Mesoscale Simulation and Machine Learning of Asphaltene Aggregation Phase Behavior and Molecular Assembly Landscapes

A Study of the Methane Oxidation Mechanism and Reaction Pathways Using Reactive Molecular Simulation and Nonlinear Manifold Learning

Methane, as the primary component of natural gas, is a vital energy resource extensively utilized through oxidation reactions. These reactions yield diverse radicals and molecules via varying intermediate reaction routes, contingent upon the oxidation conditions. In this study, we employ reactive molecular dynamics simulations to investigate the early-stage mechanism of methane oxidation across different temperatures and methane/oxygen conditions. Our analysis reveals distinct variations in species count, initial reaction times, and the spectrum of the main reactions/molecules under diverse conditions. Notably, both full oxidation of methane (FOM) and partial oxidation of methane (POM) are observed in all simulations, with FOM favored under high-temperature and fuel-lean conditions, while POM prevails in low-temperature and fuel-rich environments. Furthermore, we utilize nonlinear manifold learning techniques to extract a 2D manifold from the reaction state space, identifying two collective variables governing the reaction pathways. This research provides a systematic understanding of the initial stage mechanisms of methane oxidation under varying conditions, offering useful insights into chemical science and fuel engineering.

Electric field modulated configuration and orientation of aqueous molecule chains

Understanding how external electric fields (EFs) impact the properties of aqueous molecules is crucial for various applications in chemistry, biology, and engineering. In this paper, we present a study utilizing molecular dynamics simulation to explore how direct-current (DC) and alternative-current (AC) EFs affect hydrophobic (n-triacontane) and hydrophilic (PEG-10) oligomer chains. Through a machine learning approach, we extract a 2-dimensional free energy (FE) landscape of these molecules, revealing that electric fields modulate the FE landscape to favor stretched configurations and enhance the alignment of the chain with the electric field. Our observations indicate that DC EFs have a more prominent impact on modulation compared to AC EFs and that EFs have a stronger effect on hydrophobic chains than on hydrophilic oligomers. We analyze the orientation of water dipole moments and hydrogen bonds, finding that EFs align water molecules and induce more directional hydrogen bond networks, forming 1D water structures. This favors the stretched configuration and alignment of the studied oligomers simultaneously, as it minimizes the disruption of 1D structures. This research deepens our understanding of the mechanisms by which electric fields modulate molecular properties and could guide the broader application of EFs to control other aqueous molecules, such as proteins or biomolecules.

Data-driven prediction of αIIbβ3 integrin activation paths using manifold learning and deep generative modeling

The integrin heterodimer is a transmembrane protein critical for driving cellular process and is a therapeutic target in the treatment of multiple diseases linked to its malfunction. Activation of integrin involves conformational transitions between bent and extended states. Some of the conformations that are intermediate between bent and extended states of the heterodimer have been experimentally characterized, but the full activation pathways remain unresolved both experimentally due to their transient nature and computationally due to the challenges in simulating rare barrier crossing events in these large molecular systems. An understanding of the activation pathways can provide new fundamental understanding of the biophysical processes associated with the dynamic interconversions between bent and extended states and can unveil new putative therapeutic targets. In this work, we apply nonlinear manifold learning to coarse-grained molecular dynamics simulations of bent, extended, and two intermediate states of αIIbβ3 integrin to learn a low-dimensional embedding of the configurational phase space. We then train deep generative models to learn an inverse mapping between the low-dimensional embedding and high-dimensional molecular space and use these models to interpolate the molecular configurations constituting the activation pathways between the experimentally characterized states. This work furnishes plausible predictions of integrin activation pathways and reports a generic and transferable multiscale technique to predict transition pathways for biomolecular systems.

