Sergey Ovchinnikov

Harvard University | Harvard · FAS Center for Systems Biology

De novo design of complex protein folds using solely computational means remains a substantial challenge¹. Here we use a robust deep learning pipeline to design complex folds and soluble analogues of integral membrane proteins. Unique membrane topologies, such as those from G-protein-coupled receptors², are not found in the soluble proteome, and we...

In silico validation of de novo designed proteins with deep learning (DL)‐based structure prediction algorithms has become mainstream. However, formal evidence of the relationship between a high‐quality predicted model and the chance of experimental success is lacking. We used experimentally characterized de novo water‐soluble and transmembrane β‐b...

The exponential growth of protein sequences in the post-genomic era has revolutionized the application of generative sequence models for pivotal tasks such as contact prediction, protein design, alignment, and homology search. Despite remarkable progress in these areas, the interpretability of the modeled pairwise parameters remains limited due to...

Deciphering the relationship between a gene and its genomic context is fundamental to understanding and engineering biological systems. Machine learning has shown promise in learning latent relationships underlying the sequence-structure-function paradigm from massive protein sequence datasets. However, to date, limited attempts have been made in e...

De novo design of complex protein folds using solely computational means remains a significant challenge. Here, we use a robust deep learning pipeline to design complex folds and soluble analogues of integral membrane proteins. Unique membrane topologies, such as those from GPCRs, are not found in the soluble proteome and we demonstrate that their...

Protein language models (pLMs) have emerged as potent tools for predicting and designing protein structure and function, and the degree to which these models fundamentally understand the inherent biophysics of protein structure stands as an open question. Motivated by a discovery that pLM-based structure predictors erroneously predict non-physical...

Since its public release in 2021, AlphaFold2 (AF2) has made investigating biological questions, using predicted protein structures of single monomers or full complexes, a common practice. ColabFold-AF2 is an open-source Jupyter Notebook inside Google Colaboratory and a command-line tool, which makes it easy to use AF2, while exposing its advanced o...

AlphaFold2 (AF2) 1 has revolutionized structural biology by accurately predicting single structures of proteins. However, a protein's biological function often depends on multiple conformational substates2, and disease-causing point mutations often cause population changes within these substates3,4. We demonstrate that clustering a multiple sequenc...

Metagenomes encode an enormous diversity of proteins, reflecting a multiplicity of functions and activities 1,2 . Exploration of this vast sequence space has been limited to a comparative analysis against reference microbial genomes and protein families derived from those genomes. Here, to examine the scale of yet untapped functional diversity beyo...

The Novel Metagenome Protein Families Database (NMPFamsDB) is a database of metagenome- and metatranscriptome-derived protein families, whose members have no hits to proteins of reference genomes or Pfam domains. Each protein family is accompanied by multiple sequence alignments, Hidden Markov Models, taxonomic information, ecosystem and geolocatio...

Inferring the most probable evolutionary tree given leaf nodes is an important problem in computational biology that reveals the evolutionary relationships between species. Due to the exponential growth of possible tree topologies, finding the best tree in polynomial time becomes computationally infeasible. In this work, we propose a novel differen...

Advances in DNA sequencing and machine learning are providing insights into protein sequences and structures on an enormous scale¹. However, the energetics driving folding are invisible in these structures and remain largely unknown². The hidden thermodynamics of folding can drive disease3,4, shape protein evolution5–7 and guide protein engineering...

There has been considerable recent progress in designing new proteins using deep learning methods1-9. Despite this progress, a general deep learning framework for protein design that enables solution of a wide range of design challenges, including de novo binder design and design of higher order symmetric architectures, has yet to be described. Dif...

Prediction of protein complex structures and interfaces potentially has wide applications and can benefit the study of biological mechanisms involving protein-protein interactions. However, the surface prediction accuracy of traditional docking methods and AlphaFold-Multimer is limited. Here we present ColabDock, a framework that makes use of Colab...

Protein structures are essential to understand cellular processes in molecular detail. While advances in AI revealed the tertiary structure of proteins at scale, their quaternary structure remains mostly unknown. Here, we describe a scalable strategy based on AlphaFold2 to predict homo-oligomeric assemblies across four proteomes spanning the tree o...

In silico validation of de novo designed proteins with deep learning (DL)-based structure prediction algorithms has become a mainstream practice. However, formal evidence of the relationship between a high-quality predicted model and the chances of experimental success is lacking. We used experimentally characterized de novo designs to show that Al...

