SidechainNet: An all-atom protein structure dataset for machine learning

Deep Learning of Proteins with Local and Global Regions of Disorder

Although machine learning has transformed protein structure prediction of folded protein ground states with remarkable accuracy, intrinsically disordered proteins and regions (IDPs/IDRs) are defined by diverse and dynamical structural ensembles that are predicted with low confidence by algorithms such as AlphaFold. We present a new machine learning method, IDPForge (Intrinsically Disordered Protein, FOlded and disordered Region GEnerator), that exploits a transformer protein language diffusion model to create all-atom IDP ensembles and IDR disordered ensembles that maintains the folded domains. IDPForge does not require sequence-specific training, back transformations from coarse-grained representations, nor ensemble reweighting, as in general the created IDP/IDR conformational ensembles show good agreement with solution experimental data, and options for biasing with experimental restraints are provided if desired. We envision that IDPForge with these diverse capabilities will facilitate integrative and structural studies for proteins that contain intrinsic disorder.

Designing single-polymer-chain nanoparticles to mimic biomolecular hydration frustration

Native folded proteins rely on sculpting the local chemical environment of their active or binding sites, as well as their shapes, to achieve functionality. In particular, proteins use hydration frustration-control over the dehydration of hydrophilic residues and the hydration of hydrophobic residues-to amplify their chemical or binding activity. Here we uncover that single-polymer-chain nanoparticles formed by random heteropolymers comprising four or more components can display similar levels of hydration frustration. We categorize these nanoparticles into three types based on whether either hydrophobic or hydrophilic residues, or both types, display frustrated states. We propose a series of physicochemical rules that determine the state of these nanoparticles. We demonstrate the generality of these rules in atomistic and simplified Monte Carlo models of single-polymer-chain nanoparticles with different backbones and residues. Our work provides insights into the design of single-chain nanoparticles, an emerging polymer modality that achieves the ease and cost of fabrication of polymeric material with the functionality of biological proteins.

FlowBack: A Generalized Flow-Matching Approach for Biomolecular Backmapping

Coarse-grained models have become ubiquitous in biomolecular modeling tasks aimed at studying slow dynamical processes such as protein folding and DNA hybridization. These models can considerably accelerate sampling but it remains challenging to accurately and efficiently restore all-atom detail to the coarse-grained trajectory, which can be vital for detailed understanding of molecular mechanisms and calculation of observables contingent on all-atom coordinates. In this work, we introduce FlowBack as a deep generative model employing a flow-matching objective to map samples from a coarse-grained prior distribution to an all-atom data distribution. We construct our prior distribution to be agnostic to the coarse-grained map and molecular type. A protein-specific model trained on ∼65k structures from the Protein Data Bank achieves state-of-the-art performance on structural metrics compared to previous generative and rules-based approaches in applications to static PDB structures, all-atom simulations of fast-folding proteins, and coarse-grained trajectories generated by a machine-learned force field. A DNA-protein model trained on ∼1.5k DNA-protein complexes achieves excellent reconstruction and generative capabilities on static DNA-protein complexes from the Protein Data Bank as well as on out-of-distribution coarse-grained dynamical simulations of DNA-protein complexation. FlowBack offers an accurate, efficient, and easy-to-use tool to recover all-atom structures from coarse-grained molecular simulations with higher robustness and fewer steric clashes than previous approaches. We make FlowBack freely available to the community as an open source Python package.

Interpreting forces as deep learning gradients improves quality of predicted protein structures

Protein structure predictions from deep learning models like AlphaFold2, despite their remarkable accuracy, are likely insufficient for direct use in downstream tasks like molecular docking. The functionality of such models could be improved with a combination of increased accuracy and physical intuition. We propose a new method to train deep learning protein structure prediction models using molecular dynamics force fields to work toward these goals. Our custom PyTorch loss function, OpenMM-Loss, represents the potential energy of a predicted structure. OpenMM-Loss can be applied to any all-atom representation of a protein structure capable of mapping into our software package, SidechainNet. We demonstrate our method's efficacy by finetuning OpenFold. We show that subsequently predicted protein structures, both before and after a relaxation procedure, exhibit comparable accuracy while displaying lower potential energy and improved structural quality as assessed by MolProbity metrics.

