Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “compound screening”

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 OSTI2025

Light output and neutron detection efficiency of boron-based neutron scintillator screens for neutron imaging

Recent research has explored the development of boron-based neutron scintillator screens, which potentially offer improved spatial resolution and neutron capture efficiency compared to traditional lithium-based screens. This work builds upon previous efforts to improve boron-based neutron scintillators by assessing a newer generation of boron-based scintillator screens fabricated using different compositions and fabrication approaches compared to previous generations of screens. Some of the test screens exhibit higher light output than previous efforts and higher neutron capture efficiency than lithium-based screens. This paper describes the current state of screen development, measurement results for the most recent generation of screens, and future activities.

46 - INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AN↗
DOE OSTI2025

Thorium Bis‐Salophen Trimers as Anion Detectors and Binding Agents

Bis‐salophen ligands are the condensation product of a tetramine and a salicylaldehyde derivative. They feature two binding sites, both of which are tetradentate with mixed O/N donor atoms. Reaction with the ligand precursor and thorium nitrate tetrahydrate forms a 3:3 metal‐to‐ligand trimer with a ΔΔΔ‐chirality confirmed by X‐ray crystallography of a racemic single crystal. The structure has a pore in the center of the triangular structure measuring 6.22 Å at its narrowest point. This compound is air‐ and water‐stable as well as soluble in organic solvents. Here, this compound was screened with a series of tetrabutyl ammonium halide salts, and tetrabutylammonium (TBA) chloride showed the strongest binding to the complex. After the addition of 2 equivalents of TBACl, the complex and salt precipitate out of solution.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Detection and quantification of γ-H2AX using a dissociation enhanced lanthanide fluorescence immunoassay

Phosphorylation of the histone protein H2AX to form γ-H2AX foci directly represents DNA double-strand break formation. Traditional γ-H2AX detection involves counting individual foci within individual nuclei. The novelty of this work is the application of a time-resolved fluorescence assay using dissociation-enhanced lanthanide fluorescence immunoassay for quantitative measurements of γ-H2AX. For comparison, standard fluorescence detection was employed and analyzed either by bulk fluorescent measurements or by direct foci counting using BioTek Spot Count algorithm and Gen 5 software. Etoposide induced DNA damage in A549 carcinoma cells was compared across all test platforms. Time resolved fluorescence detection of europium as a chelated complex enabled quantitative measurement of γ-H2AX foci with nanomolar resolution. Comparative bulk fluorescent signals achieved only micromolar sensitivity. Lanthanide based immunodetection of γ-H2AX offers superior detection and a user-friendly workflow. These approaches have the potential to improve screening of compounds that either enhance DNA damage or protect against its deleterious effects.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Structure-Aware Unsupervised, Transformational Machine Learning for Drug Discovery (DTRA Basic Research Final Report)

The major goal of this project is to develop machine learning (ML) methods to enable improved predictive power on real drug discovery for novel targets. More specifically, we planned to demonstrate the capability and effectiveness of ML tools utilizing unlabeled large-volume protein-ligand datasets. We investigated multiple pre-training approaches for 3D protein-ligand structure-based foundation models, without relying on experimental binding data. We also addressed scenarios in which crystal structures are unavailable or binding data are limited. We also planned to develop a complete pipeline to screen novel compounds as well as to demonstrate the capability and effectiveness of the developed methods by testing on a realistic drug discovery task such as SARS-CoV-2. While the major goals and milestones remain consistent with the original proposal, certain technical details have been adjusted, based on the experimental results and related outcomes.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Expanding the Domain of Applicability of Machine Learning Models with Limited Data for Drug Property Prediction

Accurate machine learning models for predicting small molecule interactions with biological targets are essential for therapeutic discovery, biothreat response, and computational drug design, but their performance is often limited for understudied targets with sparse experimental data. To address this challenge, we developed and evaluated methods to improve molecular property prediction under low-data conditions, using the NimA-related kinase (NEK) family as a proof-of-concept. This work focused on two complementary goals within the ATOM Modeling PipeLine (AMPL) and the Generative Molecular Design (GMD) loop: expanding model applicability through transfer learning, representation learning, feature scaling, sampling strategies, and active-learning-inspired compound selection; and enabling efficient virtual screening to prioritize compounds that balance predicted activity, design objectives, and synthetic accessibility.

