Search NASA⌕ Search

SEARCH · Search NASA

Results for “Cheminformatics”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Leveraging Triphenylphosphine-Containing Polymers to Explore Design Principles for Protein-Mimetic Catalysts

Complex interactions between noncoordinating residues are significant yet commonly overlooked components of macromolecular catalyst function. While these interactions have been demonstrated to impact binding affinities and catalytic rates in metalloenzymes, the roles of similar structural elements in synthetic polymeric catalysts remain underexplored. Using a model Suzuki–Miyuara cross-coupling reaction, we performed a series of systematic studies to probe the interconnected effects of metal–ligand cross-links, electrostatic interactions, and local rigidity in polymer catalysts. To achieve this, a novel bifunctional triphenylphosphine acrylamide (BisTPPAm) monomer was synthesized and evaluated alongside an analogous monofunctional triphenylphosphine acrylamide (TPPAm). In model copolymer catalysts, increased initial reaction rates were observed for copolymers untethered by Pd complexation (BisTPPAm-containing) compared to Pd-cross-linked catalysts (TPPAm-containing). Further, incorporating local rigidity through secondary structure-like and electrostatic interactions revealed nonmonotonic relationships between composition and the reaction rate, demonstrating the potential for tunable behavior through secondary-sphere interactions. Lastly, through rigorous cheminformatics featurization strategies and statistical modeling, we quantitated relationships between chemical descriptors of the substrate and reaction conditions on catalytic performance. Collectively, these results provide insights into relationships among the composition, structure, and function of protein-mimetic catalytic copolymers.

Catalysts↗
DOE OSTI2021

Bond order predictions using deep neural networks

Machine learning is an extremely powerful tool for the modern theoretical chemist since it provides a method for bypassing costly algorithms for solving the Schrödinger equation. Already, it has proven able to infer molecular and atomic properties such as charges, enthalpies, dipoles, excited state energies, and others. Most of these machine learning algorithms proceed by inferring properties of individual atoms, even breaking down total molecular energy into individual atomic contributions. In this paper, we introduce a modified version of the Hierarchically Interacting Particle Neural Network (HIP-NN) capable of making predictions on the bonds between atoms rather than on the atoms themselves. We train the modified HIP-NN to infer bond orders for a large number of small organic molecules as computed via the Natural Bond Orbital package. We demonstrate that the trained model is extensible to molecules much larger than those in the training set by studying its performance on the COMP6 dataset. This method has applications in cheminformatics and force field parameterization and opens a promising future for machine learning models to predict other quantities that are defined between atoms such as density matrix elements, Hamiltonian parameters, and molecular reactivities.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

Structure-based group A streptococcal vaccine design: Helical wheel homology predicts antibody cross-reactivity among streptococcal M protein–derived peptides

Group A streptococcus (Strep A) surface M protein, an α-helical coiled-coil dimer, is a vaccine target and a major determinant of streptococcal virulence. The sequence-variable N-terminal region of the M protein defines the M type and also contains epitopes that promote opsonophagocytic killing of streptococci. Recent reports have reported considerable cross-reactivity among different M types, suggesting the prospect of identifying cross-protective epitopes that would constitute a broadly protective multivalent vaccine against Strep A isolates. In this work, we have used a combination of immunological assays, structural biology, and cheminformatics to construct a recombinant M protein–based vaccine that included six Strep A M peptides that were predicted to elicit antisera that would cross-react with an additional 15 nonvaccine M types of Strep A. Rabbit antisera against this recombinant vaccine cross-reacted with 10 of the 15 nonvaccine M peptides. Two of the five nonvaccine M peptides that did not cross-react shared high sequence identity (≥50%) with the vaccine peptides, implying that high sequence identity alone was insufficient for cross-reactivity among the M peptides. Additional structural analyses revealed that the sequence identity at corresponding polar helical-wheel heptad sites between vaccine and nonvaccine peptides accurately distinguishes cross-reactive from non–cross-reactive peptides. On the basis of these observations, we developed a scoring algorithm based on the sequence identity at polar heptad sites. When applied to all epidemiologically important M types, this algorithm should enable the selection of a minimal number of M peptide–based vaccine candidates that elicit broadly protective immunity against Strep A.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Metabolite Damage and Damage Control in a Minimal Genome

