Biomolecular Modeling Research Articles

BEMM-GEN: A Toolkit for Generating a Biomolecular Environment-Mimicking Model for Molecular Dynamics Simulation.

Understanding the influence of the cellular environment on protein conformations is crucial for elucidating protein functions within living cells. In studies using molecular dynamics (MD) simulation, carbon nanotubes and hydrophobic cages have been widely used to emulate the cellular environment inside specific large biomolecules such as ribosome tunnels and chaperones. However, recent studies suggest that these uniform hydrophobic models may not adequately capture the environmental effects inside each biomolecule. Based on these facts, it is necessary to generate spherical and cylindrical models with varied chemical properties corresponding to the components within target biomolecules. We developed a biomolecular environment-mimicking model generator (BEMM-GEN) that generates spherical and cylindrical models with user-specified chemical properties and allows the integration of arbitrary protein conformations into the generated models. BEMM-GEN provides model and protein complex structures, along with the corresponding parameter files for MD simulation (AMBER and GROMACS), and users immediately run their MD simulation based on the generated input files. BEMM-GEN can be freely downloaded and installed via a Python package manager (pip install BEMM-gen). The source code files and a user manual for operation are provided on GitHub (https://github.com/y4suda/BEMM-GEN).

Sensing the structural and conformational properties of single-stranded nucleic acids using electrometry and molecular simulations

Inferring the 3D structure and conformation of disordered biomolecules, e.g., single stranded nucleic acids (ssNAs), remains challenging due to their conformational heterogeneity in solution. Here, we use escape-time electrometry (ETe) to measure with sub elementary-charge precision the effective electrical charge in solution of short to medium chain length ssNAs in the range of 5–60 bases. We compare measurements of molecular effective charge with theoretically calculated values for simulated molecular conformations obtained from Molecular Dynamics simulations using a variety of forcefield descriptions. We demonstrate that the measured effective charge captures subtle differences in molecular structure in various nucleic acid homopolymers of identical length, and also that the experimental measurements can find agreement with computed values derived from coarse-grained molecular structure descriptions such as oxDNA, as well next generation ssNA force fields. We further show that comparing the measured effective charge with calculations for a rigid, charged rod—the simplest model of a nucleic acid—yields estimates of molecular structural dimensions such as linear charge spacings that capture molecular structural trends observed using high resolution structural analysis methods such as X-ray scattering. By sensitively probing the effective charge of a molecule, electrometry provides a powerful dimension supporting inferences of molecular structural and conformational properties, as well as the validation of biomolecular structural models. The overall approach holds promise for a high throughput, microscopy-based biomolecular analytical approach offering rapid screening and inference of molecular 3D conformation, and operating at the single molecule level in solution.

Kemeny Constant-Based Optimization of Network Clustering Using Graph Neural Networks.

The recent trend in using network and graph structures to represent a variety of different data types has renewed interest in the graph partitioning (GP) problem. This interest stems from the need for general methods that can both efficiently identify network communities and reduce the dimensionality of large graphs while satisfying various application-specific criteria. Traditional clustering algorithms often struggle to capture the complex relationships within graphs and generalize to arbitrary clustering criteria. The emergence of graph neural networks (GNNs) as a powerful framework for learning representations of graph data provides new approaches to solving the problem. Previous work has shown GNNs to be capable of proposing partitionings using a variety of criteria. However, these approaches have not yet been extended to Markov chains or kinetic networks. These arise frequently in the study of molecular systems and are of particular interest to the biomolecular modeling community. In this work, we propose several GNN-based architectures to tackle the GP problem for Markov Chains described as kinetic networks. This approach aims to maximize the Kemeny constant, which is a variational quantity and it represents the sum of time scales of the system. We propose using an encoder-decoder architecture and show how simple GraphSAGE-based GNNs with linear layers can outperform much larger and more expressive attention-based models in this context. As a proof of concept, we first demonstrate the method's ability to cluster randomly connected graphs. We also use a linear chain architecture corresponding to a 1D free energy profile as our kinetic network. Subsequently, we demonstrate the effectiveness of our method through experiments on a data set derived from molecular dynamics. We compare the performance of our method to other partitioning techniques, such as PCCA+. We explore the importance of feature and hyperparameter selection and propose a general strategy for large-scale parallel training of GNNs for discovering optimal graph partitionings.

