The Evolution of Protein Folds by Creative Destruction

Veretnik

et al. 2023

Preprint

Our views of fold space implicitly rest upon many assumptions that impact how we analyze, interpret and understand biological systems—from protein structure comparison and classification to function prediction and evolutionary analyses. For instance, is there an optimal granularity at which to view protein structural similarities (e.g., architecture, topology or some other level)? If so, how does it vary with the type of question being asked? Similarly, the discrete/continuous dichotomy of fold space is central in structural bioinformatics, but remains unresolved. Discrete views of fold space bin ‘similar’ folds into distinct, non-overlapping groups; unfortunately, such binning may inherently miss many remote relationships. While hierarchical databases like CATH, SCOP and ECOD represent major steps forward in protein classification, a scalable, objective and conceptually flexible method, with less reliance on assumptions and heuristics, could enable a more systematic and nuanced exploration of fold space, particularly as regards evolutionarily-distant relationships. Building upon a recent ‘Urfold’ model of protein structure, we have developed a new approach to analyze protein structure relationships. Termed ‘DeepUrfold’, this method is rooted in deep generative modeling via variational Bayesian inference, and we find it to be useful for comparative analysis across the protein universe. Critically, DeepUrfold leverages its deep generative model’s embeddings, which represent a distilled, lower-dimensional space of a given protein and its amalgamation of sequence, structure and biophysical properties. Notably, DeepUrfold is structure-guided, versus being purely structure-based, and its architecture allows each trained model to learn protein features (structural and otherwise) that, in a sense, ‘define’ different superfamilies. Deploying DeepUrfold with CATH suggests a new, mostly-continuous view of fold space—a view that extends beyond simple 3D structural/geometric similarity, towards the realm of integrated sequence↔structure↔function properties. We find that such an approach can quantitatively represent and detect evolutionarily-remote relationships that evade existing methods.

Section: Introductionmentioning

confidence: 99%

Deep Generative Models of Protein Structure: A Method to Uncover Distant Relationships Across a Continuous Fold Space

Veretnik

et al. 2023

Preprint

“…A recent and somewhat unique approach, Geometricus (Durairaj et al ., 2020), creates protein embeddings by taking two parallel approaches to fragmentation: (i) a k -mer based fragmentation runs along the sequence (yielding contiguous segments), while (ii) a radius-based fragmentation uses the method of spatial moment invariants to compute (potentially non-contiguous) geometric ‘fragments’ for each residue position and its neighborhood within a given radius, which are then mapped to ‘shape-mers’. Conceptually, this allowance for discontinuous fragments is a key step in allowing an algorithm to bridge more of fold space, as similarities between such non-contiguous fragments can imply an ancestral contiguous peptide that duplicated and removed its terminal SSE in a process termed ‘creative destruction’, resulting in two different folds with different topologies, but similar architecture (Alvarez-Carreño et al ., 2021, 2022).…”

Section: Introductionmentioning

confidence: 99%

“…Presumably, the protein universe—meaning the set of all proteins (known or unknown, ancestral or extent)—did not spontaneously arise with intact, full-sized domains. Rather, smaller, sub-domain–sized protein fragments likely preceded the modern domains; the genomic elements encoding these primitive fragments were subject to natural evolutionary processes of duplication, mutation and recombination to give rise to extant domains found in contemporary proteins (Alva et al ., 2015; Alvarez-Carreño et al ., 2022; Bromberg et al ., 2022; Kolodny et al ., 2021; Youkharibache, 2019). Our ability to detect common fragments, shared amongst at least two domains, relies on (i) having an accurate similarity metric and (ii) a suitable random/background distribution (i.e.…”

