Comparative molecular dynamics simulations provided insights into the mechanisms of cold-adaption of alginate lyases from the PL7 family.

Alginate is an important polysaccharide that is abundant in the marine environments, including the Polar Regions, and bacterial alginate lyases play key roles in its degradation. Many reported alginate lyases show characteristics of cold-adapted enzymes, including relatively low temperature optimum of activities (Topt) and low thermal stabilities. However, the cold-adaption mechanisms of alginate lyases remain unclear. Here, we studied the cold-adaptation mechanisms of alginate lyases by comparing four members of the PL7 family from different environments: AlyC3 from the Arctic ocean (Psychromonas sp. C-3), AlyA1 from the temperate ocean (Zobellia galactanivorans), PA1167 from the human pathogen (Pseudomonas aeruginosa PAO1), and AlyQ from the tropic ocean (Persicobacter sp. CCB-QB2). Sequence comparison and comparative molecular dynamics (MD) simulations revealed two main strategies of cold adaptation. First, the Arctic AlyC3 and temperate AlyA1 increased the flexibility of the loops close to the catalytic center by introducing insertions at these loops. Second, the Arctic AlyC3 increased the electrostatic attractions with the negatively charged substrate by introducing a high portion of positively charged lysine at three of the insertions mentioned above. Furthermore, our study also revealed that the root mean square fluctuation (RMSF) increased greatly when the temperature was increased to Topt or higher, suggesting the RMSF increase temperature as a potential indicator of the cold adaptation level of the PL7 family. This study provided new insights into the cold-adaptation mechanisms of bacterial alginate lyases and the marine carbon cycling at low temperatures.

Multi-fidelity machine learning for predicting bandgaps of nonlinear optical crystals.

Nonlinear optical (NLO) materials are of great importance in modern optics and industry because of their intrinsic capability of wavelength conversion. Bandgap is a key property of NLO crystals. In recent years, machine learning (ML) has become a powerful tool to predict the bandgaps of compounds before synthesis. However, the shortage of available experimental data of NLO crystals poses a significant challenge for the exploration of new NLO materials using ML. In this work, we proposed a new multi-fidelity ML approach based on the multilevel descriptors developed by us (Z.-Y. Zhang, X. Liu, L. Shen, L. Chen and W.-H. Fang, J. Phys. Chem. C, 2021, 125, 25175-25188) and the gradient boosting regression tree algorithm. The calculated and experimental bandgaps of NLO crystals were collected as the low- and high-fidelity labels, respectively. The experimental values were predicted based on chemical compositions of crystals without prior knowledge about crystal structures. The multi-fidelity ML model overcame the performance of single-fidelity predictor. Furthermore, it was observed that less accurate predictions on the low-fidelity label may result in more accurate prediction on the high-fidelity label, at least in the present case. Using the multi-fidelity ML model with the best performance in this work, the mean absolute error on the test set of experimental bandgaps was 0.293 eV, which is smaller than that using the single-fidelity model (0.355 eV). It is far from perfect but accurate enough as an effective computational tool in the first step to discover novel NLO materials.

Construction of Highly Accurate Machine Learning Potential Energy Surfaces for Excited-State Dynamics Simulations Based on Low-Level Data Sets.

Machine learning is capable of effectively predicting the potential energies of molecules in the presence of high-quality data sets. Its application in the construction of ground- and excited-state potential energy surfaces is attractive to accelerate nonadiabatic molecular dynamics simulations of photochemical reactions. Because of the huge computational cost of excited-state electronic structure calculations, the construction of a high-quality data set becomes a bottleneck. In the present work, we first built two data sets. One was obtained from surface hopping dynamics simulations at the semiempirical OM2/MRCI level. Another was extracted from the dynamics trajectories at the CASSCF level, which was reported previously. The ground- and excited-state potential energy surfaces of ethylene-bridged azobenzene at the CASSCF computational level were constructed based on the former low-level data set. Although non-neural network machine learning methods can achieve good or modest performance during the training process, only neural network models provide reliable predictions on the latter external test data set. The BPNN and SchNet combined with the Δ-ML scheme and the force term in the loss functions are recommended for dynamics simulations. Then, we performed excited-state dynamics simulations of the photoisomerization of ethylene-bridged azobenzene on machine learning potential energy surfaces. Compared with the lifetimes of the first excited state (S1) estimated at different computational levels, our results on the E isomer are in good agreement with the high-level estimation. However, the overestimation of the Z isomer is unimproved. It suggests that smaller errors during the training process do not necessarily translate to more accurate predictions on high-level potential energies or better performance on nonadiabatic dynamics simulations, at least in the present case.

Genomic and Metagenomic Insights into the Distribution of Nicotine-degrading Enzymes in Human Microbiota.

Introduction: Nicotine degradation is a new strategy to block nicotine-induced pathology. The potential of human microbiota to degrade nicotine has not been explored. Aims: This study aimed to uncover the genomic potentials of human microbiota to degrade nicotine. Methods: To address this issue, we performed a systematic annotation of Nicotine-Degrading Enzymes (NDEs) from genomes and metagenomes of human microbiota. A total of 26,295 genomes and 1,596 metagenomes for human microbiota were downloaded from public databases and five types of NDEs were annotated with a custom pipeline. We found 959 NdhB, 785 NdhL, 987 NicX, three NicA1, and three NicA2 homologs. Results: Genomic classification revealed that six phylum-level taxa, including Proteobacteria, Firmicutes, Firmicutes_A, Bacteroidota, Actinobacteriota, and Chloroflexota, can produce NDEs, with Proteobacteria encoding all five types of NDEs studied. Analysis of NicX prevalence revealed differences among body sites. NicX homologs were found in gut and oral samples with a high prevalence but not found in lung samples. NicX was found in samples from both smokers and non-smokers, though the prevalence might be different. Conclusion: This study represents the first systematic investigation of NDEs from the human microbiota, providing new insights into the physiology and ecological functions of human microbiota and shedding new light on the development of nicotine-degrading probiotics for the treatment of smoking-related diseases.

