Search NASA⌕ Search

SEARCH · Search NASA

Results for “Genetic prediction”

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 343 records · Page 19
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Populus_trichocarpa_Breeding_Population_SNPs

These data are from the manuscript “Application of Genomic Prediction in a Populus trichocarpa Breeding Program”, by Brian J. Stanton, David Macaya-Sanz, Chanaka Roshan Abeyratne, David Kainer, Kathy Haiby, Austin Himes, Carlos Gantz, Gerald A. Tuskan, and Stephen P. DiFazio. The data are based on genome resequencing to approximately 10X depth on two collections of Populus trichocarpa trees from Oregon, Washington, California, and British Columbia. The first collection consists of 293 genets collected by Poplar Innovations LLC for a breeding program. The second collection consists of 961 trees collected for the purpose of genome-wide association studies. These genets were sequenced using short, paired-end Illumina sequence reads (Chhetri et al. 2019). Reads were aligned to the P. trichocarpa ′Stettler-14′ reference (Hofmeister et al. 2020), with minor modifications to correct mis-assemblies (Zhou et al. 2020), and variants were called as per methods described in (Abeyratne et al. 2023). Identified variants were filtered using GATK’s VariantFiltration tool (DePristo et al. 2011), with filter expression flag set to “AF < 0.01 || AF > 0.99 || QD < 10.0 || ExcessHet > 20.0 || FS > 10.0 || MQ < 58.0”. SNPs with severe departures from Hardy−Weinberg expectations (exact-test p< 0.01) were also removed using vcftools --hwe flag (Danecek et al. 2011), resulting in 15,627,211 bi-allelic SNPs. The data included here consist of 141,903 high quality bi-allelic genome-wide SNPs obtained by further filtering the original SNP dataset using vcftools with flags --maf 0.05, --max-maf 0.95, --max-missing 0.95, --min-meanDP 10.75, --max-meanDP 43.00, --thin 2000. Collectively, these filtering parameters removed SNPs with 1) a minor allele frequency ≤ 0.05; 2) proportion of missing data for individual loci exceeding 5%; 3) sequencing depth more than 2X mean-depth or less than 0.5X mean-depth; or 4) a distance of

09 BIOMASS FUELS↗
DOE OSTI2023

Characterization and Diversification of AraC/XylS Family Regulators Guided by Transposon Sequencing

In this study, we explored the development of engineered inducible systems. Publicly available data from previous transposon sequencing assays were used to identify regulators of metabolism in Pseudomonas putida KT2440. For AraC family regulators (AFRs) represented in these data, we posited AFR/promoter/inducer groupings. Twelve promoters were characterized for a response to their proposed inducers in P. putida, and the resultant data were used to create and test nine two-plasmid sensor systems in Escherichia coli. Several of these were further developed into a palette of single-plasmid inducible systems. From these experiments, we observed an unreported inducer response from a previously characterized AFR, demonstrated that the addition of a P. putida transporter improved the sensor dynamics of an AFR in E. coli, and identified an uncharacterized AFR with a novel potential inducer specificity. Finally, targeted mutations in an AFR, informed by structural predictions, enabled the further diversification of these inducible plasmids.

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

High-throughput, dynamic, multi-dimensional: an expanding repertoire of plant respiration measurements

A recent burst of technological innovation and adaptation has greatly improved our ability to capture respiration rate data from plant sources. At the tissue level, several independent respiration measurement options are now available, each with distinct advantages and suitability, including high-throughput sampling capacity. These advancements facilitate the inclusion of respiration rate data into large-scale biological studies such as genetic screens, ecological surveys, crop breeding trials, and multi-omics molecular studies. As a result, our understanding of the correlations of respiration with other biological and biochemical measurements is rapidly increasing. Difficult questions persist concerning the interpretation and utilization of respiration data; concepts such as allocation of respiration to growth versus maintenance, the unnecessary or inefficient use of carbon and energy by respiration, and predictions of future respiration rates in response to environmental change are all insufficiently grounded in empirical data. However, we emphasize that new experimental designs involving novel combinations of respiration rate data with other measurements will flesh-out our current theories of respiration. Furthermore, dynamic recordings of respiration rate, which have long been used at the scale of mitochondria, are increasingly being used at larger scales of size and time to reflect processes of cellular signal transduction and physiological response to the environment. Herein, we also highlight how respiratory methods are being better adapted to different plant tissues including roots and seeds, which have been somewhat neglected historically.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Torque Teno Sus Virus 1: A Potential Surrogate Pathogen to Study Pig-Transmitted Transboundary Animal Diseases

