ChemRxiv
Not a member yet
27047 research outputs found
Sort by
Is BigSMILES the Friend of Polymer Machine Learning?
No full textComputational methods, exemplified by machine learning (ML), have provided theoretical guidance and solutions for the development of sustainable polymers, accelerating advancements in materials for societal needs such as equipment, environment, health, and green energy. In previous polymer ML workflows, the Simplified Molecular-Input Line-Entry System (SMILES) notation has consistently served as the primary representation of polymer structures, though the inherent randomness of polymers has long posed challenges for SMILES in the representation learning of polymer ML.
Recently, BigSMILES and its extensions have paved the way for more versatile and concise representation of polymer structures. However, whether BigSMILES outperforms SMILES in polymer ML workflows has yet to be systematically explored and demonstrated. To fill this scientific gap, we conducted extensive experiments investigating this question, encompassing a variety of polymer property prediction and inverse design tasks based on both image and text inputs. Our findings reveal that in 11 tasks involving homopolymer systems, BigSMILES-based ML workflows exhibit performance comparable to or even exceeding that of SMILES, underscoring the efficacy of BigSMILES in representing polymer structures. Furthermore, BigSMILES offers a more compact textual representation compared to SMILES, significantly reducing the computational cost of model training, particularly for large language models. Through these comprehensive experiments, we for the first time demonstrate that BigSMILES can achieve performance on par with SMILES, while also facilitating faster model training and reducing energy consumption, which could have a substantial impact on a wide range of polymer tasks in the future, including property prediction (and classification) and polymer generation across various polymer types
Powerful and Reliable Prediction using Latent Variables of Experimentally Unobservable Reactions in Organic Synthesis
No full textIn this study, a novel machine learning algorithm was designed to assist in the development of organic reactions. This algorithm addresses the complexities inherent in batch- type organic reactions, including the necessity for numerous experiments and the effects of intricate characteristics of reaction pathways. By integrating molecular relationships and actual yields from observable reactions, the algorithm is used to estimate untested yields via extrapolation. An approach based on Bayesian optimization and dual annealing optimization is employed to compute expected values and evaluate plausibility. The algorithm’s dual-loop 2 structure, incorporating latent variables and experimental values, maximizes the coefficient of determination. Physicochemical aspects of the algorithm are validated using natural bond orbital charges, and its utility in synthesizing perfluoroiodinated naphthalenes is demonstrated. The algorithm exhibits potential for application in predicting experimentally unobservable reactions, thereby advancing the field of synthetic organic chemistry
Angle-Resolved Photoemission Spectroscopy and Electrocatalysis: Is There a Bridge in Between?
No full textThe electronic interaction between molecules and a surface greatly influences the kinetics of heterogeneous electrocatalytic processes. Here, we explore the possibility of using angle-resolved photoemission spectroscopy (ARPES) to study electrocatalysts. Using oxide materials as an example, we identify pitfalls and key challenges of such experiments and discuss future opportunities
Pharmacological Evaluation of Enantiomers of AZ11645373 against the P2X7 Receptor
No full textThe P2X purinergic receptor 7 (P2X7) has an essential role in inflammation, innate immunity, tumor progression, neurodegenerative diseases, and several other diseases, leading subsequently to the development of P2X7 modulators. AZ11645373 is a frequently studied P2X7 antagonist tool com-pound, but always as a racemic mixture. Racemic AZ11645373 can be separated, into its respective enantiomers by chiral chromatography, albeit in small batches, and these were stereochemically in-tact over a year later, by chiral HPLC analysis. On a higher scale, significant decomposition is ob-served. One of the enantiomers was crystallised as a palladium complex and its (R)-configuration was determined by single crystal X-ray diffraction, further confirmed, in solution, by vibrational circular dichroism. Biological studies demonstrated that both (S)- and (R)-forms were able to fully inhibit human P2X7, but (R)-AZ11645373 was more potent, with an IC50 of 32.9 nM. Contrary to its effect on human P2X7, (S)-AZ11645373 was ineffective on mouse P2X7, while the (R)-AZ11645373 enantiomer was a full antagonist. These results demonstrated that the antagonistic effects of racemic AZ11645373 are mainly due to its (R)-enantiomer. Site-directed mutagenesis and molecular dynam-ics simulations indicated that the (R)-enantiomer may form specific interactions with Phe95 and the antagonists bound to the other P2X7 monomers. Phe95 is situated at the channel pore and appears to be the pivotal molecular gateway between AZ11645373 allosteric binding and locking of the closed state of the P2X7 channel. All together, these structure-function relationships should be helpful for drug design of P2X7 modulators
Analysis of adenosine phosphonucleotides in blood by nh-PHIP
No full textThe sensitivity of conventional NMR is limited to metabolites at mid-micromolar concentrations, however diagnostically relevant metabolites are often found at lower concentrations. nh-PHIP has shown promise for analyzing complex biological samples like urine, but adapting it to blood presents additional challenges due to its complex matrix. Herein we successfully demonstrate the use of nh-PHIP for detecting adenine phosphonucleotides in blood
Method and theory of genome profiling (GP) developed for identification and classification of organisms
No full textDetails of the method and theory on the genome profiling (GP), which provides with the method for species identification and classification of organisms, are first described. GP consists of three major steps: i.e., random PCR, micro-temperature gradient gel electrophoresis (μTGGE), and extraction and processing of featuring points contained in genome profiles (results of the former step) generating species identification dots (spiddos). Methodological, physical and biological meanings of ‘random PCR’, ‘μTGGE’, ‘spiddos’, ‘genome distance (dG)’ and others introduced in this study are originally discussed. This paper gives a base for understanding the diversity of GP achievements performed for 30 years written in an alternative review on GP (i.e., to appear in Briefings in Functional Genomics (in process)), especially underscoring the importance of the spiddos parameter for calculating closeness of species and constructing genome database.
