Using flux theory in dynamic omics data sets to identify differentially changing signals using DPoP.

BACKGROUND: Derivative profiling is a novel approach to identify differential signals from dynamic omics data sets. This approach applies variable step-size differentiation to time dynamic omics data. This work assumes that there is a general omics derivative that is a useful and descriptive feature of dynamic omics experiments. We assert that this omics derivative, or omics flux, is a valuable descriptor that can be used instead of, or with, fold change calculations. RESULTS: The results of derivative profiling are compared to established methods such as Multivariate Adaptive Regression Splines, significance versus fold change analysis (Volcano), and an adjusted ratio over intensity (M/A) analysis to find that there is a statistically significant similarity between the results. This comparison is repeated for transcriptomic and phosphoproteomic expression profiles previously characterized in Aspergillus nidulans. This method has been packaged in an open-source, GUI-based MATLAB app, the Derivative Profiling omics Package (DPoP). Gene Ontology (GO) term enrichment has been included in the app so that a user can automatically/programmatically describe the over/under-represented GO terms in the derivative profiling results using domain specific knowledge found in their organism's specific GO database file. The advantage of the DPoP analysis is that it is computationally inexpensive, it does not require fold change calculations, it describes both instantaneous as well as overall behavior, and it achieves statistical confidence with signal trajectories of a single bio-replicate over four or more points. CONCLUSIONS: While we apply this method to time dynamic transcriptomic and phosphoproteomic datasets, it is a numerically generalizable technique that can be applied to any organism and any field interested in time series data analysis. The app described in this work enables omics researchers with no computer science background to apply derivative profiling to their data sets, while also allowing multidisciplined users to build on the nascent idea of profiling derivatives in omics.

Dynamic Transcriptomic and Phosphoproteomic Analysis During Cell Wall Stress in Aspergillus nidulans.

The fungal cell-wall integrity signaling (CWIS) pathway regulates cellular response to environmental stress to enable wall repair and resumption of normal growth. This complex, interconnected, pathway has been only partially characterized in filamentous fungi. To better understand the dynamic cellular response to wall perturbation, a ß-glucan synthase inhibitor (micafungin) was added to a growing A. nidulans shake-flask culture. From this flask, transcriptomic and phosphoproteomic data were acquired over 10 and 120 min, respectively. To differentiate statistically-significant dynamic behavior from noise, a multivariate adaptive regression splines (MARS) model was applied to both data sets. Over 1800 genes were dynamically expressed and over 700 phosphorylation sites had changing phosphorylation levels upon micafungin exposure. Twelve kinases had altered phosphorylation and phenotypic profiling of all non-essential kinase deletion mutants revealed putative connections between PrkA, Hk-8-4, and Stk19 and the CWIS pathway. Our collective data implicate actin regulation, endocytosis, and septum formation as critical cellular processes responding to activation of the CWIS pathway, and connections between CWIS and calcium, HOG, and SIN signaling pathways.

In Silico Evolution of High-Performing Metal Organic Frameworks for Methane Adsorption.

The increased use of transition fuels, such as natural gas, and the resulting increase in methane emissions have resulted in a need for novel methane storage materials. Metal-organic frameworks (MOFs) have shown promise as efficient storage materials. A virtually limitless number of potential MOFs can be hypothesized, which exhibit a wide variety of different structural and chemical characteristics. Because of the numerous possibilities, identification of the best MOF for methane storage can be a potentially challenging problem. In this work, determination of the best such MOF was cast as an inverse function problem. The function, a random forest (RF) model using 12 structural and chemical descriptors, was trained on 10% of a data set consisting of 130 398 hypothetical MOFs (hMOFs) to predict simulated methane uptake. The RF model was tested on the remaining 90% of the data. After validation, a genetic algorithm (GA) was used to evolve in silico the best MOFs for methane adsorption. The RF model was imbedded into the GA as the fitness function to predict the methane uptake of the evolved MOFs (eMOFs). The best 15 eMOFs matched hMOFs found in the top 1% of the database. Nine of the 15 eMOFs were found in the top 0.1%. More impressively, two of the eMOFs matched the top two hypothetical MOFs with the highest methane uptake values out of the entire database of 130 398 MOFs. Further, by leveraging the ensemble nature of the GA, it was possible to characterize the importance of the different material properties for methane adsorption, providing fundamental insight for future material design strategies.