[Spatio-temporal Evolution of Drought and Its Impact on Key Resources in the Important Ecological Functional Area: A Case Study in the Yellow River Basin].

This study explored the characteristics of spatial and temporal changes in drought in the Yellow River Basin from 2001 to 2020 based on TVPDI, surface runoff, vegetation net primary productivity, and grain yield data. Further, the effects of drought on water resources, grain resources, and vegetation resources were also analyzed using data spatialization methods, slope trend analysis, and Pearson correlation analysis. The results showed that:① The spatial distribution of drought in the Yellow River basin was stepped from southeast to northwest, and 60.6 % of the basin was in drought. The overall trend of drought in the basin was decreasing annually, and 94 % of the basin was gradually changing from drought to wet conditions, and the trend of drought from spring to winter decreased first and then increased. ② From the spatial and temporal changes in important resources in the basin, 53 % of the key surface runoff areas showed an increasing trend and were mainly located in the southwest of the basin; the net primary productivity (NPP) of vegetation and grain yield of food resources also showed an increasing trend. ③ Drought and the three types of resources showed significant spatial correlations, and the higher the degree of drought, the more significant the effects on surface runoff, vegetation productivity, and grain yield. However, the important resources in areas that had become wetter in recent years had not increased significantly, which indicated that the effects of drought on the three types of important resources had a time lag, and their lags had significant differences in spatial distribution and geographical differentiation patterns. This study has important theoretical implications for agricultural production, drought mitigation, and ecological conservation in the Yellow River Basin.

Annotation of 2,507 Saccharomyces cerevisiae genomes.

Saccharomyces cerevisiae (baker's yeast, budding yeast) is one of the most important model organisms for biological research and is a crucial microorganism in industry. Currently, a huge number of Saccharomyces cerevisiae genome sequences are available at the public domain. However, these genomes are distributed at different websites and a large number of them are released without annotation information. To provide one complete annotated genome data resource, we collected 2,507 Saccharomyces cerevisiae genome assemblies and re-annotated 2,506 assemblies using a custom annotation pipeline, producing a total of 15,407,164 protein-coding gene models. With a custom pipeline, all these gene sequences were clustered into families. A total of 1,506 single-copy genes were selected as marker genes, which were then used to evaluate the genome completeness and base qualities of all assemblies. Pangenomic analyses were performed based on a selected subset of 847 medium-high-quality genomes. Statistical comparisons revealed a number of gene families showing copy number variations among different organism sources. To the authors' knowledge, this study represents the largest genome annotation project of S. cerevisiae so far, providing rich genomic resources for the future studies of the model organism S. cerevisiae and its relatives.IMPORTANCESaccharomyces cerevisiae (baker's yeast, budding yeast) is one of the most important model organisms for biological research and is a crucial microorganism in industry. Though a huge number of Saccharomyces cerevisiae genome sequences are available at the public domain, these genomes are distributed at different websites and most are released without annotation, hindering the efficient reuse of these genome resources. Here, we collected 2,507 genomes for Saccharomyces cerevisiae, performed genome annotation, and evaluated the genome qualities. All the obtained data have been deposited at public repositories and are freely accessible to the community. This study represents the largest genome annotation project of S. cerevisiae so far, providing one complete annotated genome data set for S. cerevisiae, an important workhorse for fundamental biology, biotechnology, and industry.

Insights into the composition and assembly mechanism of microbial communities on intertidal microsand grains.

Introduction: Marine microorganisms are essential in marine ecosystems and have always been of interest. Currently, most marine microbial communities are studied at the bulk scale (millimeters to centimeters), and the composition, function and underlying assembly mechanism of microbial communities at the microscale (sub-100 micrometers) are unclear. Methods: The microbial communities on microsand grains (40-100 µm, n = 150) from marine sediment were investigated and compared with those on macrosand grains (400-1000 µm, n = 60) and bulk sediments (n = 5) using amplicon sequencing technology. Results: The results revealed a significant difference between microsand grains and macrosand grains. Microsand grains had lower numbers of operational taxonomic units (OTUs(97%)) and predicted functional genes than macrosand grains and bulk-scale samples. Microsand grains also showed greater intersample differences in the community composition and predicted functional genes than macrosand grains, suggesting a high level of heterogeneity of microbial communities at the microscale. Analyses based on ecological models indicated that stochastic processes dominated the assembly of microbial communities on sand grains. Consistently, cooccurrence network analyses showed that most microbial cooccurrence associations on sand grains were highly unstable. Metagenomic sequencing and further genome-scale metabolic modeling revealed that only a small number (1.3%) of microbe pairs showed high cooperative potential. Discussion: This study explored the microbial community of marine sediments at the sub-100 µm scale, broadening the knowledge of the structure and assembly mechanism of marine microbial communities.