organic↗
DOE OSTI2023

AI-accelerated protein-ligand docking for SARS-CoV-2 is 100-fold faster with no significant change in detection

Protein-ligand docking is a computational method for identifying drug leads. The method is capable of narrowing a vast library of compounds down to a tractable size for downstream simulation or experimental testing and is widely used in drug discovery. While there has been progress in accelerating scoring of compounds with artificial intelligence, few works have bridged these successes back to the virtual screening community in terms of utility and forward-looking development. We demonstrate the power of high-speed ML models by scoring 1 billion molecules in under a day (50 k predictions per GPU seconds). We showcase a workflow for docking utilizing surrogate AI-based models as a pre-filter to a standard docking workflow. Our workflow is ten times faster at screening a library of compounds than the standard technique, with an error rate less than 0.01% of detecting the underlying best scoring 0.1% of compounds. Our analysis of the speedup explains that another order of magnitude speedup must come from model accuracy rather than computing speed. In order to drive another order of magnitude of acceleration, we share a benchmark dataset consisting of 200 million 3D complex structures and 2D structure scores across a consistent set of 13 million “in-stock” molecules over 15 receptors, or binding sites, across the SARS-CoV-2 proteome. We believe this is strong evidence for the community to begin focusing on improving the accuracy of surrogate models to improve the ability to screen massive compound libraries 100 × or even 1000 × faster than current techniques and reduce missing top hits. The technique outlined aims to be a fast drop-in replacement for docking for screening billion-scale molecular libraries.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Actinomycetota isolated from the sponge Hymeniacidon perlevis as a source of novel compounds with pharmacological applications: diversity, bioactivity screening, and metabolomic analysis

Abstract Aims To combat health conditions, such as multi-resistant bacterial infections, cancer, and metabolic diseases, new drugs need to be urgently found and, in this respect, marine Actinomycetota have a high potential to produce secondary metabolites with pharmacological importance. We aimed to study the cultivable Actinomycetota community associated with a marine sponge from the Portuguese coast, Hymeniacidon perlevis, and investigate the potential of the retrieved isolates to produce compounds with antimicrobial, anticancer and anti-obesity properties. Methods and results The analysis of the 16S rRNA gene revealed 79 Actinomycetota isolates affiliated with 12 genera—Brachybacterium, Dietzia, Glutamicibacter, Gordonia, Micrococcus, Micromonospora, Nocardia, Nocardiopsis, Paenoartrhobacter, Rhodococcus, Streptomyces, and Tsukamurella, most of which affiliated with the genus Streptomyces. The screening of antimicrobial activity revealed 13 strains, all belonging to the Streptomyces genus, capable of inhibiting the growth of Candida albicans, Bacillus subtilis, or Staphylococcus aureus. Forty-three extracts exhibited cytotoxic activity against at least one tested cell line (HepG2, HCT-116, and hCMEC-D3). Three extracts that were active against the two cancer cell lines tested, did not reduce the viability of the non-cancer endothelial cell line, hCMEC-D3. One Gordonia strain exhibited anti-obesity activity, revealed by its ability to reduce the neutral lipids in zebrafish larvae. Mass spectrometry-based dereplication analysis of active extracts identified several compounds associated with known Actinomycetota natural products. Nonetheless, five clusters contained metabolites that did not match any annotated natural products, suggesting they may represent new bioactive molecules. Conclusions This work contributed to increase the knowledge on the diversity and bioactive potential of Actinomycetota associated with H. perlevis.

Fonseca, Ana C.↗
DOE OSTI2025

Chemical classification program synthesis using generative artificial intelligence