Phosphoproteomic and transcriptomic analyses reveal multiple functions for Aspergillus nidulans MpkA independent of cell wall stress.

The protein kinase MpkA plays a prominent role in the cell wall integrity signaling (CWIS) pathway, acting as the terminal MAPK activating expression of genes which encode cell wall biosynthetic enzymes and other repair functions. Numerous studies focus on MpkA function during cell wall perturbation. Here, we focus on the role MpkA plays outside of cell wall stress, during steady state growth. In an effort to seek other, as yet unknown, connections to this pathway, an mpkA deletion mutant (ΔmpkA) was subjected to phosphoproteomic and transcriptomic analysis. When compared to the control (isogenic parent of ΔmpkA), there is strong evidence suggesting MpkA is involved with maintaining cell wall strength, branching regulation, and the iron starvation pathway, among others. Particle-size analysis during shake flask growth revealed ΔmpkA mycelia were about 4 times smaller than the control strain and more than 90 cell wall related genes show significantly altered expression levels. The deletion mutant had a significantly higher branching rate than the control and phosphoproteomic results show putative branching-regulation proteins, such as CotA, LagA, and Cdc24, have a significantly different level of phosphorylation. When grown in iron limited conditions, ΔmpkA had no difference in growth rate or production of siderophores, whereas the control strain showed decreased growth rate and increased siderophore production. Transcriptomic data revealed over 25 iron related genes with altered transcript levels. Results suggest MpkA is involved with regulation of broad cellular functions in the absence of stress.

Classification of Tea Aromas Using Multi-Nanoparticle Based Chemiresistor Arrays.

Nanoparticle based chemical sensor arrays with four types of organo-functionalized gold nanoparticles (AuNPs) were introduced to classify 35 different teas, including black teas, green teas, and herbal teas. Integrated sensor arrays were made using microfabrication methods including photolithography and lift-off processing. Different types of nanoparticle solutions were drop-cast on separate active regions of each sensor chip. Sensor responses, expressed as the ratio of resistance change to baseline resistance (ΔR/R0), were used as input data to discriminate different aromas by statistical analysis using multivariate techniques and machine learning algorithms. With five-fold cross validation, linear discriminant analysis (LDA) gave 99% accuracy for classification of all 35 teas, and 98% and 100% accuracy for separate datasets of herbal teas, and black and green teas, respectively. We find that classification accuracy improves significantly by using multiple types of nanoparticles compared to single type nanoparticle arrays. The results suggest a promising approach to monitor the freshness and quality of tea products.

Batch, design optimization, and DNA sequencing study for continuous 1,3-propanediol production from waste glycerol by a soil-based inoculum.

1,3-propanediol (1,3-PD) was produced with a robust fermentation process using waste glycerol feedstock from biodiesel production and a soil-based bacterial inoculum. An iterative inoculation method was developed to achieve independence from soil and selectively breed bacterial populations capable of glycerol metabolism to 1,3-PD. The inoculum showed high resistance to impurities in the feedstock. 1,3-PD selectivity and yield in batch fermentations was optimized by appropriate nutrient compositions and pH control. The batch yield of 1,3-PD was maximized to ~0.7 mol/mol for industrial glycerol which was higher than that for pure glycerin. 16S rDNA sequencing results show a systematic selective enrichment of 1,3-PD producing bacteria with iterative inoculation and subsequent process control. A statistical design of experiments was carried out on industrial glycerol batches to optimize conditions, which were used to run two continuous flow stirred-tank reactor (CSTR) experiments over a period of >500 h each. A detailed analysis of steady states at three dilution rates is presented. Enhanced specific 1,3-PD productivity was observed with faster dilution rates due to lower levels of solvent degeneration. 1,3-PD productivity, specific productivity, and yield of 1.1 g/l hr, 1.5 g/g hr, and 0.6 mol/mol of glycerol were obtained at a dilution rate of 0.1 h(-1)which is bettered only by pure strains in pure glycerin feeds.

Volatility in mRNA secondary structure as a design principle for antisense.