Understanding the epidemiology and transmission dynamics of transboundary animal diseases (TADs) among wild pigs (Sus scrofa) will aid in preventing the introduction or containment of TADs among wild populations. Given the challenges associated with studying TADs in free-ranging populations, a surrogate pathogen system may predict how pathogens may circulate and be maintained within wild free-ranging swine populations, how they may spill over into domestic populations, and how management actions may impact transmission. We assessed the suitability of Torque teno sus virus 1 (TTSuV1) to serve as a surrogate pathogen for molecular epidemiological studies in wild pigs by investigating the prevalence, persistence, correlation with host health status and genetic variability at two study areas: Archbold’s Buck Island Ranch in Florida and Savannah River Site in South Carolina. We then conducted a molecular epidemiological case study within Archbold’s Buck Island Ranch site to determine how analysis of this pathogen could inform transmission dynamics of a directly transmitted virus. Prevalence was high in both study areas (40%, n = 190), and phylogenetic analyses revealed high levels of genetic variability within and between study areas. Our case study showed that pairwise host relatedness and geographic distance were highly correlated to pairwise viral genetic similarity. Molecular epidemiological analyses revealed a distinct pattern of direct transmission from pig to pig occurring within and between family groups. Our results suggest that TTSuV1 is highly suitable for molecular epidemiological analyses and will be useful for future studies of transmission dynamics in wild free-ranging pigs.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

A novel active optimization approach for rapid and efficient design space exploration using ensemble machine learning

In this work, a novel design optimization technique based on active learning, which involves dynamic exploration and exploitation of the design space of interest using an ensemble of machine learning algorithms, is presented. In this approach, a hybrid methodology incorporating an explorative weak learner (regularized basis function model) which fits high-level information about the response surface, and an exploitative strong learner (based on committee machine) that fits finer details around promising regions identified by the weak learner, is employed. For each design iteration, an aristocratic approach is used to select a set of nominees, where points that meet a threshold merit value as predicted by the weak learner are selected to be evaluated using expensive function evaluation. In addition to these points, the global optimum as predicted by the strong learner is also evaluated to enable rapid convergence to the actual global optimum once the most promising region has been identified by the optimizer. This methodology is first tested by applying it to the optimization of a two-dimensional multi-modal surface. The performance of the new active learning approach is compared with traditional global optimization methods, namely micro-genetic algorithm (pGA) and particle swarm optimization (PSO). It is demonstrated that the new optimizer is able to reach the global optimum much faster with a significantly fewer number of function evaluations. Subsequently, the new optimizer is also applied to a complex internal combustion (IC) engine combustion optimization case with nine control parameters related to fuel injection, initial thermodynamic conditions, and in-cylinder flow. It is again found that the new approach significantly lowers the number of function evaluations that are needed to reach the optimum design configuration (by up to 80%) when compared to particle swarm and genetic algorithm-based optimization techniques.

Machine learning, active learning, design optimiza↗
DOE OSTI2024

High-throughput protein characterization by complementation using DNA barcoded fragment libraries

Abstract Our ability to predict, control, or design biological function is fundamentally limited by poorly annotated gene function. This can be particularly challenging in non-model systems. Accordingly, there is motivation for new high-throughput methods for accurate functional annotation. Here, we used co mplementation of aux otrophs and DNA barcode seq uencing (Coaux-Seq) to enable high-throughput characterization of protein function. Fragment libraries from eleven genetically diverse bacteria were tested in twenty different auxotrophic strains of Escherichia coli to identify genes that complement missing biochemical activity. We recovered 41% of expected hits, with effectiveness ranging per source genome, and observed success even with distant E. coli relatives like Bacillus subtilis and Bacteroides thetaiotaomicron . Coaux-Seq provided the first experimental validation for 53 proteins, of which 11 are less than 40% identical to an experimentally characterized protein. Among the unexpected function identified was a sulfate uptake transporter, an O-succinylhomoserine sulfhydrylase for methionine synthesis, and an aminotransferase. We also identified instances of cross-feeding wherein protein overexpression and nearby non-auxotrophic strains enabled growth. Altogether, Coaux-Seq’s utility is demonstrated, with future applications in ecology, health, and engineering.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Shared governance in the plant holobiont and implications for one health

