Multi-scale structural analysis of proteins by deep semantic segmentation

Chainsaw: protein domain segmentation with fully convolutional neural networks

MOTIVATION

Protein domains are fundamental units of protein structure and play a pivotal role in understanding folding, function, evolution, and design. The advent of accurate structure prediction techniques has resulted in an influx of new structural data, making the partitioning of these structures into domains essential for inferring evolutionary relationships and functional classification.

RESULTS

This article presents Chainsaw, a supervised learning approach to domain parsing that achieves accuracy that surpasses current state-of-the-art methods. Chainsaw uses a fully convolutional neural network which is trained to predict the probability that each pair of residues is in the same domain. Domain predictions are then derived from these pairwise predictions using an algorithm that searches for the most likely assignment of residues to domains given the set of pairwise co-membership probabilities. Chainsaw matches CATH domain annotations in 78% of protein domains versus 72% for the next closest method. When predicting on AlphaFold models, expert human evaluators were twice as likely to prefer Chainsaw's predictions versus the next best method.

AVAILABILITY AND IMPLEMENTATION

github.com/JudeWells/Chainsaw.

Merizo: a rapid and accurate protein domain segmentation method using invariant point attention

The AlphaFold Protein Structure Database, containing predictions for over 200 million proteins, has been met with enthusiasm over its potential in enriching structural biological research and beyond. Currently, access to the database is precluded by an urgent need for tools that allow the efficient traversal, discovery, and documentation of its contents. Identifying domain regions in the database is a non-trivial endeavour and doing so will aid our understanding of protein structure and function, while facilitating drug discovery and comparative genomics. Here, we describe a deep learning method for domain segmentation called Merizo, which learns to cluster residues into domains in a bottom-up manner. Merizo is trained on CATH domains and fine-tuned on AlphaFold2 models via self-distillation, enabling it to be applied to both experimental and AlphaFold2 models. As proof of concept, we apply Merizo to the human proteome, identifying 40,818 putative domains that can be matched to CATH representative domains.

A unified approach to protein domain parsing with inter-residue distance matrix

MOTIVATION

It is fundamental to cut multi-domain proteins into individual domains, for precise domain-based structural and functional studies. In the past, sequence-based and structure-based domain parsing was carried out independently with different methodologies. The recent progress in deep learning-based protein structure prediction provides the opportunity to unify sequence-based and structure-based domain parsing.

RESULTS

Based on the inter-residue distance matrix, which can be either derived from the input structure or predicted by trRosettaX, we can decode the domain boundaries under a unified framework. We name the proposed method UniDoc. The principle of UniDoc is based on the well-accepted physical concept of maximizing intra-domain interaction while minimizing inter-domain interaction. Comprehensive tests on five benchmark datasets indicate that UniDoc outperforms other state-of-the-art methods in terms of both accuracy and speed, for both sequence-based and structure-based domain parsing. The major contribution of UniDoc is providing a unified framework for structure-based and sequence-based domain parsing. We hope that UniDoc would be a convenient tool for protein domain analysis.

AVAILABILITY AND IMPLEMENTATION

https://yanglab.nankai.edu.cn/UniDoc/.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

De Novo Design of a Highly Stable Ovoid TIM Barrel: Unlocking Pocket Shape towards Functional Design

The ability to finely control the structure of protein folds is an important prerequisite to functional protein design. The TIM barrel fold is an important target for these efforts as it is highly enriched for diverse functions in nature. Although a TIM barrel protein has been designed de novo, the ability to finely alter the curvature of the central beta barrel and the overall architecture of the fold remains elusive, limiting its utility for functional design. Here, we report the de novo design of a TIM barrel with ovoid (twofold) symmetry, drawing inspiration from natural beta and TIM barrels with ovoid curvature. We use an autoregressive backbone sampling strategy to implement our hypothesis for elongated barrel curvature, followed by an iterative enrichment sequence design protocol to obtain sequences which yield a high proportion of successfully folding designs. Designed sequences are highly stable and fold to the designed barrel curvature as determined by a 2.1 Å resolution crystal structure. The designs show robustness to drastic mutations, retaining high melting temperatures even when multiple charged residues are buried in the hydrophobic core or when the hydrophobic core is ablated to alanine. As a scaffold with a greater capacity for hosting diverse hydrogen bonding networks and installation of binding pockets or active sites, the ovoid TIM barrel represents a major step towards the de novo design of functional TIM barrels.