Designing effective antisense sequences is a formidable problem. A method for predicting efficacious antisense holds the potential to provide fundamental insight into this biophysical process. More practically, such an understanding increases the chance of successful antisense design as well as saving considerable time, money and labor. The secondary structure of an mRNA molecule is believed to be in a constant state of flux, sampling several different suboptimal states. We hypothesized that particularly volatile regions might provide better accessibility for antisense targeting. A computational framework, GenAVERT was developed to evaluate this hypothesis. GenAVERT used UNAFold and RNAforester to generate and compare the predicted suboptimal structures of mRNA sequences. Subsequent analysis revealed regions that were particularly volatile in terms of intramolecular hydrogen bonding, and thus potentially superior antisense targets due to their high accessibility. Several mRNA sequences with known natural antisense target sites as well as artificial antisense target sites were evaluated. Upon comparison, antisense sequences predicted based upon the volatility hypothesis closely matched those of the naturally occurring antisense, as well as those artificial target sites that provided efficient down-regulation. These results suggest that this strategy may provide a powerful new approach to antisense design.

Semi-automated curation of metabolic models via flux balance analysis: a case study with Mycoplasma gallisepticum.

Primarily used for metabolic engineering and synthetic biology, genome-scale metabolic modeling shows tremendous potential as a tool for fundamental research and curation of metabolism. Through a novel integration of flux balance analysis and genetic algorithms, a strategy to curate metabolic networks and facilitate identification of metabolic pathways that may not be directly inferable solely from genome annotation was developed. Specifically, metabolites involved in unknown reactions can be determined, and potentially erroneous pathways can be identified. The procedure developed allows for new fundamental insight into metabolism, as well as acting as a semi-automated curation methodology for genome-scale metabolic modeling. To validate the methodology, a genome-scale metabolic model for the bacterium Mycoplasma gallisepticum was created. Several reactions not predicted by the genome annotation were postulated and validated via the literature. The model predicted an average growth rate of 0.358±0.12[Formula: see text], closely matching the experimentally determined growth rate of M. gallisepticum of 0.244±0.03[Formula: see text]. This work presents a powerful algorithm for facilitating the identification and curation of previously known and new metabolic pathways, as well as presenting the first genome-scale reconstruction of M. gallisepticum.

Incremental Inverse Design of Desired Soybean Phenotypes.

We present an application of computational inverse design, which reverses the conventional trial-and-error forward design paradigm, optimizes biological phenotype by directly modifying genotype. The limitations of inverse design in genotype-to-bulk phenotype (G-BP) mapping can be addressed via an established design paradigm: "design, build, test, learn" (DBTL), where computational inverse design automates both the design and learn phases. In any context, inverse design is limited by the fundamental "one-to-many" nature of the inverse function. G-BP inverse design is further limited by the number of single nucleotide polymorphisms that can be made to a member of the population while maintaining feasibility of genotype creation and biological viability. Considering these limitations, we propose a design paradigm based on incremental optimization of phenotype through a combined computational and experimental approach. We intend this work to be a foundational synthesis of well-known techniques applied to the context of genotype-to-bulk phenotype inverse design, which has not yet been performed in the literature. The design pipeline can optimize phenotype by either directly proposing genotypic changes, or simply by suggesting parents to be used for selective breeding. The soybean nested association matrix data set is used to present an in silico case study of the design pipeline by performing optimization that maximizes protein content while constraining other phenotypes. A random forest (RF) is used to model the genotype-to-phenotype relationship, and a genetic algorithm is used to query the RF until a feasible genotype with desired phenotype is discovered. After 20 in silico DBTL cycles, a final population of individuals with a mean protein content of 36.13%, an increase of three standard deviations above the original mean is suggested.

Determination of lethal and mutation induction doses of gamma rays for gladiolus (Gladiolus grandifloras Hort.) genotypes.