Analysis of the genes retained in the minimized Mycoplasma JCVI-Syn3A genome established that systems that repair or preempt metabolite damage are essential to life. Several genes known to have such functions were identified and experimentally validated, including 5-formyltetrahydrofolate cycloligase, coenzyme A (CoA) disulfide reductase, and certain hydrolases. Furthermore, we discovered that an enigmatic YqeK hydrolase domain fused to NadD has a novel proofreading function in NAD synthesis and could double as a MutT-like sanitizing enzyme for the nucleotide pool. Finally, we combined metabolomics and cheminformatics approaches to extend the core metabolic map of JCVI-Syn3A to include promiscuous enzymatic reactions and spontaneous side reactions. This extension revealed that several key metabolite damage control systems remain to be identified in JCVI-Syn3A, such as that for methylglyoxal.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

MACAW v1.0

The ability to embed molecules in a numeric space is essential in order to build mathematical and machine-learning models describing molecular properties or other processes affected by molecules. MACAW is a cheminformatic tool that allows embedding small molecules into a multidimensional numeric space. In the embedding, each molecule is assigned a numeric vector that captures information of the molecule in relation to other molecules, and that vector can be used as input to mathematical models. Molecules that are more similar to each other are embedded closer in this numeric space, whereas molecules that are more different are embedded further away. One advantage of the MACAW embedding technology compared to established alterantives is that it is fast and does not require extensive computational resources or expertise. In particular, MACAW embeddings can be used as input to mathematical models without the need for variable cleaning or feature selection, saving time and simplifying their use. On the other hand, MACAW also contains methods to generate new molecules and to recommend new molecules satisfying a desired molecular property. The generation of new molecules can be biased based on an input set of molecules, effectively generating molecular diversity around it. In its turn, MACAW's molecular recommendation tool is a novel method for evolving molecules in silico towards a desired molecular specification. In this method, the biased molecular generator tool is applied iteratively in combination with a molecular selection step. As a result, in each iteration the molecules selected by the software are increasingly closer to the desired specification. Both the molecular generation and the molecular recommendation tools are very fast, efficient, and intuitive to use.

Roger, VincentBlay↗
DOE OSTI2024

CACTUS

CACTUS is an innovative LLM‐based agent that combines the cognitive capabilities of large language models with domain‐specific tools. By integrating cheminformatics tools, CACTUS assists researchers in tasks such as molecular property prediction, similarity searching, and drug‐likeness assessment.

Kumar, Neeraj↗
DOE OSTI2023

ORNL_AISD-Ex: Quantum chemical prediction of UV/Vis absorption spectra for over 10 million organic molecules