Section: Introductionmentioning

confidence: 99%

“…The transition of a protein sequence from one fold to another, either naturally (via evolution) or artificially (via design/engineering), likely occurs over multiple intermediate steps, for example by combining and/or permuting short secondary structure segments, or mutating single residues (Alvarez-Carreño et al ., 2021, 2022; Grishin, 2001; Kinch and Grishin, 2002; Krishna and Grishin, 2005). Each step may correspond to the same or a different fold than its preceding step.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Generative Models of Protein Structure Uncover Distant Relationships Across a Continuous Fold Space

Veretnik

et al. 2022

Preprint

Unresolved questions about the discrete/continuous dichotomy of protein fold space permeate structural and evolutionary biology. From protein structure comparison and classification to evolutionary analyses and function prediction, our views of fold space implicitly rest upon many assumptions that impact how we analyze, interpret and come to understand biological systems. Discrete views of fold space categorize similar folds into separate groups; unfortunately, such a ‘binning’ process inherently fails to capture many remote relationships. While hierarchical databases such as CATH, SCOP, and ECOD represent major steps forward in protein classification, we believe that a scalable, objective and conceptually flexible method that is less reliant upon assumptions and heuristics could enable a more systematic and thorough exploration of fold space and evolutionary-distant relationships. Here, we develop a structure-guided, comparative analysis of proteins, leveraging embeddings derived from deep generative models, which represent a highly-compressed, lower-dimensional space of a given protein and its sequence, structure and biophysical properties. Building upon a recent ‘Urfold’ model of protein structure, the deep generative approach developed here, termed ‘DeepUrfold’, suggests a new, mostly-continuous view of fold space—a view that extends beyond simple 3D structural/geometric similarity, towards the realm of integrated sequence↔structure↔function properties. We find that such an approach can quantitatively represent and detect evolutionarily-remote relationships that are not captured by existing methods.

“…Despite having permuted strands, the architecturally-identical SBBs are often involved in nucleic acid metabolic pathways, and many of them oligomerize via residue interactions amongst similar edge-strands [8]. Recently, others have proposed that the SH3 and OB share a common ancestor that diverged via a process called 'Creative Destruction' [9,10]. Notably, the SH3 and OB are two of the most ancient and widespread protein folds, and they permeate most information-storage and information-processing pathways in cellular life, from DNA replication to transcription of DNA→RNA and translation of RNA→protein [9,11].…”

Section: Introductionmentioning

confidence: 99%

Explainable Deep Generative Models, Ancestral Fragments, and Murky Regions of the Protein Structure Universe

Bourne

2022

Preprint

Modern proteins did not arise abruptly, as singular events, but rather over the course of at least 3.5 billion years of evolution. Can machine learning teach us how this occurred? The molecular evolutionary processes that yielded the intricate three-dimensional (3D) structures of proteins involve duplication, recombination and mutation of genetic elements, corresponding to short peptide fragments. Identifying and elucidating these ancestral fragments is crucial to deciphering the interrelationships amongst proteins, as well as how evolution acts upon protein sequences, structures & functions. Traditionally, structural fragments have been found using sequence-based and 3D structural alignment approaches, but that becomes challenging when proteins have undergone extensive permutations- allowing two proteins to share a common architecture, though their topologies may drastically differ (a phenomenon termed the Urfold). We have designed a new framework to identify compact, potentially-discontinuous peptide fragments by combining (i) deep generative models of protein superfamilies with (ii) layer-wise relevance propagation (LRP) to identify atoms of great relevance in creating an embedding during an all superfamilies by all domains analysis. Our approach recapitulates known relationships amongst the evolutionarily ancient small beta-barrels (e.g. SH3 and OB folds) and amongst P-loop-containing proteins (e.g. Rossmann and P-loop NTPases), previously established via manual analysis. Because of the generality of our deep model's approach, we anticipate that it can enable the discovery of new ancestral peptides. In a sense, our framework uses LRP as an 'explainable AI' approach, in conjunction with a recent deep generative model of protein structure (termed DeepUrfold), in order to leverage decades worth of structural biology knowledge to decipher the underlying molecular bases for protein structural relationships-including those which are exceedingly remote, yet discoverable via deep learning.