Computation-guided transcription factor biosensor specificity engineering for adipic acid detection

Transcription factor (TF)-based biosensors that connect small-molecule sensing with readouts such as fluorescence have proven to be useful synthetic biology tools for applications in biotechnology. However, the development of specific TF-based biosensors is hindered by the limited repertoire of TFs specific for molecules of interest since current construction methods rely on a limited set of characterized TFs. In this study, we present an approach for engineering the specificity of TFs through a computation-based workflow using molecular docking that enables targeted alteration of TF ligand specificity. Using this method, we engineer the LysR family BenM TF to alter its specificity from its cognate ligand cis,cis-muconic acid to adipic acid through a single amino acid substitution identified by our computational workflow. When implemented in a cell-free system, the engineered biosensor shows higher ligand sensitivity, expanding the potential applications of this circuit. We further investigate ligand binding through molecular dynamics to analyze the substitution, elucidating the impact of modulating a single amino acid position on the mechanism of BenM ligand binding. This study represents the first application of biomolecular modeling methods for altering BenM specificity and for gaining insights into how mutations influence the structural dynamics of BenM. Such methods can potentially be applied to other TFs to alter specificity and analyze the dynamics responsible for these changes, highlighting the applicability of computational tools for informing experiments. In addition, our developed adipic acid biosensor can be applied for the identification and engineering of enzymes to produce adipic acid.

Machine Learning-derived Multi-omics Prognostic Signature of Pyroptosis-related lncRNA with Regard to ZKSCAN2-DT and Tumor Immune Infiltration in Colorectal Cancer.

Colorectal cancer (CRC) has become the most prevalent gastrointestinal malignant tumor, ranking third (10.2%) in incidence and second (9.2%) in death among all malignancies globally. The most common histological subtype of CRC is colon adenocarcinoma (COAD), although the cause of CRC remains unknown, as there are no valid biomarkers. A thorough investigation was used to build a credible biomolecular risk model based on the pyroptosis-associated lncRNAs discovered for COAD prediction. Furthermore, Cibersort and Tumor Immune Dysfunction and Exclusion (TIDE), the methods of exploring tumor immune infiltration, were adopted in our paper to detect the effects of differential lncRNAs on the tumor microenvironment. Finally, quantitative real-time polymerase chain reaction (qPCR), as the approach of exploring expressions, was utilized on four different cell lines. Seven pyroptosis-related lncRNAs have been identified as COAD predictive risk factors. Cox analysis, both univariate and multivariate, revealed that the established signature might serve as a novel independent factor with prognostic meaning in COAD patients. ZKSCAN2-DT was shown to be considerably overexpressed in the COAD cell line when compared to normal human colonic epithelial cells. Furthermore, ssGSEA analysis results revealed that the immune infiltration percentage of most immune cells dropped considerably as ZKSCAN2-DT expression increased, implying that ZKSCAN2-DT may play an important role in COAD immunotherapy. Our research is the first to identify pyroptosis-related lncRNAs connected with COAD patient prognosis and to construct a predictive prognosis signature, directing COAD patient prognosis in therapeutic interventions.

Generalized biomolecular modeling and design with RoseTTAFold All-Atom.

Deep-learning methods have revolutionized protein structure prediction and design but are presently limited to protein-only systems. We describe RoseTTAFold All-Atom (RFAA), which combines a residue-based representation of amino acids and DNA bases with an atomic representation of all other groups to model assemblies that contain proteins, nucleic acids, small molecules, metals, and covalent modifications, given their sequences and chemical structures. By fine-tuning on denoising tasks, we developed RFdiffusion All-Atom (RFdiffusionAA), which builds protein structures around small molecules. Starting from random distributions of amino acid residues surrounding target small molecules, we designed and experimentally validated, through crystallography and binding measurements, proteins that bind the cardiac disease therapeutic digoxigenin, the enzymatic cofactor heme, and the light-harvesting molecule bilin.

Significance of genetic code module structure in gene expression and GC content enhancement in RNA sequences

The existent algebraic models of the genetic code contribute to the understanding of the physio-chemical characteristics of the amino acids. However, the process of translating a gene into a phenotype is highly complex. Moreover, the intricacy of gene expression gets further multiplied due to the biases in the codon usage. This paper explores an algebraic structure called module on the set of codons as well as on that of RNA sequences. We study the potential implications of these structures on gene expression and the GC content of an RNA sequence. The base order {C,U,G,A} appears to possess greater biological significance than many of the orders previously studied. We have developed a novel algorithm to generate RNA sequences with high GC content, aiming to enhance the thermostability of biomolecules. The insights gained from this investigation may have applications in biomolecular modeling and docking, protein engineering, drug development, and related fields.

Validating and building biomolecular structure models for cryo-EM maps using deep learning

Validating and building biomolecular structure models for cryo-EM maps using deep learning

Quality assessment and biomolecular structure modeling for cryo-EM using deep learning

Quality assessment and biomolecular structure modeling for cryo-EM using deep learning

Solving Stochastic Nonlinear Poisson-Boltzmann Equations Using a Collocation Method Based on RBFs

In this paper, we present a numerical scheme based on a collocation method to solve stochastic non-linear Poisson–Boltzmann equations (PBE). This equation is a generalized version of the non-linear Poisson–Boltzmann equations arising from a form of biomolecular modeling to the stochastic case. Applying the collocation method based on radial basis functions (RBFs) allows us to deal with the difficulties arising from the complexity of the domain. To indicate the accuracy of the RBF method, we present numerical results for two-dimensional models, we also study the stability of this method numerically. We examine our results with the RBF-reference value and the Chebyshev Spectral Collocation (CSC) method. Furthermore, we discuss finding the appropriate shape parameter to obtain an accurate numerical solution besides greatest stability. We have exerted the Newton–Raphson approach for solving the system of non-linear equations resulting from discretization by the RBF technique.