We performed calculations of electronic excitation energies and associated oscillator strengths based on the time-dependent density-functional tight-binding (TD-DFTB) method [1]. The SMILES (Simplified molecular-input line-entry system) strings of the molecules from the AISD HOMO-LUMO database [2] were converted to a 3D atomistic structure and stored in a PDB file after preliminary geometry optimization using the Merck Molecular Force Field (MMFF94) in RDKit [3,4]. The primary information stored in the PDB file archive consists of Cartesian coordinates for each atom of the molecule in their 3D location in space, along with summary information about the structure, sequence, and experiment. We then performed molecular geometry optimization using the density-functional tight-binding (DFTB) method [5] in the electronic ground state, followed by single-point excited states calculations, as described below. We note that, since RDKit employs a random choice for the generation of molecular conformers, the molecular geometries obtained in this dataset could be different from the ones that were generated when the AISD HOMO-LUMO dataset was generated. The computed excitation energies and associated oscillator strengths can be converted to predict UV/Vis absorption spectra, where excitation energies correspond to absorption peak positions, and oscillator strengths are a good measure of the probability of absorption of visible or UV light in transitions between electronic ground and excited states. The conversion of SMILES strings to 3D Cartesian coordinates of fully DFTB-optimized molecules was successful for 10,502,904 out of 10,502,917 molecules. For these molecules, both geometry optimizations and excited states calculations were successful. The DFTB calculations did not complete for 13 molecules of the original AISD HOMO-LUMO dataset. We still provide information about the geometry of these molecules. The molecules are diverse for chemical compositions (which span 5 non-hydrogen elements: oxygen, carbon, nitrogen, fluorine, sulfur) and molecular size (the smallest molecule contains 5 non-hydrogen atoms, and the largest molecule contains 71 non-hydrogen atoms). The DFTB method [5] is an approximation to density functional theory (DFT), utilizing a minimal basis set in conjunction with a two-center approximation to the electronic Hamiltonian and overlap matrix elements. The DFTB total energy is the sum of an electronic and a repulsive energy contribution, and their calculation requires optimized electronic parameters and diatomic repulsive potential energy functions. All DFTB calculations were performed using the DFTB+ code [6] (version 21.2) and the wrapper for DFTB+ in the Atomic Simulation Environment (ASE) (version 3.22.1) [7], which performed an internal conversion of Cartesian coordinates from PDB to the .gen file format. For the geometry optimizations on the electronic ground state potential energy surface of the molecules, we have chosen the third-order DFTB (DFTB3) method [5c] and employed the matching 3ob set of electronic parameters and repulsive potentials [8]. The empirical γ-damping for hydrogen bond correction, and Grimme's D3 empirical dispersion correction with Becke-Johnson damping (D3(BJ)) [9] dispersion correction was included to improve the description of non-covalent interactions. For excited states single-point energy calculations, we employed the TD-DFTB method in conjunction with the DFTB2 method [5b] and the matching mio [5b,10] and halorg [11] parameter sets. We opted to request the simultaneous calculation of 50 excited states for singlet transition to investigate sufficient number of excited states, based on linear response theory using the Casida equation [Ref: T. A. Niehaus, S. Suhai, F. Della Sala, P Lugli, M. Elstner, G. Seifert, and Th. Frauenheim. Tight-binding approach to time-dependent density-functional response theory. Phys. Rev. B, 63:085108, 2001] and the ARPACK diagonalizer [R. B. Lehoucq, D. C. Sorensen, and C. Yang. Arpack users guide: Solution of large-scale eigenvalue problems by implicitly restarted arnoldi methods, 1997. 46, 51]. The dataset contains 1001 tar.gz files. Tar files are named as “ornl_aisd_ex_1.tar.gz†through “ornl_aisd_ex_1000.tar.gzâ€. Additionally, the 13 failed molecules are in “ornl_aisd_ex_unprocessed.tar.gzâ€. Except for the tar files listed below, each tar file contains 10,500 molecules. Tar files numbered 34, 121, 128, 352, 360, 429, 495, 509, 518, 627, 676, 668, and 862 contain 10,499 molecules each. The last tar file numbered 1000 contains 13,417 molecules. The total size of the uncompressed dataset is over 283 Gigabytes. The code for calculating the electronic excitation energies and statistical analysis of the dataset is provided at the following GitLab repository: https://github.com/ORNL/Analysis-of-Large-Scale-Molecular-Datasets-with-Python Calculating the UV spectrum of a molecule requires performing 3 main operations: 1. Converting the smiles string representation of a molecule into a geometric structure where each atom is assigned XYZ coordinates. The geometric structure is written to the file smiles.pdb. 2. Using smiles.pdb to compute the relaxed geometry of the molecule, which corresponds with the position of the atoms at the position of equilibrium at the ground state. This generates the files band.out, detailed.out, and geo_end.gen. 3. Using geo_end.gen to calculate the UV spectrum of the molecule which is written into the file EXC.DAT. Every molecule in the dataset has its own directory. The files contained in each molecule directory are as follows: 1. geo_end.gen 2. detailed.out 3. band.out 4. EXC.DAT 5. smiles.pdb REFERENCES [1] Niehaus, T. A.; Suhai, S.; Della Salla, F.; Lugli, P.; Elstner, M.; Seifert, G.; Frauenheim, Th. Tight-binding approach to time-dependent density-functional response theory. Phys. Rev. B, 2001, 63, 085108/1-9. [2] Blanchard, A.; Gounley, J.; Metha, K.; Yoo, P.; Irle, S. AISD HOMO-LUMO. DOI: 10.13139/ORNLNCCS/1869409 [3] RDKit: Cheminformatics and Machine Learning Software. 2013, [http://www.rdkit.org] [4] Tosco, P.; Stiefl, N. and Landrum, G. Bringing the MMFF force field to the RDKit: implementation and validation. J Cheminform. 2014, 6, 1–4. [5] a) Porezag, D.; Frauenheim, T.; Kohler, T.; Seifert, G.; Kaschner, Construction of tight-binding-like potentials on the basis of density-functional theory: Application to carbon, R. Phys. Rev. B 1995, 51, 12947-12957; b) Elstner, M.; Porezag, D.; Jungnickel, G.; Elsner, J.; Haugk, M.; Frauenheim, Th.; Suhai, S.; Seifert, G.; Phys. Rev. B 1998, 58, 7260-7268; c) Gaus, M.; Cui, Q.; Elstner, M. DFTB3: Extension of the Self-Consistent-Charge Density-Functional Tight-Binding Method (SCC-DFTB), J. Chem. Theory Comput. 2011, 7, 931-948; d) Cui, Q.; Elstner, M. Density functional tight binding: values of semi-empirical methods in an ab initio era, Phys. Chem. Chem. Phys. 2014, 16, 14368-14377. [6] Hourahine, B. et al. DFTB+, a software package for efficient approximate density functional theory based atomistic simulations, J. Chem. Phys. 2020, 152, 124101/1-19. [7] Larsen, A. H. et al. The atomic simulation environment—a Python library for working with atoms. J. Phys.: Cond. Matter 2017, 29, 273002. [8] Kubillus, M.; Kubar, T.; Gaus, M.; Rezac, J.; Elstner, M. Parameterization of the DFTB3 Method for Br, Ca, Cl, F, I, K, and Na in Organic and Biological Systems, J. Chem. Theory Comput. 2015, 11, 332-342. [9] Brandenburg, J. G.; Grimme, S. Accurate Modeling of Organic Molecular Crystals by Dispersion-Corrected Density Functional Tight Binding (DFTB), J. Phys. Chem. Lett. 2014, 5, 1785−1789. [10] a) Niehaus, T. A.; Elstner, M.; Frauenheim, Th.; Suhai, S. Application of an approximate density-functional method to sulfur containing compounds. J. Mol. Struct.: THEOCHEM 2001, 541, 185-94; b) Elstner, M.; Hobza, P.; Frauenheim, Th.; Suhai, S.; Kaxiras, E. Hydrogen bonding and stacking interactions of nucleic acid base pairs: A density-functional-theory based treatment. J. Chem. Phys. 2001, 114, 5149-55. [11] Kubar, T.; Bodrog, Z.; Gaus, M.; Köhler, C.; Aradi, B.; Frauenheim, Th.; Elstner, M. Parametrization of the SCC-DFTB Method for Halogens. J. Chem. Theory Comput. 2013, 9, 2939-49.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Predicting drug-metagenome interactions: Variation in the microbial β-glucuronidase level in the human gut metagenomes