Gladiolus is a highly allogamous flower plant, but owing to the prolonged juvenile phase, asexual propagation is preferred, which acts as a barrier for the induction of natural genetic variability in gladiolus. Therefore, the induced mutagenesis could be utilized for the creation of desirable genotypes, without altering their basic agronomic features. An analysis of the optimum doses of γ radiation for the induction of fruitful mutations could be achieved in short period of time, compared with the conventional method of breeding. The objectives of this study were to perform radiosensitivity tests on various gladiolus genotypes using different doses of gamma rays and to determine the optimal dose of radiation dose for obtaining the greatest number of mutants. The present experiment was carried out during the winter-spring seasons, for the four consecutive years of 2017-18, 2018-19, 2019-20, and 2020-21. The seven genotypes of gladiolus were exposed to seven doses of gamma rays (60Cobalt). Plants irradiated with radiation doses lower than 4.5 Kr (G1) had greater plant survivability than the higher doses of gamma rays (≥5.0 Kr). The radiation of G0 (0 Kr) result in highest plant survivability, while radiation dose of G6 (6.5 Kr) resulted lowest survivability. LD25 and BD50 for all the genotypes were achieved except for V5 and V7, similarly the median lethal doses (LD50) for V3 and V4 genotypes had been achieved. The highest flower blindness percent and percent abnormal plants were observed at G5 and G6 and between the 4.0 Kr (G1) and 5.5 Kr (G4) gamma ray doses, respectively. The flower colour mutation frequency was recorded highest in genotypes Tiger Flame at 5.0 Kr (V7G3), while the Flower colour mutation spectrum was identified between 4.0 Kr (G1) to 5.5 Kr (G4) in all the genotypes except for genotypes V5 and V7. For the generation of higher phenotypic variations, radiation dose between 4.0 Kr (G1) and 5.5 Kr (G4) were found the most prominent. Specifically the gamma rays radiation dose of 5.5 Kr (G4) resulted in the highest flower colour mutation frequency. These isolated mutant lines will broaden the gladiolus gene pool and support future gladiolus breeding experiments.

Aspergillus nidulans Septa Are Indispensable for Surviving Cell Wall Stress.

Septation in filamentous fungi is a normal part of development, which involves the formation of cross-hyphal bulkheads, typically containing pores, allowing cytoplasmic streaming between compartments. Based on previous findings regarding septa and cell wall stress, we hypothesized that septa are critical for survival during cell wall stress. To test this hypothesis, we used known Aspergillus nidulans septation-deficient mutants (ΔsepH, Δbud3, Δbud4, and Δrho4) and six antifungal compounds. Three of these compounds (micafungin, Congo red, and calcofluor white) are known cell wall stressors which activate the cell wall integrity signaling pathway (CWIS), while the three others (cycloheximide, miconazole, and 2,3-butanedione monoxime) perturb specific cellular processes not explicitly related to the cell wall. Our results show that deficiencies in septation lead to fungi which are more susceptible to cell wall-perturbing compounds but are no more susceptible to other antifungal compounds than a control. This implies that septa play a critical role in surviving cell wall stress. IMPORTANCE The ability to compartmentalize potentially lethal damage via septation appears to provide filamentous fungi with a facile means to tolerate diverse forms of stress. However, it remains unknown whether this mechanism is deployed in response to all forms of stress or is limited to specific perturbations. Our results support the latter possibility by showing that presence of septa promotes survival in response to cell wall damage but plays no apparent role in coping with other unrelated forms of stress. Given that cell wall damage is a primary effect caused by exposure to the echinocandin class of antifungal agents, our results emphasize the important role that septa might play in enabling resistance to these drugs. Accordingly, the inhibition of septum formation could conceivably represent an attractive approach to potentiating the effects of echinocandins and mitigating resistance in human fungal pathogens.

Enhancing iCVD Modification of Electrospun Membranes for Membrane Distillation Using a 3D Printed Scaffold.

Electrospun membranes have shown promise for use in membrane distillation (MD) as they exhibit exceptionally low vapor transport. Their high porosity coupled with the occasional large pore can make them prone to wetting. In this work, initiated chemical vapor deposition (iCVD) is used to modify for electrospun membranes with increased hydrophobicity of the fiber network. To demonstrate conformal coating, we demonstrate the approach on intrinsically hydrophilic electrospun fibers and render the fibers suitable for MD. We enable conformal coating using a unique coating procedure, which provides convective flow of deposited polymers during iCVD. This is made possible by using a 3D printed scaffold, which changed the orientation of the membrane during the coating process. The new coating orientation allows both sides as well as the interior of the membrane to be coated simultaneously and reduced the coating time by a factor of 10 compared to conventional CVD approaches. MD testing confirmed the hydrophobicity of the material as 100% salt rejections were obtained.

Energy-Geometry Dependency of Molecular Structures: A Multistep Machine Learning Approach.

There is growing interest in estimating quantum observables while circumventing expensive computational overhead for facile in silico materials screening. Machine learning (ML) methods are implemented to perform such calculations in shorter times. Here, we introduce a multistep method based on machine learning algorithms to estimate total energy on the basis of spatial coordinates and charges for various chemical structures, including organic molecules, inorganic molecules, and ions. This method quickly calculates total energy with 0.76 au in root-mean-square error (RMSE) and 1.5% in mean absolute percent error (MAPE) when tested on a database of optimized and unoptimized structures. Using similar molecular representations, experimental thermochemical properties were estimated, with MAPE as low as 6% and RMSE of 8 cal/mol·K for heat capacity in a 10-fold cross-validation.