Spiddos contains such a kind of information termed as SIOWS (sequence information obtained without sequencing), which is essentially unique and important for the GP technology, derived from the sequence-specific DNA melting phenomenon together with DNA melting theory. Succeedingly, one can understand why GP enables us to draw the sufficient amount of information without sequencing of genomic DNA. This also explains why identification/classification of species can be so readily and universally done by GP. It requires the perspective on the nature of spiddos. Since this paper first provides detailed procedures of GP and in-depth theoretical explanation of the GP-related phenomena, a wide range of scientists (Bacteriology, Infectious Disease Medicine, Epidemiology, Environmental Science, Biodiversity Science, Mutagen Research, Taxonomy, Bio-database Science, and others) can engage in applications of GP, which is difficult without sufficient understanding of the method and theory of GP
A conformation-specific approach to native top-down mass spectrometry
No full textNative top-down mass spectrometry is a powerful approach for characterizing proteoforms, and has recently been applied to provide similarly powerful insights into protein conformation. Current approaches, however, are limited such that struc-tural insights can only be obtained for the entire conformational landscape in bulk, or without any direct conformational measurement. We report a new ion mobility-enabled method for performing native top-down MS in a conformation-specific manner. Our approach identified conformation-linked differences in backbone dissociation for the model protein calmodulin, which simultaneously inform upon proteoform variations and provide structural insights. We also illustrate that our method can be applied to protein-ligand complexes, either to identify components or to probe ligand-induced structural changes
Equilibrium and Non-equilibrium Ensemble Methods for Accurate, Precise and Reproducible Absolute Binding Free Energy Calculations
No full textFree energy calculations for protein-ligand complexes have become widespread in recent years owing to several conceptual, methodological and technological advances. Central among these is the use of ensemble methods which permits accurate, precise and reproducible predictions and are necessary for uncertainty quantification. Absolute binding free energies (ABFEs) are challenging to predict using alchemical methods and their routine application in drug discovery has remained out of reach until now. Here, we apply ensemble alchemical ABFE methods to a large dataset comprising 219 ligand-protein complexes and obtain statistically robust results with high accuracy (< 1 kcal/mol). We compare equilibrium and non-equilibrium methods for ABFE predictions at large scale and provide a systematic critical assessment of each method. The equilibrium method is more accurate, precise, faster, computationally more cost-effective and requires a much simpler protocol, making it preferable for large scale and blind applications. We find that the calculated free energy distributions are non-normal and discuss the consequences. We recommend a definitive protocol to perform ABFE calculations optimally. Using this protocol, it is possible to perform thousands of ABFE calculations within a few hours on modern exascale machines
Efficient Training of Neural Network Potentials for Chemical and Enzymatic Reactions by Continual Learning
No full textThe machine learning (ML) method has emerged as an efficient surrogate for high-level electronic structure theory, offering precision and computational efficiency. However, the construction of a general force field remains challenging due to the vast conformational and chemical space. Training data sets typically cover only a limited region of this space, resulting in poor extrapolation performance. Traditional strategies inadequately address this problem by training models from scratch using both old and new datasets. In addition, model transferability is crucial for general force field construction. Existing ML force fields, designed for closed systems with no external environmental potential, exhibit limited transferability to complex condensed phase systems such as enzymatic reactions, resulting in inferior performance and high memory costs. Our ML/MM model based on the Taylor expansion of the electrostatic operator showed high transferability between reactions in several simple solvents. In this work, we extend the strategy to enzymatic reactions to explore transferability between more complex heterogeneous environments. In addition, we also apply continual learning strategies based on memory datasets to enable autonomous and on-the-fly training on a continuous stream of new data. By combining these two methods, we can construct a more general force field more efficiently
Template-assisted electrospinning and 3D printing of multilayered hierarchical vascular grafts
No full textFabricating complex hierarchical structures mimicking natural vessels and arteries is pivotal for addressing problems of cardiovascular diseases. Various fabrication strategies have been explored to achieve this goal, each contributing unique advantages and challenges to the development of functional vascular grafts. In this study, a three-layered tubular structure resembling vascular grafts was fabricated using biocompatible and biodegradable copolymers of poly(butylene succinate)(PBS) using advanced manufacturing techniques. The outer layer was fabricated by template-assisted electrospinning utilizing 3D printed scaffold with a precise hexagonal pore design as the template, and the inner layer was coated with gelatin through perfusion. Cellulose nanocrystals (CNC) were incorporated into electrospun fibers to enhance mechanical properties. Gelatin coating was applied to the lumen using perfusion coating, resembling inner layer. Integration of 3D printed structures with electrospun fibers via template-assisted electrospinning, and gelatin coating resulted in a seamless multilayered scaffold. Mechanical testing demonstrated robustness, surpassing natural arteries in some aspects, while gelatin coating significantly reduced liquid leakage, ensuring leak-free functionality. Cytotoxicity assessment confirmed biocompatibility of processed materials with fibroblast cells, supporting potential for medical applications