Abstract The holobiont Holobiont theory is more than 80 years old, while the importance of microbial communities for plant holobionts was already identified by Lorenz Hiltner more than a century ago. Both concepts are strongly supported by results from the new field of microbiome research. Here, we present ecological and genetic features of the plant holobiont that underpin principles of a shared governance between hosts and microbes and summarize the relevance of plant holobionts in the context of global change. Moreover, we uncover knowledge gaps that arise when integrating plant holobionts in the broader perspective of the holobiome as well as one and planetary health concepts. Action is needed to consider interacting holobionts at the holobiome scale, for prediction and control of microbiome function to improve human and environmental health outcomes.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Predicting interface structure using the minima hopping method

Here, we adapt the minima hopping method (MHM) to the problem of interfacial structure prediction and apply it to study a canonical problem, the tilt grain boundaries in SrTiO 3 . Our method employs a hybrid approach by first exploring the potential energy surface (PES) of different grain boundary samplings with an empirical force field, among which the fifteen candidates with lower energies are then refined using ab initio density functional theory (DFT) calculations. During the exploratory stage, we bias the search using a local order parameter to primarily sample various reconstructions in the vicinity of the interface, while preserving the crystallinity of the bulk regions. We further enhance the search by incorporating initial structures with rigid body displacements to account for translational variations between bulk phases, enabling the MHM to effectively generate both stoichiometric and nonstoichiometric SrTiO 3 Σ⁢3(111)[110] and Σ⁢3(112)[110] grain boundaries. From an algorithmic standpoint, MHM outperforms earlier studies based on genetic algorithms (GA) by identifying more stable interfacial structures of several SrTiO 3 grain boundaries. The performance of the present implementation of the MHM approach is primarily limited by exploring an approximate description of the PES with a rather simple Buckingham potential. This limitation leads to variations in performance when compared to approaches utilizing more advanced surrogate PES models, such as direct DFT-PES sampling or GA with the embedded atom method (EAM). Despite the present limitations, the MHM approach is able to yield interfacial structures with comparable or lower interfacial energies in specific cases, such as Σ⁢3(111)[110] Γ=1, ±0.5 and Σ⁢3(112)[110] Γ= ±1, −2, underscoring the robustness of the MHM approach even with a simple approximation of the DFT PES. The MHM interfacial structure prediction method thus offers an efficient approach to understanding the grain boundaries and heterointerfaces at the atomic scale, providing an important prerequisite for effective materials design.

density functional theory↗
DOE OSTI2020

Deep neural network improves the estimation of polygenic risk scores for breast cancer

Polygenic risk scores (PRS) estimate the genetic risk of an individual for a complex disease based on many genetic variants across the whole genome. Here, we compared a series of computational models for estimation of breast cancer PRS. A deep neural network (DNN) was found to outperform alternative machine learning techniques and established statistical algorithms, including BLUP, BayesA, and LDpred. In the test cohort with 50% prevalence, the Area Under the receiver operating characteristic Curve (AUC) were 67.4% for DNN, 64.2% for BLUP, 64.5% for BayesA, and 62.4% for LDpred. BLUP, BayesA, and LPpred all generated PRS that followed a normal distribution in the case population. However, the PRS generated by DNN in the case population followed a bimodal distribution composed of two normal distributions with distinctly different means. This suggests that DNN was able to separate the case population into a high-genetic-risk case subpopulation with an average PRS significantly higher than the control population and a normal-genetic-risk case subpopulation with an average PRS similar to the control population. This allowed DNN to achieve 18.8% recall at 90% precision in the test cohort with 50% prevalence, which can be extrapolated to 65.4% recall at 20% precision in a general population with 12% prevalence. Interpretation of the DNN model identified salient variants that were assigned insignificant p values by association studies, but were important for DNN prediction. These variants may be associated with the phenotype through nonlinear relationships.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Quantile-dependent expressivity of plasma adiponectin concentrations may explain its sex-specific heritability, gene-environment interactions, and genotype-specific response to postprandial lipemia