Characterizing the gut microbiota in terms of their capacity to interfere with drug metabolism is necessary to achieve drug efficacy and safety. Although examples of drug-microbiome interactions are well-documented, little has been reported about a computational pipeline for systematically identifying and characterizing bacterial enzymes that process particular classes of drugs. The goal of our study is to develop a computational approach that compiles drugs whose metabolism may be influenced by a particular class of microbial enzymes and that quantifies the variability in the collective level of those enzymes among individuals. The present paper describes this approach, with microbial β-glucuronidases as an example, which break down drug-glucuronide conjugates and reactivate the drugs or their metabolites. We identified 100 medications that may be metabolized by β-glucuronidases from the gut microbiome. These medications included morphine, estrogen, ibuprofen, midazolam, and their structural analogues. The analysis of metagenomic data available through the Sequence Read Archive (SRA) showed that the level of β-glucuronidase in the gut metagenomes was higher in males than in females, which provides a potential explanation for the sex-based differences in efficacy and toxicity for several drugs, reported in previous studies. Our analysis also showed that infant gut metagenomes at birth and 12 months of age have higher levels of β-glucuronidase than the metagenomes of their mothers and the implication of this observed variability was discussed in the context of breastfeeding as well as infant hyperbilirubinemia. Overall, despite important limitations discussed in this paper, our analysis provided useful insights on the role of the human gut metagenome in the variability in drug response among individuals. Importantly, this approach exploits drug and metagenome data available in public databases as well as open-source cheminformatics and bioinformatics tools to predict drug-metagenome interactions.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Combined Experimental and Computational Efforts to Establish Ion Mobility, Solubility and Stability of Functional Liquids for Electrochemical Energy Storage

