution to substrate binding pocket formation, potentially enhancing substrate specificity. Structure-based phylogenetic analysis provides insights into the evolution of diverse regulatory domains in the GNAT superfamily across bacterial taxonomy. This first visualization of PatZ advances our mechanistic understanding of bacterial physiology, offering novel insights into GNAT-mediated bacterial adaptations. HIGHLIGHTS- E. coli PatZ forms a homotetramer, with each subunit possessing a GNAT catalytic domain and a regulatory domain. - Cooperative binding of acetyl-CoA to the regulatory domains is a prerequisite for inducing the structural compatibility of the catalytic domain with a substrate. - Diverse regulatory domains in GNATs evolved to adapt to varied metabolic conditions across bacterial taxonomy.

biochemistry↗

bioRxiv2024

BFVD - a large repository of predicted viral protein structures

The AlphaFold Protein Structure Database (AFDB) is the largest repository of accurately predicted structures with taxonomic labels. Despite providing predictions for over 214 million UniProt entries, the AFDB does not cover viral sequences, severely limiting their study. To bridge this gap, we created the Big Fantastic Virus Database (BFVD), a repository of 351,242 protein structures predicted by applying ColabFold to the viral sequence representatives of the UniRef30 clusters. BFVD holds a unique repertoire of protein structures as over 63% of its entries show no or low structural similarity to existing repositories. We demonstrate how BFVD substantially enhances the fraction of annotated bacteriophage proteins compared to sequence-based annotation using Bakta. In that, BFVD is on par with the AFDB, while holding nearly three orders of magnitude fewer structures. BFVD is an important virus-specific expansion to protein structure repositories, offering new opportunities to advance viral research. BFVD is freely available at https://bfvd.steineggerlab.workers.dev/

bioinformatics↗

bioRxiv2024

CarpeDeam: A De Novo Metagenome Assemblerfor Heavily Damaged Ancient Datasets

De novo assembly of ancient metagenomic datasets is a challenging task. Ultra-short fragment size and characteristic postmortem damage patterns of sequenced ancient DNA molecules leave current tools ill-equipped for ideal assembly. We present CarpeDeam, a novel damage-aware de novo assembler designed specifically for ancient metagenomic samples. Utilizing maximum-likelihood frameworks that integrate sample-specific damage patterns, CarpeDeam demonstrates improved recovery of longer continuous sequences and protein sequences in many simulated and empirical datasets compared to existing assemblers. As a pioneering ancient metagenome assembler, CarpeDeam opens the door for new opportunities in functional and taxonomic analyses of ancient microbial communities.

genomics↗

bioRxiv2024

Multiple Protein Structure Alignment at Scale with FoldMason

Protein structure is conserved beyond sequence, making multiple structural alignment (MSTA) essential for analyzing distantly related proteins. Computational prediction methods have vastly extended our repository of available proteins structures, requiring fast and accurate MSTA methods. Here, we introduce FoldMason, a progressive MSTA method that leverages the structural alphabet from Foldseek, a pairwise structural aligner, for multiple alignment of hundreds of thousands of protein structures, exceeding alignment quality of state-of-the-art methods, while two orders of magnitudes faster than other MSTA methods. FoldMason computes confidence scores, offers interactive visualizations, and provides essential speed and accuracy for large-scale protein structure analysis in the era of accurate structure prediction. Using Flaviviridae glycoproteins, we demonstrate how FoldMasons MSTAs support phylogenetic analysis below the twilight zone. FoldMason is free open-source software: foldmason.foldseek.com and webserver: search.foldseek.com/foldmason.

bioinformatics↗

bioRxiv2024

Molecular Display of the Animal Meta-Venome for Discovery of Novel Therapeutic Peptides

Animal venoms, distinguished by their unique structural features and potent bioactivities, represent a vast and relatively untapped reservoir of therapeutic molecules. However, limitations associated with extracting or expressing large numbers of individual venoms and venom-like molecules have precluded their therapeutic evaluation via high throughput screening. Here, we developed an innovative computational approach to design a highly diverse library of animal venoms and "metavenoms". We employed programmable M13 hyperphage display to preserve critical disulfide-bonded structures for highly parallelized single-round biopanning with quantitation via high-throughput DNA sequencing. Our approach led to the discovery of Kunitz type domain containing proteins that target the human itch receptor Mas-related G protein-coupled receptor X4 (MRGPRX4), which plays a crucial role in itch perception. Deep learning-based structural homology mining identified two endogenous human homologs, tissue factor pathway inhibitor (TFPI) and serine peptidase inhibitor, Kunitz type 2 (SPINT2), which exhibit agonist-dependent potentiation of MRGPRX4. Highly multiplexed screening of animal venoms and metavenoms is therefore a promising approach to uncover new drug candidates.