Unsupervised Learning Methods for Molecular Simulation Data

Unsupervised learning is becoming an essential tool to analyze the increasingly large amounts of data produced by atomistic and molecular simulations, in material science, solid state physics, biophysics, and biochemistry. In this Review, we provide a comprehensive overview of the methods of unsupervised learning that have been most commonly used to investigate simulation data and indicate likely directions for further developments in the field. In particular, we discuss feature representation of molecular systems and present state-of-the-art algorithms of dimensionality reduction, density estimation, and clustering, and kinetic models. We divide our discussion into self-contained sections, each discussing a specific method. In each section, we briefly touch upon the mathematical and algorithmic foundations of the method, highlight its strengths and limitations, and describe the specific ways in which it has been used-or can be used-to analyze molecular simulation data.

Tetranucleosome Interactions Drive Chromatin Folding

The multiscale organizational structure of chromatin in eukaryotic cells is instrumental to DNA transcription, replication, and repair. At mesoscopic length scales, nucleosomes pack in a manner that serves to regulate gene expression through condensation and expansion of the genome. The particular structures that arise and their respective thermodynamic stabilities, however, have yet to be fully resolved. In this study, we combine molecular modeling using the 1CPN mesoscale model of chromatin with nonlinear manifold learning to identify and characterize the structure and free energy of metastable states of short chromatin segments comprising between 4- and 16-nucleosomes. Our results reveal the formation of two previously characterized tetranucleosomal conformations, the "α-tetrahedron" and the "β-rhombus", which have been suggested to play an important role in the accessibility of DNA and, respectively, induce local chromatin compaction or elongation. The spontaneous formation of these motifs is potentially responsible for the slow nucleosome dynamics observed in experimental studies. Increases of the nucleosome repeat length are accompanied by more pronounced structural irregularity and flexibility and, ultimately, a dynamic liquid-like behavior that allows for frequent structural reorganization. Our findings indicate that tetranucleosome motifs are intrinsically stable structural states, driven by local internucleosomal interactions, and support a mechanistic picture of chromatin packing, dynamics, and accessibility that is strongly influenced by emergent local mesoscale structure.

The Martini Model in Materials Science

The Martini model, a coarse-grained force field initially developed with biomolecular simulations in mind, has found an increasing number of applications in the field of soft materials science. The model's underlying building block principle does not pose restrictions on its application beyond biomolecular systems. Here, the main applications to date of the Martini model in materials science are highlighted, and a perspective for the future developments in this field is given, particularly in light of recent developments such as the new version of the model, Martini 3.

Discovery of Self-Assembling π-Conjugated Peptides by Active Learning-Directed Coarse-Grained Molecular Simulation

Electronically active organic molecules have demonstrated great promise as novel soft materials for energy harvesting and transport. Self-assembled nanoaggregates formed from π-conjugated oligopeptides composed of an aromatic core flanked by oligopeptide wings offer emergent optoelectronic properties within a water-soluble and biocompatible substrate. Nanoaggregate properties can be controlled by tuning core chemistry and peptide composition, but the sequence-structure-function relations remain poorly characterized. In this work, we employ coarse-grained molecular dynamics simulations within an active learning protocol employing deep representational learning and Bayesian optimization to efficiently identify molecules capable of assembling pseudo-1D nanoaggregates with good stacking of the electronically active π-cores. We consider the DXXX-OPV3-XXXD oligopeptide family, where D is an Asp residue and OPV3 is an oligophenylenevinylene oligomer (1,4-distyrylbenzene), to identify the top performing XXX tripeptides within all 20³ = 8000 possible sequences. By direct simulation of only 2.3% of this space, we identify molecules predicted to exhibit superior assembly relative to those reported in prior work. Spectral clustering of the top candidates reveals new design rules governing assembly. This work establishes new understanding of DXXX-OPV3-XXXD assembly, identifies promising new candidates for experimental testing, and presents a computational design platform that can be generically extended to other peptide-based and peptide-like systems.