This work provides a computation-driven investigation of the stability of organic electrolytes for lithium-air batteries. Electrolyte instability is currently a key challenge that limits practical use of aprotic Li-air batteries, and the chemical processes that cause this instability are often kinetically-driven. Computational screening for kinetic stability involves the determination of reaction barriers for the numerous potential reaction mechanisms, barriers that are challenging to calculate due to the difficulty of locating transition state structures. Here we screen a broad set of substituted electrolytes for susceptibility to nucleophilic attack by superoxide. We find that carbonates are not typically expected to be stable and that sulfones are generally stable, validating literature trends. We study the effects of chemical functionalization with electron-donating and withdrawing groups and their interplay with steric factors, identifying functional groups and other chemical modifications that increase stability in these groups. User-input driven transition state identification is used for these initial calculations, and an automated computational pipeline is subsequently presented and validated as a means to perform further high-throughput searches across mechanisms and chemistries. The pipeline integrates cheminformatics-based reaction encoding, relaxed potential energy scans, and nudged elastic band calculations for an end-to-end approach to barrier calculations. We review this automated search approach and its current limitations, and discuss challenges and further work.

25 ENERGY STORAGE↗
DOE OSTI2024

Spin-Controllable Dynamics in Defect-Engineered Carbon Nanotubes as Single Photon Emitters: Data-Driven Modeling and Computations

Quantum technologies, such as quantum computing and sensing, require efficient single-photon emission (SPE) sources that operate at room temperature in telecom wavelengths. While several materials can serve as SPE sources, no single platform meets all the criteria for efficiency, ambient operation, and scalability. Single-walled carbon nanotubes (SWCNTs) with covalently attached molecules offer a promising solution. Their SPE can be easily tuned via modifications of the SWCNT's diameter, chirality, and bonded molecules, enabling emission across near-IR to telecom wavelengths at ambient conditions. However, to fully realize the potential of SWCNTs and unlock their quantum capabilities, a deeper understanding of how structural defects from molecular adducts affect their emission and competing photoexcited processes is essential. To address this gap in our knowledge, this project combined quantum chemistry calculations with data-driven methods of cheminformatics (QSAR) and machine learning (ML). The developed computational approaches have provided several design strategies for covalent functionalization of SWCNTs to improve their optical response. The collaboration with Los Alamos National Lab (LANL) enabled direct comparison of computational and experimental data, facilitating method validation. This partnership was enhanced through access to LANL's Center for Integrated Nanotechnologies (CINT) utilizing User Facility Program and summer internships, which provided three NDSU graduate students with hands-on experience at LANL. The outcomes of this project included (1) Advancing the current stage of computational methods in accurate modeling of non-adiabatic spin-dependent photoexcited dynamics and its applicability to nanosystems consisting of thousands of atoms, realized as open-access codes linked to existing DFT-based software; (2) Establishing the relationship between the structure of adducts and SWCNTs and intrinsic excitonic and spin properties of defect states for guiding novel synthetic strategies and experimental probes of chemically functionalized SWCNTs as near-IR emitting materials; (3) Generating virtual libraries of hypothetical functionalized SWCNTs for virtual screening of their chemical structures and optical properties, leveraging new functionalities of SWCNTs; (4) Offering a unique experience for NDSU graduate students that prepared them for future scientific careers related to materials modeling and big data processing. These results were summarized in 12 published journal papers and 3 recently submitted papers. One of a key finding is that the position of defect sites on the SWCNT surface primarily drives the emission redshift (up to 100 meV), while the polarity of the defect-inducing molecules has a much smaller effect (~10 meV). However, the electron-donating or withdrawing properties of a molecule influence selecting reactivity of defect sites. These insights important for optimizing synthetic protocols for desired emissions in SWCNTs. We also revealed that the interaction between two defects at various positions on the SWCNT enhances the redshift and optical activity of states, favoring strong near-IR emission. This suggests that manipulations in defect concentrations is a promising strategy for controlling efficient emission. Mostly important, the defect position was found controllable by the spin states of photoexcited intermediates: Excited aromatic molecules form ortho defects with SWCNTs at their singlet states in the presence of oxygen, while oxygen-free conditions favor para defects via the triplet-state mechanism. Additionally, a heat-activated [2+2] cycloaddition reaction facilitates divalent defect formation with fewer bonding positions that narrows emission bands. These groundbreaking findings have been experimentally validated and significantly advance our understanding of defect chemistry in SWCNTs. Using a novel encoding technique and 3D-MoRSE descriptors, we developed highly accurate ML/QSAR models to predict both the 3D structure and optical properties of SWCNTs with chemical defects. This model enabled the creation of a virtual library of 125,556 structures, providing new insights into the relationship between SWCNT-defect structure and emission.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2021