“Quantile-dependent expressivity” occurs when the effect size of a genetic variant depends upon whether the phenotype (e.g. adiponectin) is high or low relative to its distribution. We have previously shown that the heritability ( h 2 ) of adiposity, lipoproteins, postprandial lipemia, pulmonary function, and coffee and alcohol consumption are quantile-specific. Whether adiponectin heritability is quantile specific remains to be determined. Plasma adiponectin concentrations from 4,182 offspring-parent pairs and 1,662 sibships from the Framingham Heart Study were analyzed. Quantile-specific heritability from offspring-parent (β OP , h 2 = 2 β OP /(1 + r spouse )) and full-sib regression slopes ( β FS , h 2 = {(1 + 8r spouse β FS ) 0.05 -1}/(2r spouse )) were robustly estimated by quantile regression with nonparametric significance assigned from 1,000 bootstrap samples. Quantile-specific h 2 (± SE) increased with increasing percentiles of the offspring’s age- and sex-adjusted adiponectin distribution when estimated from β OP ( P trend = 2.2 × 10 -6 ): 0.30 ± 0.03 at the 10th, 0.33 ± 0.04 at the 25th, 0.43 ± 0.04 at the 50th, 0.55 ± 0.05 at the 75th, and 0.57 ± 0.08 at the 90th percentile, and when estimated from β FS (P trend = 7.6 × 10 -7 ): 0.42 ± 0.03 at the 10th, 0.44 ± 0.04 at the 25th, 0.56 ± 0.05 at the 50th, 0.73 ± 0.08 at the 75th, and 0.79 ± 0.11 at the 90th percentile. Consistent with quantile-dependent expressivity, adiponectin’s: (1) heritability was greater in women in accordance with their higher adiponection concentrations; (2) relationships to ADIPOQ polymorphisms were modified by adiposity in accordance with its adiponectin-lowering effect; (3) response to rosiglitazone was predicted by the 45T> G ADIPOQ polymorphism; (4) difference by ADIPOQ haplotypes increased linearly with increasing postprandial adiponectin concentrations. Adiponectin heritability is quantile dependent, which may explain sex-specific heritability, gene-environment and gene-drug interactions, and postprandial response by haplotypes.

Postprandial↗
DOE OSTI2023

In silico analyses of diversity and dissemination of antimicrobial resistance genes and mobile genetics elements, for plasmids of enteric pathogens

The antimicrobial resistance (AMR) mobilome plays a key role in the dissemination of resistance genes encoded by mobile genetics elements (MGEs) including plasmids, transposons (Tns), and insertion sequences (ISs). These MGEs contribute to the dissemination of multidrug resistance (MDR) in enteric bacterial pathogens which have been considered as a global public health risk. To further understand the diversity and distribution of AMR genes and MGEs across different plasmid types, we utilized multiple sequence-based computational approaches to evaluate AMR-associated plasmid genetics. A collection of 1,309 complete plasmid sequences from Gammaproteobacterial species, including 100 plasmids from each of the following 14 incompatibility (Inc) types: A/C, BO, FIA, FIB, FIC, FIIA, HI1, HI2, I1, K, M, N, P except W, where only 9 sequences were available, was extracted from the National Center for Biotechnology Information (NCBI) GenBank database using BLAST tools. The extracted FASTA files were analyzed using the AMRFinderPlus web-based tools to detect antimicrobial, disinfectant, biocide, and heavy metal resistance genes and ISFinder to identify IS/Tn MGEs within the plasmid sequences. In silico prediction based on plasmid replicon types showed that the resistance genes were diverse among plasmids, yet multiple genes were widely distributed across the plasmids from enteric bacterial species. These findings provide insights into the diversity of resistance genes and that MGEs mediate potential transmission of these genes across multiple plasmid replicon types. This notion was supported by the observation that many IS/Tn MGEs and resistance genes known to be associated with them were common across multiple different plasmid types. Our results provide critical insights about how the diverse population of resistance genes that are carried by the different plasmid types can allow for the dissemination of AMR across enteric bacteria. The results also highlight the value of computational-based approaches and in silico analyses for the assessment of AMR and MGEs, which are important elements of molecular epidemiology and public health outcomes.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Controlling the Self-Assembly of DNA Origami Octahedra via Manipulation of Inter-Vertex Interactions