Editorial for the special issue “Control‐theoretic approaches for systems in the life sciences”

Editorial for the special issue “Control<scp>‐theoretic</scp> approaches for systems in the life sciences”

Real-time in vivo cancer staging of nasopharyngeal carcinoma patients with rapid fiberoptic Raman endoscopy

Cancer staging is important to guide treatment and for prognostication. This work aims to demonstrate the ability of rapid fiberoptic Raman endoscopy for real-time in vivo cancer staging of nasopharyngeal cancer (NPC) patients. We interrogate 278 tissue sites on the primary NPC with different cancer stages from 61 NPC patients and 50 healthy volunteers using rapid fiberoptic Raman endoscopy examination. Distinct Raman spectral differences of NPC at different cancer stages are observed through simultaneous fingerprint and high-wavenumber (FP/HW) Raman spectral measurements, reflecting the biomolecular differences of NPC tumor across various cancer stages. Raman staging model is established based on in vivo FP/HW tissue Raman spectra together with partial-least-squares linear-discriminant-analysis (PLS-LDA) and leave-one-tissue-site-out cross-validation (LOOCV). In vivo FP/HW Raman endoscopy provides an overall diagnostic accuracy of 92.81% for identifying different stages of NPC (i.e., NPC stage I&II and NPC stage III&IV) from normal nasopharynx. Specifically, the diagnostic sensitivity of 91.18% is obtained for identifying NPC stage I& II; and the sensitivity of 93.04% is achieved for classifying NPC stage III&IV from normal tissue. The key tissue biomolecular variations responsible for different NPC stages have been identified using biomolecular Raman modeling developed based on non-negative linear regression. The essential biomolecules (chondroitin sulfate, glucose, hemoglobin, oleic acid and triolein) are uncovered from the Raman spectra of NPC tissues through biomolecular modeling with significant variations (p < 0.05) between early-stage NPC (stage I and stage II) and late-stage NPC patients (stage III and stage IV). Our pivotal work demonstrates for the first time that fiberoptic Raman endoscopy is a robust analytical tool for real-time in vivo NPC staging in clinical settings.

SynBio in 3D: The first synthetic genetic circuit as a 3D printed STEM educational resource

Synthetic biology is a new area of science that operates at the intersection of engineering and biology and aims to design and synthesize living organisms and systems to perform new or improved functions. Despite the important role it plays in resolving global issues, instructing synthetic biology can be challenged by a limited availability of specific educational materials and techniques for explaining complex molecular mechanisms. On the other hand, digital fabrication tools, which allow the creation of 3D objects, are increasingly used for educational purposes, and several computational structures of molecular components commonly used in synthetic biology processes are deposited in open databases. Therefore, we hypothesized that the use of computer-assisted design (CAD) and 3D printing to create biomolecular structural models through hands-on interaction, followed by reflective observation, critical and analytical thinking, could enhance students’ learning in synthetic biology. In this sense, the present work describes the design, 3D printing process, and evaluation in classrooms of the molecular models of the first synthetic biological circuit, the genetic toggle switch. The 3D printed molecular structures can be freely downloaded and used by teachers to facilitate the training of STEM students in synthetic biology. Most importantly, the results demonstrated that our resource showed a significant positive impact (p &amp;lt; 0.05) on students’ learning process, indicating that the proposed method helped them better understand the genetic toggle switch.

Molecular dynamics simulations reveal the impact of NUDT15 R139C and R139H variants in structural conformation and dynamics

NUDT15, also known as MTH2, is a member of the NUDIX protein family that catalyzes the hydrolysis of nucleotides and deoxynucleotides, as well as thioguanine analogues. NUDT15 has been reported as a DNA sanitizer in humans, and more recent studies have shown that some genetic variants are related to a poor prognosis in neoplastic and immunologic diseases treated with thioguanine drugs. Despite this, the role of NUDT15 in physiology and molecular biology is quite unclear, as is the mechanism of action of this enzyme. The existence of clinically relevant variants has prompted the study of these enzymes, whose capacity to bind and hydrolyze thioguanine nucleotides is still poorly understood. By using a combination of biomolecular modeling techniques and molecular dynamics, we have studied the monomeric wild type NUDT15 as well as two important variants, R139C and R139H. Our findings reveal not only how nucleotide binding stabilizes the enzyme but also how two loops are responsible for keeping the enzyme in a packed, close conformation. Mutations in α2 helix affect a network of hydrophobic and π-interactions that enclose the active site. This knowledge contributes to the understanding of NUDT15 structural dynamics and will be valuable for the design of new chemical probes and drugs targeting this protein. Communicated by Ramaswamy H. Sarma