De novo design of protein folds with complex topologies and intricate structural features using solely computational means remains a significant challenge. Here, we use a robust deep learning pipeline to design soluble analogues of integral membrane proteins. Some unique protein topologies, such as the GPCR fold, are not found in the soluble proteo...

Deciphering the relationship between a gene and its genomic context is fundamental to understanding and engineering biological systems. Machine learning has shown promise in learning latent relationships underlying the sequence-structure-function paradigm from massive protein sequence datasets. However, to date, limited attempts have been made in e...

Small beta barrel proteins are attractive targets for computational design because of their considerable functional diversity despite their very small size (<70 amino acids). However, there are considerable challenges to designing such structures, and there has been little success thus far. Because of the small size, the hydrophobic core stabilizin...

Metagenomics has enabled accessing the genetic repertoire of natural microbial communities. Metagenome shotgun sequencing has become the method of choice for studying and classifying microorganisms from various environments. To this end, several methods have been developed to process and analyze the sequence data from raw reads to end-products such...

Deep learning networks offer considerable opportunities for accurate structure prediction and design of biomolecules. While cyclic peptides have gained significant traction as a therapeutic modality, developing deep learning methods for designing such peptides has been slow, mostly due to the small number of available structures for molecules in th...

Deep learning techniques are being used to design new proteins by creating target backbone geometries and finding sequences that can fold into those shapes. While methods like ProteinMPNN provide an efficient algorithm for generating sequences for a given protein backbone, there is still room for improving the scope and computational efficiency of...

Metal ions have various important biological roles in proteins, including structural maintenance, molecular recognition and catalysis. Previous methods of predicting metal-binding sites in proteomes were based on either sequence or structural motifs. Here we developed a co-evolution-based pipeline named ‘MetalNetʼ to systematically predict metal-bi...

Cyclic GMP–AMP synthase (cGAS) is a pattern recognition receptor critical for the innate immune response to intracellular pathogens, DNA damage, tumorigenesis and senescence. Binding to double-stranded DNA (dsDNA) induces conformational changes in cGAS that activate the enzyme to produce 2′-3′ cyclic GMP–AMP (cGAMP), a second messenger that initiat...

Learning the design patterns of proteins from sequences across evolution may have promise toward generative protein design. However it is unknown whether language models, trained on sequences of natural proteins, will be capable of more than memorization of existing protein families. Here we show that language models generalize beyond natural prote...

Advances in DNA sequencing and machine learning are illuminating protein sequences and structures on an enormous scale. However, the energetics driving folding are invisible in these structures and remain largely unknown. The hidden thermodynamics of folding can drive disease, shape protein evolution, and guide protein engineering, and new approach...

The problem of predicting a protein’s 3D structure from its primary amino acid sequence is a longstanding challenge in structural biology. Recently, approaches like alphafold have achieved remarkable performance on this task by combining deep learning techniques with coevolutionary data from multiple sequence alignments of related protein sequences...

Motivation Multiple sequence alignments (MSAs) of homologous sequences contain information on structural and functional constraints and their evolutionary histories. Despite their importance for many downstream tasks, such as structure prediction, MSA generation is often treated as a separate pre-processing step, without any guidance from the appli...

Most proteins fold into 3D structures that determine how they function and orchestrate the biological processes of the cell. Recent developments in computational methods for protein structure predictions have reached the accuracy of experimentally determined models. Although this has been independently verified, the implementation of these methods...

AlphaFold2 (AF2) has revolutionized structural biology by accurately predicting single structures of proteins and protein-protein complexes. However, biological function is rooted in a protein’s ability to sample different conformational substates, and disease-causing point mutations are often due to population changes of these substates. This has...

The binding and catalytic functions of proteins are generally mediated by a small number of functional residues held in place by the overall protein structure. Here, we describe deep learning approaches for scaffolding such functional sites without needing to prespecify the fold or secondary structure of the scaffold. The first approach, “constrain...

ColabFold offers accelerated prediction of protein structures and complexes by combining the fast homology search of MMseqs2 with AlphaFold2 or RoseTTAFold. ColabFold’s 40−60-fold faster search and optimized model utilization enables prediction of close to 1,000 structures per day on a server with one graphics processing unit. Coupled with Google C...