Ig-VAE: Generative modeling of protein structure by direct 3D coordinate generation

While deep learning models have seen increasing applications in protein science, few have been implemented for protein backbone generation—an important task in structure-based problems such as active site and interface design. We present a new approach to building class-specific backbones, using a variational auto-encoder to directly generate the 3D coordinates of immunoglobulins. Our model is torsion- and distance-aware, learns a high-resolution embedding of the dataset, and generates novel, high-quality structures compatible with existing design tools. We show that the Ig-VAE can be used with Rosetta to create a computational model of a SARS-CoV2-RBD binder via latent space sampling. We further demonstrate that the model’s generative prior is a powerful tool for guiding computational protein design, motivating a new paradigm under which backbone design is solved as constrained optimization problem in the latent space of a generative model.

Many essential biochemical processes are governed by protein-protein interactions (PPIs), and our ability to make binding proteins that modulate PPIs is crucial to the creation of therapeutics and the study of cell-signaling. One critical aspect of PPI design is to capture protein conformational flexibility. Deep generative models are a class of mathematical models that are able to synthesize novel data from a finite set of training examples. Here, we make advances in computational protein design methodology by developing a deep generative model that creates protein backbones adopting the immunoglobulin fold, which is found in natural binding proteins such as antibodies. While generative models have been powerful in tasks such as image generation, using them to create proteins has remained a challenge. We solve this problem with a new model that allows for the direct generation of novel 3D molecules and show that they are of high chemical accuracy. Generated structures work well with existing protein design methods such as Rosetta, providing access to a large collection of novel immunoglobulin structures. Finally, we present a new protein design framework, called “generative design,” that shows how deep generative models such as ours can be applied to virtually any protein design problem.

Deep learning approaches for conformational flexibility and switching properties in protein design

Following the hugely successful application of deep learning methods to protein structure prediction, an increasing number of design methods seek to leverage generative models to design proteins with improved functionality over native proteins or novel structure and function. The inherent flexibility of proteins, from side-chain motion to larger conformational reshuffling, poses a challenge to design methods, where the ideal approach must consider both the spatial and temporal evolution of proteins in the context of their functional capacity. In this review, we highlight existing methods for protein design before discussing how methods at the forefront of deep learning-based design accommodate flexibility and where the field could evolve in the future.

Structure-based protein design with deep learning

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. In recent years, it has been demonstrated that proteins with new functionality or shapes can be designed via structure-based modeling methods, and the design strategies have combined all available information - but largely piece-by-piece - from sequence derived statistics to the detailed atomic-level modeling of chemical interactions. Despite the significant progress, incorporating data-derived approaches through the use of deep learning methods can be a game changer. In this review, we summarize current progress, compare the arc of developing the deep learning approaches with the conventional methods, and describe the motivation and concepts behind current strategies that may lead to potential future opportunities.

Repeat proteins: designing new shapes and functions for solenoid folds

The modular nature of repeat proteins has inspired the design of regular and completely novel sequences and structures. Research in the past years has provided a broad set of design approaches and new repeat proteins that have found applications in molecular recognition, taking advantage of the natural ability of some of these families to bind proteins, peptides and nucleic acids. Here, we provide an overview on the recent trends in design of repeat proteins, particularly solenoid folds, and their applications. By exploiting the intrinsic modularity of repeats, new architectures have been designed that combine different types of repeat, are easily scalable by changing the number of repeats and can be quickly generated by using existing modular building blocks.

De novo protein design, a retrospective

Proteins are molecular machines whose function depends on their ability to achieve complex folds with precisely defined structural and dynamic properties. The rational design of proteins from first-principles, or de novo, was once considered to be impossible, but today proteins with a variety of folds and functions have been realized. We review the evolution of the field from its earliest days, placing particular emphasis on how this endeavor has illuminated our understanding of the principles underlying the folding and function of natural proteins, and is informing the design of macromolecules with unprecedented structures and properties. An initial set of milestones in de novo protein design focused on the construction of sequences that folded in water and membranes to adopt folded conformations. The first proteins were designed from first-principles using very simple physical models. As computers became more powerful, the use of the rotamer approximation allowed one to discover amino acid sequences that stabilize the desired fold. As the crystallographic database of protein structures expanded in subsequent years, it became possible to construct proteins by assembling short backbone fragments that frequently recur in Nature. The second set of milestones in de novo design involves the discovery of complex functions. Proteins have been designed to bind a variety of metals, porphyrins, and other cofactors. The design of proteins that catalyze hydrolysis and oxygen-dependent reactions has progressed significantly. However, de novo design of catalysts for energetically demanding reactions, or even proteins that bind with high affinity and specificity to highly functionalized complex polar molecules remains an importnant challenge that is now being achieved. Finally, the protein design contributed significantly to our understanding of membrane protein folding and transport of ions across membranes. The area of membrane protein design, or more generally of biomimetic polymers that function in mixed or non-aqueous environments, is now becoming increasingly possible.