Recent studies have demonstrated novel strategies for the organization of nanomaterials into three-dimensional (3D) ordered arrays with prescribed lattice symmetries using DNA-based self-assembly strategies. In one approach, the nanomaterial is sequestered into DNA origami frames or “material voxels” and then coordinated into ordered arrays based on the voxel geometry and the corresponding directional interactions based on its valency. While the lattice symmetry is defined by the valency of the bonds, a larger-scale morphological development is affected by assembly processes and differences in energies of anisotropic bonds. To facilely model this assembly process, we investigate the self-assembly behavior of hard particles with six interacting vertices via theory and Monte Carlo simulations and exploration of corresponding experimental systems. We demonstrate that assemblies with different 3D crystalline morphologies, but the same lattice symmetry can be formed depending on the relative strength of vertex-to-vertex interactions in orthogonal directions. We observed three distinct assembly morphologies for such systems: cube-like, sheet-like, and cylinder-like. A simple analytical theory inspired by well-established ideas in the areas of protein crystallization, based on calculating the second virial coefficient of patchy hard spheres, captures the simulation results and thus represents a straightforward means of modeling this self-assembly process. To complement the theory and simulations, experimental studies were performed to investigate the assembly of octahedral DNA origami frames with varying binding energies at their vertices. In conclusion, x-ray scattering confirms the robustness of the formed nanoscale lattices for different binding energies, while both optical and electron microscopy imaging validated the theoretical predictions on the dependence of the distinct morphologies of assembled state on the interaction strengths in the three orthogonal directions.

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

Biosystems Design by Machine Learning

Biosystems such as enzymes, pathways, and whole cells have been increasingly explored for biotechnological applications. Yet, the intricate connectivity and complexity of biosystems pose a major hurdle in designing biosystems with desired features. As -omics and other high throughput technologies have been rapidly developed, the promise of applying machine learning (ML) techniques in biosystems design has started to become a reality. ML models enable the identification of patterns within complicated biological data across multiple scales of analysis and can augment biosystems design applications by predicting new candidates for optimized performance. ML is being used at every stage of biosystems design to help find non-obvious engineering solutions with fewer design iterations. In this review, we first describe commonly used models and modeling paradigms within ML. We then discuss some applications of these models that have already shown success in biotechnological applications. Moreover, we discuss successful applications at all scales of biosystems design, including nucleic acids, genetic circuits, proteins, pathways, genomes, and bioprocess. Lastly, we discuss some limitations of these methods and potential solutions as well as prospects of the combination of ML and biosystems design.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Harnessing the predicted maize pan-interactome for putative gene function prediction and prioritization of candidate genes for important traits

Abstract The recent assembly and annotation of the 26 maize nested association mapping population founder inbreds have enabled large-scale pan-genomic comparative studies. These studies have expanded our understanding of agronomically important traits by integrating pan-transcriptomic data with trait-specific gene candidates from previous association mapping results. In contrast to the availability of pan-transcriptomic data, obtaining reliable protein–protein interaction (PPI) data has remained a challenge due to its high cost and complexity. We generated predicted PPI networks for each of the 26 genomes using the established STRING database. The individual genome-interactomes were then integrated to generate core- and pan-interactomes. We deployed the PPI clustering algorithm ClusterONE to identify numerous PPI clusters that were functionally annotated using gene ontology (GO) functional enrichment, demonstrating a diverse range of enriched GO terms across different clusters. Additional cluster annotations were generated by integrating gene coexpression data and gene description annotations, providing additional useful information. We show that the functionally annotated PPI clusters establish a useful framework for protein function prediction and prioritization of candidate genes of interest. Our study not only provides a comprehensive resource of predicted PPI networks for 26 maize genomes but also offers annotated interactome clusters for predicting protein functions and prioritizing gene candidates. The source code for the Python implementation of the analysis workflow and a standalone web application for accessing the analysis results are available at https://github.com/eporetsky/PanPPI.