Extracting collective motions underlying nucleosome dynamics via nonlinear manifold learning

The identification of effective collective variables remains a challenge in molecular simulations of complex systems. Here, we use a nonlinear manifold learning technique known as the diffusion map to extract key dynamical motions from a complex biomolecular system known as the nucleosome: a DNA-protein complex consisting of a DNA segment wrapped around a disc-shaped group of eight histone proteins. We show that without any a priori information, diffusion maps can identify and extract meaningful collective variables that characterize the motion of the nucleosome complex. We find excellent agreement between the collective variables identified by the diffusion map and those obtained manually using a free energy-based analysis. Notably, diffusion maps are shown to also identify subtle features of nucleosome dynamics that did not appear in those manually specified collective variables. For example, diffusion maps identify the importance of looped conformations in which DNA bulges away from the histone complex that are important for the motion of DNA around the nucleosome. This work demonstrates that diffusion maps can be a promising tool for analyzing very large molecular systems and for identifying their characteristic slow modes.

Aggregation Behavior of Model Asphaltenes Revealed from Large-Scale Coarse-Grained Molecular Simulations

Fully atomistic simulations of models of asphaltenes in simple solvents have allowed the study of trends in aggregation phenomena to understand the underlying role played by molecular structure. The detail included at this scale of molecular modeling is, however, at odds with the required spatial and temporal resolution needed to fully understand asphaltene aggregation. The computational cost required to explore the relevant scales can be reduced by employing coarse-grained (CG) models, which consist of lumping a few atoms into a single segment that is characterized by effective interactions. In this work, CG force fields developed via the statistical associating fluid theory (SAFT-γ) [ Müller , E. A. ; Jackson , G. Annu. Rev. Chem. Biomol. Eng. 5 , 2014 , 405 - 427 ] equation of state (EoS) provide a reliable pathway to link the molecular description with macroscopic thermophysical data. A recent modification of the SAFT-VR EoS [ Müller , E. A. ; Mejía , A. Langmuir 33 , 2017 , 11518 - 11529 ], which allows for the parameterization of homonuclear rings, is selected as the starting point to develop CG models for polycyclic aromatic hydrocarbons. The new aromatic-core models, along with others published for simpler organic molecules, are adopted for the construction of asphaltene models by combining different chemical moieties in a group-contribution fashion. We apply the procedure to two previously reported asphaltene models and perform molecular dynamics simulations to validate the coarse-grained representation against benchmark systems of 27 asphaltenes in a pure solvent (toluene or heptane) described in a fully atomistic fashion. An excellent match between both levels of description is observed for the cluster size, radii of gyration, and relative-shape-anisotropy-factor distributions. We exploit the advantages of the CG representation by simulating systems containing up to 2000 asphaltene molecules in an explicit solvent investigating the effect of asphaltene concentration, solvent composition, and temperature on aggregation. By studying large systems facilitated by the use of CG models, we observe stable continuous distributions of molecular aggregates at conditions away from the two-phase precipitation point. As a further example application, a widely accepted interpretation of cluster-size distributions in asphaltenic systems is challenged by performing system-size tests, reversibility checks, and a time-dependence analysis. The proposed coarse-graining procedure is seen to be general and predictive and, hence, can be applied to other asphaltenic molecular structures.

Recovery of Protein Folding Funnels from Single-Molecule Time Series by Delay Embeddings and Manifold Learning