Trends in Solid Adsorbent Materials Development for CO2 Capture.

A recent report from the United Nations has warned about the excessive CO2 emissions and the necessity of making efforts to keep the increase in global temperature below 2 °C. Current CO2 capture technologies are inadequate for reaching that goal, and effective mitigation strategies must be pursued. In this work, we summarize trends in materials development for CO2 adsorption with focus on recent studies. We put adsorbent materials into four main groups: (I) carbon-based materials, (II) silica/alumina/zeolites, (III) porous crystalline solids, and (IV) metal oxides. Trends in computational investigations along with experimental findings are covered to find promising candidates in light of practical challenges imposed by process economics.

Comprehensive Analysis of Aspergillus nidulans PKA Phosphorylome Identifies a Novel Mode of CreA Regulation.

In filamentous fungi, an important kinase responsible for adaptation to changes in available nutrients is cyclic AMP (cAMP)-dependent protein kinase (protein kinase A [PKA]). This kinase has been well characterized at a molecular level, but its systemic action and direct/indirect targets are generally not well understood in filamentous fungi. In this work, we used a pkaA deletion strain (ΔpkaA) to identify Aspergillus nidulans proteins for which phosphorylation is dependent (either directly or indirectly) on PKA. A combination of phosphoproteomic and transcriptomic analyses revealed both direct and indirect targets of PKA and provided a global perspective on its function. One of these targets was the transcription factor CreA, the main repressor responsible for carbon catabolite repression (CCR). In the ΔpkaA strain, we identified a previously unreported phosphosite in CreA, S319, which (based on motif analysis) appears to be a direct target of Stk22 kinase (AN5728). Upon replacement of CreA S319 with an alanine (i.e., phosphonull mutant), the dynamics of CreA import to the nucleus are affected. Collectively, this work provides a global overview of PKA function while also providing novel insight regarding significance of a specific PKA-mediated phosphorylation event.IMPORTANCE The cyclic AMP (cAMP)-dependent protein kinase A (PKA) signaling pathway is well conserved across eukaryotes, and previous work has shown that it plays an important role in regulating development, growth, and virulence in a number of fungi. PKA is activated in response to extracellular nutrients and acts to regulate metabolism and growth. While a number of components in the PKA pathway have been defined in filamentous fungi, current understanding does not provide a global perspective on PKA function. Thus, this work is significant in that it comprehensively identifies proteins and functional pathways regulated by PKA in a model filamentous fungus. This information enhances our understanding of PKA action and may provide information on how to manipulate it for specific purposes.

Bayesian-based selection of metabolic objective functions.

MOTIVATION: A critical component of in silico analysis of underdetermined metabolic systems is the identification of the appropriate objective function. A common assumption is that the objective of the cell is to maximize growth. This objective function has been shown to be consistent in a few limited experimental cases, but may not be universally appropriate. Here a method is presented to quantitatively determine the most probable objective function. RESULTS: The genome-scale metabolism of Escherichia coli growing on succinate was used as a case-study for analysis. Five different objective functions, including maximization of growth rate, were chosen based on biological plausibility. A combination of flux balance analysis and linear programming was used to simulate cellular metabolism, which was then compared to independent experimental data using a Bayesian objective function discrimination technique. After comparing rates of oxygen uptake and acetate production, minimization of the production rate of redox potential was determined to be the most probable objective function. Given the appropriate reaction network and experimental data, the discrimination technique can be applied to any bacterium to test a variety of different possible objective functions. SUPPLEMENTARY INFORMATION: Additional files, code and a program for carrying out model discrimination are available at http://www.engr.uconn.edu/~srivasta/modisc.html.

Low-abundant bacteria drive compositional changes in the gut microbiota after dietary alteration.