Recent methods have shown promise in using pairwise sequence coevolution predictions to illuminate physical interactions and functional relationships between pairs of protein residues. As a result, there has been an increased interest in identifying higher-order correlations between sequence positions in an effort to further understand how the mult...

A bstract The problem of predicting a protein’s 3D structure from its primary amino acid sequence is a longstanding challenge in structural biology. Recently, approaches like AlphaFold have achieved remarkable performance on this task by combining deep learning techniques with coevolutionary data from multiple sequence alignments of related protein...

The rise in the number of protein sequences in the post-genomic era has led to a major breakthrough in fitting generative sequence models for contact prediction, protein design, alignment, and homology search. Despite this success, the interpretability of the modeled pairwise parameters continues to be limited due to the entanglement of coevolution...

The established approach to unsupervised protein contact prediction estimates coevolving positions using undirected graphical models. This approach trains a Potts model on a Multiple Sequence Alignment. Increasingly large Transformers are being pretrained on unlabeled, unaligned protein sequence databases and showing competitive performance on prot...

There has been considerable recent progress in protein structure prediction using deep neural networks to predict inter-residue distances from amino acid sequences1,2,3. Here we investigate whether the information captured by such networks is sufficiently rich to generate new folded proteins with sequences unrelated to those of the naturally occurr...

Current approaches to de novo design of proteins harboring a desired binding or catalytic motif require pre-specification of an overall fold or secondary structure composition, and hence considerable trial and error can be required to identify protein structures capable of scaffolding an arbitrary functional site. Here we describe two complementary...

Deep learning for protein interactions The use of deep learning has revolutionized the field of protein modeling. Humphreys et al . combined this approach with proteome-wide, coevolution-guided protein interaction identification to conduct a large-scale screen of protein-protein interactions in yeast (see the Perspective by Pereira and Schwede). Th...

ColabFold offers accelerated protein structure and complex predictions by combining the fast homology search of MMseqs2 with AlphaFold2 or RoseTTAFold. ColabFold's 20-30x faster search and optimized model use allows predicting thousands of proteins per day on a server with one GPU. Coupled with Google Colaboratory, ColabFold becomes a free and acce...

Multiple Sequence Alignments (MSAs) of homologous sequences contain information on structural and functional constraints and their evolutionary histories. Despite their importance for many downstream tasks, such as structure prediction, MSA generation is often treated as a separate pre-processing step, without any guidance from the application it w...

Protein-protein interactions play critical roles in biology, but despite decades of effort, the structures of many eukaryotic protein complexes are unknown, and there are likely many interactions that have not yet been identified. Here, we take advantage of recent advances in proteome-wide amino acid coevolution analysis and deep-learning-based str...

Most proteins fold into 3D structures that determine how they function and orchestrate the biological processes of the cell. Recent developments in computational methods have led to protein structure predictions that have reached the accuracy of experimentally determined models. While this has been independently verified, the implementation of thes...

Since the first revelation of proteins functioning as macromolecular machines through their three dimensional structures, researchers have been intrigued by the marvelous ways the biochemical processes are carried out by proteins. The aspiration to understand protein structures has fueled extensive efforts across different scientific disciplines. I...

ColabFold is an easy-to-use Notebook based environment for fast and convenient protein structure predictions. Its structure prediction is powered by AlphaFold2 and RoseTTAFold combined with a fast multiple sequence alignment generation stage using MMseqs2. MMseqs2’s MSAs produce more accurate predictions while being ~16 faster compared to the Alpha...

Deep learning takes on protein folding In 1972, Anfinsen won a Nobel prize for demonstrating a connection between a protein’s amino acid sequence and its three-dimensional structure. Since 1994, scientists have competed in the biannual Critical Assessment of Structure Prediction (CASP) protein-folding challenge. Deep learning methods took center st...

DeepMind presented remarkably accurate protein structure predictions at the CASP14 conference. We explored network architectures incorporating related ideas and obtained the best performance with a 3-track network in which information at the 1D sequence level, the 2D distance map level, and the 3D coordinate level is successively transformed and in...

Significance Almost all proteins fold to their lowest free energy state, which is determined by their amino acid sequence. Computational protein design has primarily focused on finding sequences that have very low energy in the target designed structure. However, what is most relevant during folding is not the absolute energy of the folded state bu...