The stability and folding of proteins is governed by the underlying single-molecule free energy surface (smFES) mapping the free energy of the molecule as a function of configurational state. Ascertaining the smFES is of great value in understanding and engineering protein structure and function. By integrating tools from dynamical systems theory and nonlinear manifold learning, we describe an approach to reconstruct the multidimensional smFES for a protein from a time series in a single experimentally measurable observable. We employ Takens' delay embeddings to project the time series into a high-dimensional space in which the projected dynamics are C¹-equivalent to the true system dynamics and employ diffusion maps to recover a low-dimensional reconstruction of the smFES that is equivalent to the true smFES up to a smooth and invertible transformation. We validate the approach in molecular dynamics simulations of Trp-cage, Villin, and BBA to demonstrate that landscapes recovered from univariate time series in the head-to-tail distance are topologically identical-they precisely preserve the metastable states and folding pathways-and topographically approximate-the free energy barrier heights and well depths are approximately preserved-to the true landscapes determined from complete knowledge of all atomic coordinates. We go on to show that the reconstructed landscapes reliably predict temperature denaturation and identify point mutations and groups of mutations critical to folding. These results demonstrate that protein folding funnels can be reconstructed from experimentally measurable time series and used to understand and engineer folding.

Asphaltene Mesoscale Aggregation Behavior in Organic Solvents-A Brownian Dynamics Study

Significant advances have been achieved in understanding the main molecular mechanisms leading to asphaltene aggregation. However, the existing computational deficiency of molecular dynamics simulations did not allow full reproduction of the complex aggregation behavior of asphaltene in the past. In this work, we use the Brownian dynamics simulation to investigate asphaltene aggregation behavior on larger length and time scales that have not been previously accessed by molecular simulations. This enabled us to completely render the formation of clusters of asphaltene nanoaggregates and the resulting fractal or network of aggregates during the aggregation process. Asphaltene aggregation is studied at several volume fractions (ϕ = 1-7%) of asphaltene nanoaggregates in two solvents including heptane and heptol (i.e., a mixture of heptane and toluene). Our simulation results support the aggregation hierarchy proposed in the Yen-Mullins model (Mullins, Annu. Rev. Anal. Chem. 2011, 4, 393-418.) by demonstrating that asphaltene nanoaggregates form small clusters with an aggregation number of 7-8 and an average gyration radius of ∼4.0 nm capable of forming either fractal aggregates with a fractal dimension of 1.93-2.04 at low ϕ or percolating networks of aggregates at high ϕ. Percolating structures are observed at ϕ = 7% in both solvents. In heptol, the structures mainly percolate along two directions, whereas in heptane, they can percolate along three directions (i.e., x, y, and z). The self-diffusion coefficient ( D) significantly decreases as ϕ increases. Generally, D is larger in heptol than in heptane, but this difference diminishes as ϕ increases, approaching to almost the same value at ϕ = 7%.

Automated crystal characterization with a fast neighborhood graph analysis method

We present a significantly improved, very fast implementation of the Neighborhood Graph Analysis technique for template-free characterization of crystal structures [W. F. Reinhart et al., Soft Matter, 2017, 13, 4733]. By comparing local neighborhoods in terms of their relative graphlet frequencies, we reduce the computational cost by four orders of magnitude compared to the original stochastic method. Furthermore, we present protocols for the detection of topologically important structures and assignment of visually informative colors, providing a fully automated procedure for characterization of crystal structures from particle tracking data. We demonstrate the flexibility of our method on a wide range of crystal structures which have proven difficult to classify by previously available techniques.

Machine learning and data science in soft materials engineering

In many branches of materials science it is now routine to generate data sets of such large size and dimensionality that conventional methods of analysis fail. Paradigms and tools from data science and machine learning can provide scalable approaches to identify and extract trends and patterns within voluminous data sets, perform guided traversals of high-dimensional phase spaces, and furnish data-driven strategies for inverse materials design. This topical review provides an accessible introduction to machine learning tools in the context of soft and biological materials by 'de-jargonizing' data science terminology, presenting a taxonomy of machine learning techniques, and surveying the mathematical underpinnings and software implementations of popular tools, including principal component analysis, independent component analysis, diffusion maps, support vector machines, and relative entropy. We present illustrative examples of machine learning applications in soft matter, including inverse design of self-assembling materials, nonlinear learning of protein folding landscapes, high-throughput antimicrobial peptide design, and data-driven materials design engines. We close with an outlook on the challenges and opportunities for the field.