BACKGROUND: As the importance of beneficial bacteria is better recognized, understanding the dynamics of symbioses becomes increasingly crucial. In many gut symbioses, it is essential to understand whether changes in host diet play a role in the persistence of the bacterial gut community. In this study, termites were fed six dietary sources and the microbial community was monitored over a 49-day period using 16S rRNA gene sequencing. A deep backpropagation artificial neural network (ANN) was used to learn how the six different lignocellulose food sources affected the temporal composition of the hindgut microbiota of the termite as well as taxon-taxon and taxon-substrate interactions. RESULTS: Shifts in the termite gut microbiota after diet change in each colony were observed using 16S rRNA gene sequencing and beta diversity analyses. The artificial neural network accurately predicted the relative abundances of taxa at random points in the temporal study and showed that low-abundant taxa maintain community driving correlations in the hindgut. CONCLUSIONS: This combinatorial approach utilizing 16S rRNA gene sequencing and deep learning revealed that low-abundant bacteria that often do not belong to the core community are drivers of the termite hindgut bacterial community composition.

Altered secretion patterns and cell wall organization caused by loss of PodB function in the filamentous fungus Aspergillus nidulans.

Filamentous fungi are widely used in the production of a variety of industrially relevant enzymes and proteins as they have the unique ability to secrete tremendous amounts of proteins. However, the secretory pathways in filamentous fungi are not completely understood. Here, we investigated the role of a mutation in the POlarity Defective (podB) gene on growth, protein secretion, and cell wall organization in Aspergillus nidulans using a temperature sensitive (Ts) mutant. At restrictive temperature, the mutation resulted in lack of biomass accumulation, but led to a significant increase in specific protein productivity. Proteomic analysis of the secretome showed that the relative abundance of 584 (out of 747 identified) proteins was altered due to the mutation. Of these, 517 were secreted at higher levels. Other phenotypic differences observed in the mutant include up-regulation of unfolded protein response (UPR), deformation of Golgi apparatus and uneven cell wall thickness. Furthermore, proteomic analysis of cell wall components in the mutant revealed the presence of intracellular proteins in higher abundance accompanied by lower levels of most cell wall proteins. Taken together, results from this study suggest the importance of PodB as a target when engineering fungal strains for enhanced secretion of valuable biomolecules.

Systematic tuning of parameters in support vector clustering.

Clustering algorithms divide a set of observations into groups so that members of the same group share common features. In most of the algorithms, tunable parameters are set arbitrarily or by trial and error, resulting in less than optimal clustering. This paper presents a global optimization strategy for the systematic and optimal selection of parameter values associated with a clustering method. In the process, a performance criterion for the optimization model is proposed and benchmarked against popular performance criteria from the literature (namely, the Silhouette coefficient, Dunn's index, and Davies-Bouldin index). The tuning strategy is illustrated using the support vector clustering (SVC) algorithm and simulated annealing. In order to reduce the computational burden, the paper also proposes an alternative to the adjacency matrix method (used for the assignment of cluster labels), namely the contour plotting approach. Datasets tested include the iris and the thyroid datasets from the UCI repository, as well as lymphoma and breast cancer data. The optimal tuning parameters are determined efficiently, while the contour plotting approach leads to significant reductions in computational effort (CPU time) especially for large datasets. The performance criteria comparisons indicate mixed results. Specifically, the Silhouette coefficient and the Davies-Bouldin index perform better, while the Dunn's index is worse on average than the proposed performance index.

Machine Learning Using Combined Structural and Chemical Descriptors for Prediction of Methane Adsorption Performance of Metal Organic Frameworks (MOFs).

Using molecular simulation for adsorbent screening is computationally expensive and thus prohibitive to materials discovery. Machine learning (ML) algorithms trained on fundamental material properties can potentially provide quick and accurate methods for screening purposes. Prior efforts have focused on structural descriptors for use with ML. In this work, the use of chemical descriptors, in addition to structural descriptors, was introduced for adsorption analysis. Evaluation of structural and chemical descriptors coupled with various ML algorithms, including decision tree, Poisson regression, support vector machine and random forest, were carried out to predict methane uptake on hypothetical metal organic frameworks. To highlight their predictive capabilities, ML models were trained on 8% of a data set consisting of 130,398 MOFs and then tested on the remaining 92% to predict methane adsorption capacities. When structural and chemical descriptors were jointly used as ML input, the random forest model with 10-fold cross validation proved to be superior to the other ML approaches, with an R2 of 0.98 and a mean absolute percent error of about 7%. The training and prediction using the random forest algorithm for adsorption capacity estimation of all 130,398 MOFs took approximately 2 h on a single personal computer, several orders of magnitude faster than actual molecular simulations on high-performance computing clusters.