Accurately classifying chemical structures is essential for cheminformatics and bioinformatics, including tasks such as identifying bioactive compounds of interest, screening molecules for toxicity to humans, finding non-organic compounds with desirable material properties, or organizing large chemical libraries for drug discovery or environmental monitoring. However, manual classification is labor-intensive and difficult to scale to large chemical databases. Existing automated approaches either rely on manually constructed classification rules, or are deep learning methods that lack explainability. This work presents an approach that uses generative artificial intelligence to automatically write chemical classifier programs for classes in the Chemical Entities of Biological Interest (ChEBI) database. These programs can be used for efficient deterministic run-time classification of SMILES structures, with natural language explanations. The programs themselves constitute an explainable computable ontological model of chemical class nomenclature, which we call the ChEBI Chemical Class Program Ontology (C3PO). We validated our approach against the ChEBI database, and compared our results against deep learning models and a naive SMARTS pattern based classifier. C3PO outperforms the naive classifier, but does not reach the performance of state of the art deep learning methods. However, C3PO has a number of strengths that complement deep learning methods, including explainability and reduced data dependence. C3PO can be used alongside deep learning classifiers to provide an explanation of the classification, where both methods agree. The programs can be used as part of the ontology development process, and iteratively refined by expert human curators.

Artificial Intelligence↗
DOE OSTI2023

Advancing energy storage through solubility prediction: leveraging the potential of deep learning

Solubility prediction plays a crucial role in energy storage applications, such as redox flow batteries, because it directly affects the efficiency and reliability. Researchers have developed various methods that utilize quantum calculations and descriptors to predict the aqueous solubilities of organic molecules. Notably, machine learning models based on descriptors have shown promise for solubility prediction. As deep learning tools, graph neural networks (GNNs) have emerged to capture complex structure–property relationships for material property prediction. Specifically, MolGAT, a type of GNN model, was designed to incorporate n-dimensional edge attributes, enabling the modeling of intricacies in molecular graphs and enhancing the prediction capabilities. In a previous study, MolGAT successfully screened 23 467 promising redox-active molecules from a database of over 500 000 compounds, based on redox potential predictions. This study focused on applying the MolGAT model to predict the aqueous solubility (log S) of a broad range of organic compounds, including those previously screened for redox activity. The model was trained on a diverse sample of 8494 organic molecules from AqSolDB and benchmarked against literature data, demonstrating superior accuracy compared with other state of the art graph-based and descriptor-based models. Subsequently, the trained MolGAT model was employed to screen redox-active organic compounds identified in the first phase of high-throughput virtual screening, targeting favorable solubility in energy storage applications. The second round of screening, which considered solubility, yielded 12 332 promising redox-active and soluble organic molecules suitable for use in aqueous redox flow batteries. Thus, the two-phase high-throughput virtual screening approach utilizing MolGAT, specifically trained for redox potential and solubility, is an effective strategy for selecting suitable intrinsically soluble redox-active molecules from extensive databases, potentially advancing energy storage through reliable material development. This indicates that the model is reliable for predicting the solubility of various molecules and provides valuable insights for energy storage, pharmaceutical, environmental, and chemical applications.

25 ENERGY STORAGE↗
DOE OSTI2024

An artificial intelligence accelerated virtual screening platform for drug discovery

Abstract Structure-based virtual screening is a key tool in early drug discovery, with growing interest in the screening of multi-billion chemical compound libraries. However, the success of virtual screening crucially depends on the accuracy of the binding pose and binding affinity predicted by computational docking. Here we develop a highly accurate structure-based virtual screen method, RosettaVS, for predicting docking poses and binding affinities. Our approach outperforms other state-of-the-art methods on a wide range of benchmarks, partially due to our ability to model receptor flexibility. We incorporate this into a new open-source artificial intelligence accelerated virtual screening platform for drug discovery. Using this platform, we screen multi-billion compound libraries against two unrelated targets, a ubiquitin ligase target KLHDC2 and the human voltage-gated sodium channel Na V 1.7. For both targets, we discover hit compounds, including seven hits (14% hit rate) to KLHDC2 and four hits (44% hit rate) to Na V 1.7, all with single digit micromolar binding affinities. Screening in both cases is completed in less than seven days. Finally, a high resolution X-ray crystallographic structure validates the predicted docking pose for the KLHDC2 ligand complex, demonstrating the effectiveness of our method in lead discovery.