ThermoFinder: A sequence-based thermophilic proteins prediction framework

Thermophilic proteins are important for academic research and industrial processes, and various computational methods have been developed to identify and screen them. However, their performance has been limited due to the lack of high-quality labeled data and efficient models for representing protein. Here, we proposed a novel sequence-based thermophilic proteins prediction framework, called ThermoFinder. The results demonstrated that ThermoFinder outperforms previous state-of-the-art tools on two benchmark datasets, and feature ablation experiments confirmed the effectiveness of our approach. Additionally, ThermoFinder exhibited exceptional performance and consistency across two newly constructed datasets, one of these was specifically constructed for the regression-based prediction of temperature optimum values directly derived from protein sequences. The feature importance analysis, using shapley additive explanations, further validated the advantages of ThermoFinder. We believe that ThermoFinder will be a valuable and comprehensive framework for predicting thermophilic proteins, and we have made our model open source and available on Github at https://github.com/Luo-SynBioLab/ThermoFinder.

Structure-Infused Protein Language Models

Embeddings from protein language models (PLM's) capture intricate patterns for protein sequences, enabling more accurate and efficient prediction of protein properties. Incorporating protein structure information as direct input into PLMs results in an improvement on the predictive ability of protein embeddings on downstream tasks. In this work we demonstrate that indirectly infusing structure information into PLMs also leads to performance gains on structure related tasks. The key difference between this framework and others is that at inference time the model does not require access to structure to produce its embeddings.

DiAMoNDBack: Diffusion-Denoising Autoregressive Model for Non-Deterministic Backmapping of Cα Protein Traces

Coarse-grained molecular models of proteins permit access to length and time scales unattainable by all-atom models and the simulation of processes that occur on long time scales, such as aggregation and folding. The reduced resolution realizes computational accelerations, but an atomistic representation can be vital for a complete understanding of mechanistic details. Backmapping is the process of restoring all-atom resolution to coarse-grained molecular models. In this work, we report DiAMoNDBack (Diffusion-denoising Autoregressive Model for Non-Deterministic Backmapping) as an autoregressive denoising diffusion probability model to restore all-atom details to coarse-grained protein representations retaining only Cα coordinates. The autoregressive generation process proceeds from the protein N-terminus to C-terminus in a residue-by-residue fashion conditioned on the Cα trace and previously backmapped backbone and side-chain atoms within the local neighborhood. The local and autoregressive nature of our model makes it transferable between proteins. The stochastic nature of the denoising diffusion process means that the model generates a realistic ensemble of backbone and side-chain all-atom configurations consistent with the coarse-grained Cα trace. We train DiAMoNDBack over 65k+ structures from the Protein Data Bank (PDB) and validate it in applications to a hold-out PDB test set, intrinsically disordered protein structures from the Protein Ensemble Database (PED), molecular dynamics simulations of fast-folding mini-proteins from DE Shaw Research, and coarse-grained simulation data. We achieve state-of-the-art reconstruction performance in terms of correct bond formation, avoidance of side-chain clashes, and the diversity of the generated side-chain configurational states. We make the DiAMoNDBack model publicly available as a free and open-source Python package.