A bstract The established approach to unsupervised protein contact prediction estimates co-evolving positions using undirected graphical models. This approach trains a Potts model on a Multiple Sequence Alignment, then predicts that the edges with highest weight correspond to contacts in the 3D structure. On the other hand, increasingly large Trans...

A bstract Unsupervised contact prediction is central to uncovering physical, structural, and functional constraints for protein structure determination and design. For decades, the predominant approach has been to infer evolutionary constraints from a set of related sequences. In the past year, protein language models have emerged as a potential al...

If disentangled properly, patterns distilled from evolutionarily related sequences of a given protein family can inform their traits - such as their structure and function. Recent years have seen an increase in the complexity of generative models towards capturing these patterns; from sitewise to pairwise to deep and variational. In this study we e...

An outstanding challenge in protein design is the design of binders against therapeutically relevant target proteins via scaffolding the discontinuous binding interfaces present in their often large and complex binding partners. There is currently no method for sampling through the almost unlimited number of possible protein structures for those ca...

Biofilms are accumulations of microorganisms embedded in extracellular matrices that protect against external factors and stressful environments. Cyanobacterial biofilms are ubiquitous and have potential for treatment of wastewater and sustainable production of biofuels. But the underlying mechanisms regulating cyanobacterial biofilm formation are...

Nucleosomes package genomic DNA into chromatin. By regulating DNA access for transcription, replication, DNA repair, and epigenetic modification, chromatin forms the nexus of most nuclear processes. In addition, dynamic organization of chromatin underlies both regulation of gene expression and evolution of chromosomes into individualized sister obj...

The protein design problem is to identify an amino acid sequence which folds to a desired structure. Given Anfinsen’s thermodynamic hypothesis of folding, this can be recast as finding an amino acid sequence for which the lowest energy conformation is that structure. As this calculation involves not only all possible amino acid sequences but also a...

There has been considerable recent progress in protein structure prediction using deep neural networks to infer distance constraints from amino acid residue co-evolution. We investigated whether the information captured by such networks is sufficiently rich to generate new folded proteins with sequences unrelated to those of the naturally occurring...

Bacteria and archaea possessing the hgcAB gene pair methylate inorganic mercury (Hg) to form highly toxic methylmercury. HgcA consists of a corrinoid binding domain and a transmembrane domain, and HgcB is a dicluster ferredoxin. However, their detailed structure and function have not been thoroughly characterized. We modeled the HgcAB complex by co...

The Rosetta software for macromolecular modeling, docking and design is extensively used in laboratories worldwide. During two decades of development by a community of laboratories at more than 60 institutions, Rosetta has been continuously refactored and extended. Its advantages are its performance and interoperability between broad modeling capab...

A close-up view of the retrotranslocon Misfolded endoplasmic reticulum (ER) proteins are retrotranslocated into the cytosol, polyubiquitinated, and degraded by the proteasome in a process known as ER-associated protein degradation (ERAD). ERAD of misfolded luminal ER proteins (ERAD-L) is mediated by the Hrd1 complex, composed of the ubiquitin ligas...

The prediction of interresidue contacts and distances from coevolutionary data using deep learning has considerably advanced protein structure prediction. Here, we build on these advances by developing a deep residual network for predicting interresidue orientations, in addition to distances, and a Rosetta-constrained energy-minimization protocol f...

The prediction of inter-residue contacts and distances from co-evolutionary data using deep learning has considerably advanced protein structure prediction. Here we build on these advances by developing a deep residual network for predicting inter-residue orientations in addition to distances, and a Rosetta constrained energy minimization protocol...

As a participant in the joint CASP13‐CAPRI46 assessment, the ClusPro server debuted its new template‐based modeling functionality. The addition of this feature, called ClusPro TBM, was motivated by the previous CASP‐CAPRI assessments and by the proven ability of template‐based methods to produce higher quality models, provided templates are availab...

Predicting protein pairs Biological function is driven by interaction between proteins. High-throughput experimental techniques have provided large datasets of protein interactions in several organisms; however, much combinatorial space remains uncharted. Cong et al. predict protein interfaces by identifying coevolving residues in aligned protein s...

Revealing the functional sites of biological sequences, such as evolutionary conserved, structurally interacting or co-evolving protein sites, is a fundamental, and yet challenging task. Different frameworks and models were developed to approach this challenge, including Position-Specific Scoring Matrices, Markov Random Fields, Multivariate Gaussia...