Science & Technology - Other Topics↗
DOE OSTI2024

An expedited screening platform for the discovery of anti-ageing compounds in vitro and in vivo

Background: Restraining or slowing ageing hallmarks at the cellular level have been proposed as a route to increased organismal lifespan and healthspan. Consequently, there is great interest in anti-ageing drug discovery. However, this currently requires laborious and lengthy longevity analysis. Here, we present a novel screening readout for the expedited discovery of compounds that restrain ageing of cell populations in vitro and enable extension of in vivo lifespan. Methods: Using Illumina methylation arrays, we monitored DNA methylation changes accompanying long-term passaging of adult primary human cells in culture. This enabled us to develop, test, and validate the CellPopAge Clock, an epigenetic clock with underlying algorithm, unique among existing epigenetic clocks for its design to detect anti-ageing compounds in vitro. Additionally, we measured markers of senescence and performed longevity experiments in vivo in Drosophila, to further validate our approach to discover novel anti-ageing compounds. Finally, we bench mark our epigenetic clock with other available epigenetic clocks to consolidate its usefulness and specialisation for primary cells in culture. Results: We developed a novel epigenetic clock, the CellPopAge Clock, to accurately monitor the age of a population of adult human primary cells. We find that the CellPopAge Clock can detect decelerated passage-based ageing of human primary cells treated with rapamycin or trametinib, well-established longevity drugs. We then utilise the CellPopAge Clock as a screening tool for the identification of compounds which decelerate ageing of cell populations, uncovering novel anti-ageing drugs, torin2 and dactolisib (BEZ-235). We demonstrate that delayed epigenetic ageing in human primary cells treated with anti-ageing compounds is accompanied by a reduction in senescence and ageing biomarkers. Finally, we extend our screening platform in vivo by taking advantage of a specially formulated holidic medium for increased drug bioavailability in Drosophila. We show that the novel anti-ageing drugs, torin2 and dactolisib (BEZ-235), increase longevity in vivo. Conclusions: Our method expands the scope of CpG methylation profiling to accurately and rapidly detecting anti-ageing potential of drugs using human cells in vitro, and in vivo, providing a novel accelerated discovery platform to test sought after anti-ageing compounds and geroprotectors.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

Screening and Development of Sacrificial Cathode Additives for Lithium‐Ion Batteries

Abstract This work presents a computational screening approach to identify Li‐rich transition‐metal oxide sacrificial cathode additives and provides experimental validation of antifluorite‐structured Li 6 MnO 4 as a potential candidate. Initial attempts to synthesize this compound result in low purity (≤40% by weight) owing to close thermodynamic competition with Li 2 O and MnO at low temperature. However, it is shown that a much higher purity of 85% by weight can be achieved by combining Li excess with rapid cooling from high temperature, which effectively stabilizes the Li 6 MnO 4 phase. The synthesized product delivers a high irreversible Li release capacity that exceeds 700 mAh g −1 by utilizing combined Mn oxidation (Mn 2+/3+ and Mn 3+/4+ ) and O oxidation. These results demonstrate that Li 6 MnO 4 may therefore be useful as a potential sacrificial cathode additive in Li‐ion batteries and motivate further investigation of other structurally‐related compounds. While attempts were made to synthesize two additional compounds among computationally screened candidates, it was not successful to experimentally realize the two candidates. The difficulty of experimental realization of the newly predicted materials remains a challenge and it is suggested that more efforts need to be devoted to developing computational techniques to precisely predict synthesizability and propose potential synthetic routes of the predicted materials.