The moderately defficient enzyme: Catalysis-related damage in vivo and its repair

Enzymes have in vivo lifespans. Analysis of lifespans – lifetime totals of catalytic turnovers – suggests that non-survivable collateral chemical damage from the very reactions that enzymes catalyze is a common but underdiagnosed cause of enzyme death. Analysis also implies that many enzymes are moderately deficient in that their active-site regions are not naturally as hardened against such collateral damage as they could be, leaving room for improvement by rational design or directed evolution. Enzyme lifespan might also be improved by engineering systems that repair otherwise fatal active-site damage, of which a handful are known and more are inferred to exist. Unfortunately, the data needed to design and execute such improvements is lacking: there are too few measurements of in vivo lifespan, and existing information on the extent, nature, and mechanisms of active-site damage and repair during normal enzyme operation is too scarce, anecdotal, and speculative to act on. Fortunately, advances in proteomics, metabolomics, cheminformatics, comparative genomics, and structural biochemistry now empower a systematic, data-driven approach to identify, predict, and validate instances of active-site damage and its repair. These capabilities would be practically useful in enzyme redesign and improvement of in-use stability, and could change thinking about which enzymes die young in vivo, and why.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

The moderately defficient enzyme: Catalysis-related damage in vivo and its repair

Enzymes have in vivo lifespans. Analysis of lifespans – lifetime totals of catalytic turnovers – suggests that non-survivable collateral chemical damage from the very reactions that enzymes catalyze is a common but underdiagnosed cause of enzyme death. Analysis also implies that many enzymes are moderately deficient in that their active-site regions are not naturally as hardened against such collateral damage as they could be, leaving room for improvement by rational design or directed evolution. Enzyme lifespan might also be improved by engineering systems that repair otherwise fatal active-site damage, of which a handful are known and more are inferred to exist. Unfortunately, the data needed to design and execute such improvements is lacking: there are too few measurements of in vivo lifespan, and existing information on the extent, nature, and mechanisms of active-site damage and repair during normal enzyme operation is too scarce, anecdotal, and speculative to act on. Fortunately, advances in proteomics, metabolomics, cheminformatics, comparative genomics, and structural biochemistry now empower a systematic, data-driven approach to identify, predict, and validate instances of active-site damage and its repair. These capabilities would be practically useful in enzyme redesign and improvement of in-use stability, and could change thinking about which enzymes die young in vivo, and why.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Evaluating the Use of Foundational Chemical Language Models in Multimodal Graph Fusion

Rapid and accurate prediction of the physicochemical properties of molecules given their structures remains a key challenge in cheminformatics. Machine learning approaches offer high-throughput options, but the optimality of inductive biases and data representations are up for debate. For example, BERT-based masked language models (MLMs) can be trained in a self-supervised way on hundreds of millions to billions of readily available SMILES strings. Another option is graph neural networks (GNNs), which can operate directly on molecular structures. Yet, generating accurate molecular geometry is computationally expensive, leading to a relative scarcity in data compared to SMILES strings. It is attractive to combine these two paradigms by pre-training an LM on a large corpus of SMILES strings and embedding these representation into a geometric graph neural network. Despite the promise of such an approach, and contrary to previous studies, we find mixed results with the combination of the LMs and GNNs on several molecule datasets. In particular, we found evidence for improvement on the FreeSolv and QM7 benchmarks, but degraded performance on the ESOL, LIPO and QM9 datasets compared to a GNN baseline.