Functional manipulation of biosynthetic enzymes such as cytochrome P450s (or P450s) has attracted great interest in metabolic engineering of plant natural products. Cucurbitacins and mogrosides are plant triterpenoids that share the same backbone but display contrasting bioactivities. This structural and functional diversity of the two metabolites...

Targeted antibody blocking enables characterization of binding sites on immunoglobulin G (IgG), and can efficiently eliminate harmful antibodies from organisms. In this report, we present a novel peptide—denoted dual-functional conjugate of antigenic peptide and Fc-III mimetics (DCAF)—for targeted blocking of antibodies. Synthesis of DCAF was achie...

Every two years groups worldwide participate in the Critical Assessment of Protein Structure Prediction (CASP) experiment to blindly test the strengths and weaknesses of their computational methods. CASP has significantly advanced the field but many hurdles still remain, which may require new ideas and collaborations. In 2012 a web-based effort cal...

One major challenge to delimiting species with genetic data is successfully differentiating species divergences from population structure, with some current methods biased towards overestimating species numbers. Many fields of science are now utilizing machine learning (ML) approaches, and in systematics and evolutionary biology, supervised ML algo...

The regular arrangements of β-strands around a central axis in β-barrels and of α-helices in coiled coils contrast with the irregular tertiary structures of most globular proteins, and have fascinated structural biologists since they were first discovered. Simple parametric models have been used to design a wide range of α-helical coiled-coil struc...

Significance Protein structure refinement by direct global energy optimization has been a longstanding challenge in computational structural biology due to limitations in both energy function accuracy and conformational sampling. This manuscript demonstrates that with recent advances in both areas, refinement can significantly improve protein compa...

Although many putative heme transporters have been discovered, it has been challenging to prove that these proteins are directly involved with heme trafficking in vivo and to identify their heme binding domains. The prokaryotic pathways for cytochrome c biogenesis, Systems I and II, transport heme from inside the cell to outside for stereochemical...

We describe several notable aspects of our structure predictions using Rosetta in CASP12 in the free modeling (FM) and refinement (TR) categories. First, we had previously generated (and published) models for most large protein families lacking experimentally determined structures using Rosetta guided by co-evolution based contact predictions, and...

Many naturally occurring protein systems function primarily as symmetric assemblies. Prediction of the quaternary structure of these assemblies is an important biological problem. This manuscript describes automated tools we have developed for predicting the structure of symmetric protein assemblies in the Robetta structure prediction server. We as...

Significance Coevolution-derived contact predictions are enabling accurate protein structure modeling. However, coevolving residues are not always in contact, and this is a potential source of error in such modeling efforts. To investigate the sources of such errors and, more generally, the origins of coevolution in protein structures, we provide a...

Misfolded endoplasmic reticulum (ER) proteins are retro-translocated through the membrane into the cytosol, where they are poly-ubiquitinated, extracted from the ER membrane, and degraded by the proteasome(1-4), a pathway termed ER-associated protein degradation (ERAD). Proteins with misfolded domains in the ER lumen or membrane are discarded throu...

Evolutionary pressure on residue interactions, intramolecular or intermolecular, that are important for protein structure or function can lead to covariance between the two positions. Recent methodological advances allow much more accurate contact predictions to be derived from this evolutionary covariance signal. The practical application of conta...

How phospholipids are trafficked between the bacterial inner and outer membranes through the hydrophilic space of the periplasm is not known. We report that members of the mammalian cell entry (MCE) protein family form hexameric assemblies with a central channel capable of mediating lipid transport. The E. coli MCE protein, MlaD, forms a ring assoc...

Improving an enzyme's initially low catalytic efficiency with a new target substrate by an order of magnitude or two may require only a few rounds of mutagenesis and screening or selection. However, subsequent rounds of optimization tend to yield decreasing degrees of improvement (diminishing returns) eventually leading to an optimization plateau....

Filling in the protein fold picture Fewer than a third of the 14,849 known protein families have at least one member with an experimentally determined structure. This leaves more than 5000 protein families with no structural information. Protein modeling using residue-residue contacts inferred from evolutionary data has been successful in modeling...

The nearly 200,000 fatalities following exposure to organophosphorus (OP) pesticides each year and the omnipresent danger of a terroristic attack with OP nerve agents emphasize the demand for the development of effective OP antidotes. Standard treatments for intoxicated patients with a combination of atropine and an oxime are limited in their effic...