Kim, Haegyeom↗
DOE OSTI2022

Decoding the protein–ligand interactions using parallel graph neural networks

Abstract Protein–ligand interactions (PLIs) are essential for biochemical functionality and their identification is crucial for estimating biophysical properties for rational therapeutic design. Currently, experimental characterization of these properties is the most accurate method, however, this is very time-consuming and labor-intensive. A number of computational methods have been developed in this context but most of the existing PLI prediction heavily depends on 2D protein sequence data. Here, we present a novel parallel graph neural network (GNN) to integrate knowledge representation and reasoning for PLI prediction to perform deep learning guided by expert knowledge and informed by 3D structural data. We develop two distinct GNN architectures: $$\hbox {GNN}_{\mathrm{F}}$$ GNN F is the base implementation that employs distinct featurization to enhance domain-awareness, while $$\hbox {GNN}_{\mathrm{P}}$$ GNN P is a novel implementation that can predict with no prior knowledge of the intermolecular interactions. The comprehensive evaluation demonstrated that GNN can successfully capture the binary interactions between ligand and protein’s 3D structure with 0.979 test accuracy for $$\hbox {GNN}_{\mathrm{F}}$$ GNN F and 0.958 for $$\hbox {GNN}_{\mathrm{P}}$$ GNN P for predicting activity of a protein–ligand complex. These models are further adapted for regression tasks to predict experimental binding affinities and $$\hbox {pIC}_{\mathrm{50}}$$ pIC 50 crucial for compound’s potency and efficacy. We achieve a Pearson correlation coefficient of 0.66 and 0.65 on experimental affinity and 0.50 and 0.51 on $$\hbox {pIC}_{\mathrm{50}}$$ pIC 50 with $$\hbox {GNN}_{\mathrm{F}}$$ GNN F and $$\hbox {GNN}_{\mathrm{P}}$$ GNN P , respectively, outperforming similar 2D sequence based models. Our method can serve as an interpretable and explainable artificial intelligence (AI) tool for predicted activity, potency, and biophysical properties of lead candidates. To this end, we show the utility of $$\hbox {GNN}_{\mathrm{P}}$$ GNN P on SARS-Cov-2 protein targets by screening a large compound library and comparing the prediction with the experimentally measured data.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Screening a knowledge‐based library of low molecular weight compounds against the proline biosynthetic enzyme 1‐pyrroline‐5‐carboxylate 1 ( PYCR1)

Abstract Δ 1 ‐pyrroline‐5‐carboxylate reductase isoform 1 (PYCR1) is the last enzyme of proline biosynthesis and catalyzes the NAD(P)H‐dependent reduction of Δ 1 ‐pyrroline‐5‐carboxylate toL‐proline. High PYCR1 gene expression is observed in many cancers and linked to poor patient outcomes and tumor aggressiveness. The knockdown of thePYCR1gene or the inhibition of PYCR1 enzyme has been shown to inhibit tumorigenesis in cancer cells and animal models of cancer, motivating inhibitor discovery. We screened a library of 71 low molecular weight compounds (average MW of 131 Da) against PYCR1 using an enzyme activity assay. Hit compounds were validated with X‐ray crystallography and kinetic assays to determine affinity parameters. The library was counter‐screened against human Δ 1 ‐pyrroline‐5‐carboxylate reductase isoform 3 and proline dehydrogenase (PRODH) to assess specificity/promiscuity. Twelve PYCR1 and one PRODH inhibitor crystal structures were determined. Three compounds inhibit PYCR1 with competitive inhibition parameter of 100 μM or lower. Among these, (S)‐tetrahydro‐2H‐pyran‐2‐carboxylic acid (70 μM) has higher affinity than the current best tool compoundN‐formyl‐l‐proline, is 30 times more specific for PYCR1 over human Δ 1 ‐pyrroline‐5‐carboxylate reductase isoform 3, and negligibly inhibits PRODH. Structure‐affinity relationships suggest that hydrogen bonding of the heteroatom of this compound is important for binding to PYCR1. The structures of PYCR1 and PRODH complexed with 1‐hydroxyethane‐1‐sulfonate demonstrate that the sulfonate group is a suitable replacement for the carboxylate anchor. This result suggests that the exploration of carboxylic acid isosteres may be a promising strategy for discovering new classes of PYCR1 and PRODH inhibitors. The structure of PYCR1 complexed withl‐pipecolate and NADH supports the hypothesis that PYCR1 has an alternative function in lysine metabolism.

Biochemistry & Molecular Biology↗
DOE OSTI2023

Core-Level Spectroscopy with Hard and Soft X-rays on Phosphorus-Containing Compounds for Energy Conversion and Storage