Francel, Collin [University of Alabama]↗
DOE OSTI2025

Model-free estimation of completeness, uncertainties, and outliers in atomistic machine learning using information theory

Abstract An accurate description of information is relevant for a range of problems in atomistic machine learning (ML), such as crafting training sets, performing uncertainty quantification (UQ), or extracting physical insights from large datasets. However, atomistic ML often relies on unsupervised learning or model predictions to analyze information contents from simulation or training data. Here, we introduce a theoretical framework that provides a rigorous, model-free tool to quantify information contents in atomistic simulations. We demonstrate that the information entropy of a distribution of atom-centered environments explains known heuristics in ML potential developments, from training set sizes to dataset optimality. Using this tool, we propose a model-free UQ method that reliably predicts epistemic uncertainty and detects out-of-distribution samples, including rare events in systems such as nucleation. This method provides a general tool for data-driven atomistic modeling and combines efforts in ML, simulations, and physical explainability.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Internal calibration of transient kinetic data via machine learning

The temporal analysis of products (TAP) reactor provides a vast amount of transient kinetic information that may be used to describe a variety of chemical features including residence time distributions, kinetic coefficients, number of active sites, reaction mechanism, etc. However, as with any measurement device, the TAP reactor signal is convoluted with noise and drift is common. In order to reduce the uncertainty of the kinetic measurement and any derived parameters or mechanisms, proper preprocessing must be performed prior to any advanced type of analysis. This preprocessing includes baseline correction, i.e., a shift in the voltage response, and calibration, i.e., a scaling of the flux response based on prior experiments. The traditional methodology of preprocessing requires significant user discretion and reliance on separate calibration experiments that may drift over time. Herein we use machine learning techniques combined with physical constraints to understand the noise and drift that is being generated within and between experiments for enhancement of the chemical kinetic signal. As such, the proposed methodology demonstrates clear benefits over the traditional preprocessing approach by eliminating the need for separate calibration experiments or heuristic input from the user.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Chemical Signatures of Seasonally Unique Anthropogenic Influences on Organic Aerosol Composition in the Central Amazon

Urbanization and fires perturb the quantities and composition of fine organic aerosol in the central Amazon, with ramifications for radiative forcing and public health. These disturbances include not only direct emissions of particulates and secondary organic aerosol (SOA) precursors, but also changes in the pathways through which biogenic precursors form SOA. The composition of ambient organic aerosol is complex and incompletely characterized, encompassing millions of potential structures relatively few of which have been synthesized and characterized. Through analysis of submicron aerosol samples from the Green Ocean Amazon (GoAmazon2014/5) field campaign by two-dimensional gas chromatography coupled with machine learning, ~1300 unique compounds were traced and characterized over two seasons. Fires and urban emissions produced chemically and interseasonally distinct impacts on product signatures, with only ~50% of compounds observed in both seasons. Seasonally unique populations point to the importance of aqueous processing in Amazonian aerosol ageing, but further mechanistic insights are impeded by limited product identity knowledge. Further, less than 10% of compounds were identifiable at an isomer-specific level. Overall, the findings (i) provide compositional characterization of anthropogenic influence on submicron organic aerosol in the Amazon, (ii) identify key season-to-season differences in chemical signatures, and (iii) highlight high-priority knowledge gaps in current speciated knowledge.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Bias free multiobjective active learning for materials design and discovery

The design rules for materials are clear for applications with a single objective. For most applications, however, there are often multiple, sometimes competing objectives where there is no single best material and the design rules change to finding the set of Pareto optimal materials. In this work, we leverage an active learning algorithm that directly uses the Pareto dominance relation to compute the set of Pareto optimal materials with desirable accuracy. We apply our algorithm to de novo polymer design with a prohibitively large search space. Using molecular simulations, we compute key descriptors for dispersant applications and drastically reduce the number of materials that need to be evaluated to reconstruct the Pareto front with a desired confidence. This work showcases how simulation and machine learning techniques can be coupled to discover materials within a design space that would be intractable using conventional screening approaches.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