Learning Correlations between Internal Coordinates to Improve 3D Cartesian Coordinates for Proteins

We consider a generic representation problem of internal coordinates (bond lengths, valence angles, and dihedral angles) and their transformation to 3-dimensional Cartesian coordinates of a biomolecule. We show that the internal-to-Cartesian process relies on correctly predicting chemically subtle correlations among the internal coordinates themselves, and learning these correlations increases the fidelity of the Cartesian representation. We developed a machine learning algorithm, Int2Cart, to predict bond lengths and bond angles from backbone torsion angles and residue types of a protein, which allows reconstruction of protein structures better than using fixed bond lengths and bond angles or a static library method that relies on backbone torsion angles and residue types in a local environment. The method is able to be used for structure validation, as we show that the agreement between Int2Cart-predicted bond geometries and those from an AlphaFold 2 model can be used to estimate model quality. Additionally, by using Int2Cart to reconstruct an IDP ensemble, we are able to decrease the clash rate during modeling. The Int2Cart algorithm has been implemented as a publicly accessible python package at https://github.com/THGLab/int2cart.

Learning to evolve structural ensembles of unfolded and disordered proteins using experimental solution data

The structural characterization of proteins with a disorder requires a computational approach backed by experiments to model their diverse and dynamic structural ensembles. The selection of conformational ensembles consistent with solution experiments of disordered proteins highly depends on the initial pool of conformers, with currently available tools limited by conformational sampling. We have developed a Generative Recurrent Neural Network (GRNN) that uses supervised learning to bias the probability distributions of torsions to take advantage of experimental data types such as nuclear magnetic resonance J-couplings, nuclear Overhauser effects, and paramagnetic resonance enhancements. We show that updating the generative model parameters according to the reward feedback on the basis of the agreement between experimental data and probabilistic selection of torsions from learned distributions provides an alternative to existing approaches that simply reweight conformers of a static structural pool for disordered proteins. Instead, the biased GRNN, DynamICE, learns to physically change the conformations of the underlying pool of the disordered protein to those that better agree with experiments.

Protein Function Analysis through Machine Learning

Machine learning (ML) has been an important arsenal in computational biology used to elucidate protein function for decades. With the recent burgeoning of novel ML methods and applications, new ML approaches have been incorporated into many areas of computational biology dealing with protein function. We examine how ML has been integrated into a wide range of computational models to improve prediction accuracy and gain a better understanding of protein function. The applications discussed are protein structure prediction, protein engineering using sequence modifications to achieve stability and druggability characteristics, molecular docking in terms of protein-ligand binding, including allosteric effects, protein-protein interactions and protein-centric drug discovery. To quantify the mechanisms underlying protein function, a holistic approach that takes structure, flexibility, stability, and dynamics into account is required, as these aspects become inseparable through their interdependence. Another key component of protein function is conformational dynamics, which often manifest as protein kinetics. Computational methods that use ML to generate representative conformational ensembles and quantify differences in conformational ensembles important for function are included in this review. Future opportunities are highlighted for each of these topics.

Facilitating Machine Learning-Guided Protein Engineering with Smart Library Design and Massively Parallel Assays

Protein design plays an important role in recent medical advances from antibody therapy to vaccine design. Typically, exhaustive mutational screens or directed evolution experiments are used for the identification of the best design or for improvements to the wild-type variant. Even with a high-throughput screening on pooled libraries and Next-Generation Sequencing to boost the scale of read-outs, surveying all the variants with combinatorial mutations for their empirical fitness scores is still of magnitudes beyond the capacity of existing experimental settings. To tackle this challenge, in-silico approaches using machine learning to predict the fitness of novel variants based on a subset of empirical measurements are now employed. These machine learning models turn out to be useful in many cases, with the premise that the experimentally determined fitness scores and the amino-acid descriptors of the models are informative. The machine learning models can guide the search for the highest fitness variants, resolve complex epistatic relationships, and highlight bio-physical rules for protein folding. Using machine learning-guided approaches, researchers can build more focused libraries, thus relieving themselves from labor-intensive screens and fast-tracking the optimization process. Here, we describe the current advances in massive-scale variant screens, and how machine learning and mutagenesis strategies can be integrated to accelerate protein engineering. More specifically, we examine strategies to make screens more economical, informative, and effective in discovery of useful variants.