The electronic properties of nine solid phosphorus (P)-containing compounds with varying oxidation states and chemical environments, including GaP (–III) , InP (–III) , red-P (0) , H 3 P (III) O 3 , Na 2 H 2 P 2 (IV) O 6 , H 3 P (V) O 4 , KH 2 P (V) O 4 , Na 2 HP (V) O 4 , and InP (V) O 4 , are investigated using X-ray absorption near-edge structure (XANES) spectroscopy in the hard (P K-edge) and soft X-ray regime (P L 2,3 - edge). We find shifts in the absorption-edge positions and correlate them with the ligands surrounding the P atom, likely causing a different core–hole interaction screening for different compounds. Complementing the experimental analysis, ab initio many-body calculations of XANES spectra provide insights into the excitonic nature of the observed spectral features and their impact on the electronic structure of the materials. Furthermore, we report on P K-edge XANES measurements on aqueous phosphorus-containing acids, including H 3 PO 3 , H 3 PO 4 , and their mixtures. At first sight, the spectra of the aqueous acids are similar to those of their solid counterparts. However, close inspection reveals a slight red shift of the absorption edge and the presence of fewer spectral features compared with spectra of the respective solids. Mixtures of aqueous acids display spectral features corresponding to the individual components, indicating the potential for speciation and quantification through fingerprinting.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

The structure of importin α and the nuclear localization peptide of ChREBP, and small compound inhibitors of ChREBP–importin α interactions

The carbohydrate response element binding protein (ChREBP) is a glucose-responsive transcription factor that plays a critical role in glucose-mediated induction of genes involved in hepatic glycolysis and lipogenesis. In response to fluctuating blood glucose levels ChREBP activity is regulated mainly by nucleocytoplasmic shuttling of ChREBP. Under high glucose ChREBP binds to importin α and importin β and translocates into the nucleus to initiate transcription. We have previously shown that the nuclear localization signal site (NLS) for ChREBP is bipartite with the NLS extending from Arg158 to Lys190. Here, we report the 2.5 Å crystal structure of the ChREBP-NLS peptide bound to importin α. The structure revealed that the NLS binding is monopartite, with the amino acid residues K171RRI174 from the ChREBP-NLS interacting with ARM2–ARM5 on importin α. We discovered that importin α also binds to the primary binding site of the 14-3-3 proteins with high affinity, which suggests that both importin α and 14-3-3 are each competing with the other for this broad-binding region (residues 117–196) on ChREBP. We screened a small compound library and identified two novel compounds that inhibit the ChREBP-NLS/importin α interaction, nuclear localization, and transcription activities of ChREBP. These candidate molecules support developing inhibitors of ChREBP that may be useful in treatment of obesity and the associated diseases.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Multivariate chemogenomic screening prioritizes new macrofilaricidal leads

Development of direct acting macrofilaricides for the treatment of human filariases is hampered by limitations in screening throughput imposed by the parasite life cycle. In vitro adult screens typically assess single phenotypes without prior enrichment for chemicals with antifilarial potential. We developed a multivariate screen that identified dozens of compounds with submicromolar macrofilaricidal activity, achieving a hit rate of >50% by leveraging abundantly accessible microfilariae. Adult assays were multiplexed to thoroughly characterize compound activity across relevant parasite fitness traits, including neuromuscular control, fecundity, metabolism, and viability. Seventeen compounds from a diverse chemogenomic library elicited strong effects on at least one adult trait, with differential potency against microfilariae and adults. Our screen identified five compounds with high potency against adults but low potency or slow-acting microfilaricidal effects, at least one of which acts through a novel mechanism. We show that the use of microfilariae in a primary screen outperforms model nematode developmental assays and virtual screening of protein structures inferred with deep learning. These data provide new leads for drug development, and the high-content and multiplex assays set a new foundation for antifilarial discovery.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Machine learning-enhanced design of lead-free halide perovskite materials using density functional theory

The investigation of emerging non-toxic perovskite materials has been undertaken to advance the fabrication of environmentally sustainable lead-free perovskite solar cells. This study introduces a machine learning methodology aimed at predicting innovative halide perovskite materials that hold promise for use in photovoltaic applications. The seven newly predicted materials are as follows: CsMnCl 4 , Rb 3 Mn 2 Cl 9 , Rb 4 MnCl 6 , Rb 3 MnCl 5 , RbMn 2 Cl 7 , RbMn 4 Cl 9 , and CsIn 2 Cl 7 . The predicted compounds are first screened using a machine learning approach, and their validity is subsequently verified through density functional theory calculations. CsMnCl 4 is notable among them, displaying a bandgap of 1.37 eV, falling within the Shockley-Queisser limit, making it suitable for photovoltaic applications. Here, through the integration of machine learning and density functional theory, this study presents a methodology that is more effective and thorough for the discovery and design of materials.

36 MATERIALS SCIENCE↗