molecular biology↗

bioRxiv2024

Rapid and Sensitive Protein Complex Alignment with Foldseek-Multimer

Advances in computational structure prediction will vastly augment the hundreds of thousands of currently-available protein complex structures. Translating these into discoveries requires aligning them, which is computationally prohibitive. Foldseek-Multimer computes complex alignments from compatible chain-to-chain alignments, identified by efficiently clustering their superposition vectors. Foldseek-Multimer is 3-4 orders of magnitudes faster than the gold standard, while producing comparable alignments; allowing it to compare billions of complex-pairs in 11 hours. Foldseek-Multimer is open-source software: github.com/steineggerlab/foldseek, webserver: search.foldseek.com and the BFMD database.

bioinformatics↗

bioRxiv2024

Strain-resolved de-novo metagenomic assembly of viral genomes and microbial 16S rRNAs

BackgroundMetagenomics is a powerful approach to study environmental and human-associated microbial communities and, in particular, the role of viruses in shaping them. Viral genomes are challenging to assemble from metagenomic samples due to their genomic diversity caused by high mutation rates. In the standard de Bruijn graph assemblers, this genomic diversity leads to complex k-mer assembly graphs with a plethora of loops and bulges that are challenging to resolve into strains or haplotypes because variants more than the k-mer size apart cannot be phased. In contrast, overlap assemblers can phase variants as long as they are covered by a single read. ResultsHere, we present PenguiN, a software for strain resolved assembly of viral DNA and RNA genomes and bacterial 16S rRNA from shotgun metagenomics. Its exhaustive detection of all read overlaps in linear time combined with a Bayesian model to select strain-resolved extensions allow it to assemble longer viral contigs, less fragmented genomes, and more strains than existing assembly tools, on both real and simulated datasets. We show a 3-40-fold increase in complete viral genomes and a 6-fold increase in bacterial 16S rRNA genes. ConclusionPenguiN is the first overlap-based assembler for viral genome and 16S rRNA assembly from large and complex metagenomic datasets, which we hope will facilitate studying the key roles of viruses in microbial communities.

bioinformatics↗

bioRxiv2024

Discovery of Highly Active Kynureninases for Cancer Immunotherapy through Protein Language Model

Overcoming the immunosuppressive tumor microenvironment is a promising strategy in anticancer therapy. L-kynurenine, a strong immunosuppressive metabolite can be degraded through kynureninases. Through homology searches and protein language models, we identified and then experimentally determined the efficacy of four top-ranked kynureninases. The catalytically most active one nearly doubles turnover number over the prior best, reducing tumor weight by 3.42 times in mouse model comparisons, and thus, presenting substantial therapeutic potential.

cancer biology↗

bioRxiv2023

Tertiary-interaction characters enable fast, model-based structural phylogenetics beyond the twilight zone

Protein structure is more conserved than protein sequence, and therefore may be useful for phylogenetic inference beyond the "twilight zone" where sequence similarity is highly decayed. Until recently, structural phylogenetics was constrained by the lack of solved structures for most proteins, and the reliance on phylogenetic distance methods which made it difficult to treat inference and uncertainty statistically. AlphaFold has mostly overcome the first problem by making structural predictions readily available. We address the second problem by redeploying a structural alphabet recently developed for Foldseek, a highly-efficient deep homology search program. For each residue in a structure, Foldseek identifies a tertiary interaction closest-neighbor residue in the structure, and classifies it into one of twenty "3Di" states. We test the hypothesis that 3Dis can be used as standard phylogenetic characters using a dataset of 53 structures from the ferritin-like superfamily. We performed 60 IQtree Maximum Likelihood runs to compare structure-free, PDB, and AlphaFold analyses, and default versus custom model sets that include a 3DI-specific rate matrix. Analyses that combine amino acids, 3Di characters, partitioning, and custom models produce the closest match to the structural distances tree of Malik et al. (2020), avoiding the long-branch attraction errors of structure-free analyses. Analyses include standard ultrafast bootstrapping confidence measures, and take minutes instead of weeks to run on desktop computers. These results suggest that structural phylogenetics could soon be routine practice in protein phylogenetics, allowing the re-exploration of many fundamental phylogenetic problems.

evolutionary biology↗

bioRxiv2023

Learning sequence, structure, and function representations of proteins with language models

The sequence-structure-function relationships that ultimately generate the diversity of extant observed proteins is complex, as proteins bridge the gap between multiple informational and physical scales involved in nearly all cellular processes. One limitation of existing protein annotation databases such as UniProt is that less than 1% of proteins have experimentally verified functions, and computational methods are needed to fill in the missing information. Here, we demonstrate that a multi-aspect framework based on protein language models can learn sequence-structure-function representations of amino acid sequences, and can provide the foundation for sensitive sequence-structure-function aware protein sequence search and annotation. Based on this model, we introduce a multi-aspect information retrieval system for proteins, Protein-Vec, covering sequence, structure, and function aspects, that enables computational protein annotation and function prediction at tree-of-life scales.

bioinformatics↗