Genetics & Heredity↗
DOE OSTI2024

Diverse signatures of convergent evolution in cactus-associated yeasts

Many distantly related organisms have convergently evolved traits and lifestyles that enable them to live in similar ecological environments. However, the extent of phenotypic convergence evolving through the same or distinct genetic trajectories remains an open question. Here, we leverage a comprehensive dataset of genomic and phenotypic data from 1,049 yeast species in the subphylum Saccharomycotina (Kingdom Fungi, Phylum Ascomycota) to explore signatures of convergent evolution in cactophilic yeasts, ecological specialists associated with cacti. We inferred that the ecological association of yeasts with cacti arose independently approximately 17 times. Using a machine learning–based approach, we further found that cactophily can be predicted with 76% accuracy from both functional genomic and phenotypic data. The most informative feature for predicting cactophily was thermotolerance, which we found to be likely associated with altered evolutionary rates of genes impacting the cell envelope in several cactophilic lineages. We also identified horizontal gene transfer and duplication events of plant cell wall–degrading enzymes in distantly related cactophilic clades, suggesting that putatively adaptive traits evolved independently through disparate molecular mechanisms. Notably, we found that multiple cactophilic species and their close relatives have been reported as emerging human opportunistic pathogens, suggesting that the cactophilic lifestyle—and perhaps more generally lifestyles favoring thermotolerance—might preadapt yeasts to cause human disease. This work underscores the potential of a multifaceted approach involving high-throughput genomic and phenotypic data to shed light onto ecological adaptation and highlights how convergent evolution to wild environments could facilitate the transition to human pathogenicity.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Telomere-to-telomere assemblies of chromosome 10 reveal complex adaptive variation of 3-ketoacyl-CoA-synthases in Populus trichocarpa likely driven by Helitrons

The model woody plant Populus trichocarpa displays an atypical alkene-diverse wax cuticle likely driven by copy number variation (CNV) of 3-ketoacyl-CoA synthases ( KCS ), which has been difficult to confirm with short-read assemblies. Long-read sequencing enables the development of telomere-to-telomere resources to detect cryptic variation, including CNVs, which are currently missed. Integrating this information can improve genomic prediction for breeding and provide insights into the evolutionary basis of important traits. Our analysis of 78 long-read haplotypes from chromosome 10 identified more than twice as many KCS genes as previously reported, and numerous intragenic non-synonymous substitutions. Random Forest predictive models highlighted the importance of Potri.010G079500 in producing very long chain alkenes; however, its absence did not predict previously reported alkene-deficient phenotypes. Instead, alkene levels are best predicted by the combinations of KCS copies. Additionally, amino acid substitutions clustered around ligand and donor binding pockets, suggesting they contribute to differing wax cuticle composition. Finally, each KCS gene and copy was linked to a Helitron transposon. A phylogenetic analysis suggests Helitrons are the evolutionary mechanism for generating KCS tandem arrays. Long-read generated telomere-to-telomere assemblies of P. trichocarpa chromosome 10 revealed large-effect loci critical to genetic studies that are unattainable from short-reads. This new resource produced novel insights into genome structure and function, and a novel mechanism for generating tandem gene duplication. Our results highlight that, given current challenges in annotation and assembly, detailed and focused long-read sequences are key to interpreting complex genomic regions that contain tandem copy number variants.

09 BIOMASS FUELS↗
DOE OSTI2025

Hybrid data-driven cement-stabilized soil design: An integration of machine learning, multi-objective optimization, and life cycle assessment

Soil stabilization is crucial in geotechnical engineering, yet conventional methods are often time-consuming, resource-intensive, and environmentally unsustainable. Despite growing interest in Machine Learning (ML) and optimization tools for mix design, few studies integrate these methods with decision-making techniques and environmental assessment to support practical implementation. This study proposes a hybrid data-driven framework for predicting strength, optimizing mix compositions, and evaluating environmental impacts via life cycle assessment of cement-stabilized soft soils. Six ML models were evaluated, and the top-performing eXtreme Gradient Boosting (XGB) model was further improved using the Grey Wolf Optimizer (GWO). The optimized XGB-GWO model, integrated with a polynomial cost function, served as the objective function in a multi-objective optimization problem solved via the Non-Dominated Sorting Genetic Algorithm II (NSGA-II), with final mix selection guided by the entropy-weighted TOPSIS method. Validation through a case study produced mix designs offering superior strength-cost trade-offs, with the optimal mix achieving 2243.2 kPa unconfined compressive strength and a 16.07 % reduction in carbon emissions compared to the highest-cost design. In conclusion, this study offers a sustainable, scalable approach to soil stabilization and supports informed decision-making in construction.

Life cycle assessment↗