Recent Advances in Biomolecular Structure Modeling and Validation using Deep Learning

Since its appearance in 2021, Alphafold2, an accurate protein structure prediction method, has quickly been adopted and substantially impacted biology research, particularly in the field of structural biology. In this review, we discuss recent advancements in the development of computational structural modeling tools related to Alphafold2. Additionally, we discuss deep learning-based structure modeling methods for cryo-electron microscopy.

20 years of crystal hits: progress and promise in ultrahigh-throughput crystallization screening.

Diffraction-based structural methods contribute a large fraction of the biomolecular structural models available, providing a critical understanding of macromolecular architecture. These methods require crystallization of the target molecule, which remains a primary bottleneck in crystal-based structure determination. The National High-Throughput Crystallization Center at Hauptman-Woodward Medical Research Institute has focused on overcoming obstacles to crystallization through a combination of robotics-enabled high-throughput screening and advanced imaging to increase the success of finding crystallization conditions. This paper will describe the lessons learned from over 20 years of operation of our high-throughput crystallization services. The current experimental pipelines, instrumentation, imaging capabilities and software for image viewing and crystal scoring are detailed. New developments in the field and opportunities for further improvements in biomolecular crystallization are reflected on.

Insights from the SAMPL8 physical properties blind prediction challenge.

The SAMPL community challenges help draw attention to major challenges impairing accuracy of biomolecular and physical models for drug discovery. The eighth iteration of SAMPL physical properties focused on prediction of pKa and logD. The pKa predictions were focused on compounds with multiple titratable functional groups. The logD portion of the challenge focused on making logD predictions for the same compounds in 7 different organic solvents such as between water and ethyl acetate or water and heptane. Here we report on the work we did for the SAMPL8 physical properties challenge which included the development of an automated high throughput to measure the pKa and the logD. We then used the acquired dataset to stage a blind challenge we used to benchmark different empirical and physical models. We analyzed the performance of various methods using error metrics such as RMSE, MAE and ME as well as correlational metrics such Rsquared, slope and Kendall Tau. Here, we discuss the SAMPL8 dataset and the results of the SAMPL8 challenge, including the pKa prediction and logD components. We will also discuss insights gained from the challenge, including which methods performed well and which performed poorly, and give suggestions for the design of future SAMPL challenges.

Simulations reveal molecular mechanisms of high affinity and selective binding of the tarantula venom protoxin-2 to the human NaV1.7 channel.

Human voltage-gated sodium channels (hNaV) are responsible for initiating and propagating action potentials in excitable cells and have been associated with numerous cardiac and neurological disorders. hNaV1.7 channels, present at the endings of pain-sensing nerves, have garnered a lot of scientific interest as potential targets for pain therapeutics by channel inhibition. Certain natural peptide toxins, such as the tarantula protoxin-2 (ProTx2), have high selectivity for hNaV1.7, and hence they serve as valuable scaffolds for peptide design in pain therapeutics. Here, we present the mechanisms by which ProTx2 can bind and modulate hNaV1.7 channel gating kinetics. Using Rosetta biomolecular modeling software, we constructed atomistic structural models of the hNaV1.7 Voltage-Sensing Domain (VSD) 2 and 4 in the activated and deactivated states. Then, we docked ProTx2 onto the VSDs, embedded the structures into lipid bilayers solvated by aqueous NaCl solution, and performed microsecond-long all-atom molecular dynamics (MD) simulations of those systems using the CHARMM36m force field and Amber PMEMD software. Afterward, we performed binding energetics analysis using molecular mechanics Poisson-Boltzmann surface area (MM/PBSA) calculations and conducted Brownian Dynamics simulations using BrownDye to calculate the second-order association rate constants. Our analysis revealed numerous important inter-residue contacts that contribute to the toxin's high affinity for the channel. By understanding the various state- and subtype-specific mechanisms of toxin binding to the channel, we can develop strategies to improve the selectivity and potency of peptide toxins to create effective and safe pain therapeutics. In addition, the results could be incorporated into functional kinetic models to elucidate how ProTx2 can modulate electrical signal propagation in dorsal root ganglion neurons and thus pain sensing.

Editorial: Tensor numerical methods and their application in scientific computing and data science

Editorial: Tensor numerical methods and their application in scientific computing and data science

Biomolecular Transition Models Applied to Water Decontamination Processes in Crops Suitable for Human Consumption

Biomolecular Transition Models Applied to Water Decontamination Processes in Crops Suitable for Human Consumption