Artificial intelligence for geoscience: Progress, challenges, and perspectives.

This paper explores the evolution of geoscientific inquiry, tracing the progression from traditional physics-based models to modern data-driven approaches facilitated by significant advancements in artificial intelligence (AI) and data collection techniques. Traditional models, which are grounded in physical and numerical frameworks, provide robust explanations by explicitly reconstructing underlying physical processes. However, their limitations in comprehensively capturing Earth's complexities and uncertainties pose challenges in optimization and real-world applicability. In contrast, contemporary data-driven models, particularly those utilizing machine learning (ML) and deep learning (DL), leverage extensive geoscience data to glean insights without requiring exhaustive theoretical knowledge. ML techniques have shown promise in addressing Earth science-related questions. Nevertheless, challenges such as data scarcity, computational demands, data privacy concerns, and the "black-box" nature of AI models hinder their seamless integration into geoscience. The integration of physics-based and data-driven methodologies into hybrid models presents an alternative paradigm. These models, which incorporate domain knowledge to guide AI methodologies, demonstrate enhanced efficiency and performance with reduced training data requirements. This review provides a comprehensive overview of geoscientific research paradigms, emphasizing untapped opportunities at the intersection of advanced AI techniques and geoscience. It examines major methodologies, showcases advances in large-scale models, and discusses the challenges and prospects that will shape the future landscape of AI in geoscience. The paper outlines a dynamic field ripe with possibilities, poised to unlock new understandings of Earth's complexities and further advance geoscience exploration.

Land use change alters carbon and nitrogen dynamics mediated by fungal functional guilds within soil aggregates.

Land use change is one of the greatest threats to soil biodiversity and ecological functions; however, how such a transition affects soil carbon (C) and nitrogen (N) dynamics driven by fungal communities at the aggregate level remains unclear. Here, we explored the variation in soil C and N pools, specific enzyme activities and fungal communities and functional guilds within three aggregate sizes (megaaggregates, > 2 mm; macroaggregates, 0.25-2 mm; microaggregates, < 0.25 mm) in a natural forest, 12- and 24-year-old rubber monocultures and corresponding agroforestry systems in tropical China. Tropical forest conversion to rubber monocultures generally reduced C and N pools in all aggregates, while agroforestry systems decreased microbial biomass C and N. Carbon- and N-degrading enzyme activities responded differently to forest conversion and were enhanced in agroforestry systems. The levels of C and N pools and their related enzyme activities increased as the aggregate size decreased. Moreover, fungal compositional shifts in dominance from copiotrophic Ascomycota and Basidiomycota (r-strategists) into oligotrophic Zygomycota (K-strategists) were noted following forest conversion, resulting in more pathogenic fungi at the expense of saprotrophic and arbuscular mycorrhizal fungi. Pathogenic fungi were greatly inhibited due to abundant Mortierella after the establishment of 12-year-old agroforestry systems. The diversity of saprotrophic fungi was the highest in microaggregates. Regardless of land use type, aggregate-associated C and N pools, especially DOC, MBC, NO3--N and DON in microaggregates, were interactively mediated by functional guilds of fungi, which was primarily driven by soil pH. These results highlight the importance of fungal functional guilds in determining C and N dynamics at the aggregate level and provide insights into the sustainable management of cash tree plantations.

Effect of short-term warming and drought on the methanogenic communities in degraded peatlands in Zoige Plateau.

Peatlands in Qinghai-Tibetan are degrading with climate change and human activities. Peatland degradation and climate change affect methane emissions. Methanogens are key functional microbes during methane production; however, knowledge of methanogens in degraded peatlands is lacking. Here, we investigated the effects of short-term (1 year) warming (OTC), drought (20%), and their combination on methanogens in the degraded peatlands on the Zoige Plateau of China via qPCR and clone library analysis. The results showed that Methanomicrobiales and Methanobacteriales were predominant in all the treatments. Non-metric multidimensional scaling (NMDS) and PERMANOVA analyses showed that the methanogenic community structure among the climate change treatments was not significantly different. The relative abundance of methanogen communities showed insignificant variation among the climate change treatments. The copy number and Shannon diversity of methanogens were significantly different within the climate change treatments, and drought significantly decreased the copy number of methanogens when compared to the control. The Redundancy analysis (RDA) results and correlation analysis showed that the environmental variables measured had no significant effect on methanogenic community structure and Shannon diversity. These results indicate that methanogens are insensitive to short-term climate change in degraded peatlands. This study provides insight into methane emissions from the Zoige Plateau peatlands by focusing on the possible responses of the methanogens to climate-driven changes.

Variations in bacterial and archaeal community structure and diversity along the soil profiles of a peatland in Southwest China.

As bacteria and archaea are key components in the ecosystem, information on their dynamics in soil profiles is important for understanding the biogeochemical cycles in peatlands. However, little is known about the vertical distribution patterns of bacteria and archaea in the Bitahai peatland, or about their relationships with soil chemical properties. Here, bacterial and archaeal abundance, diversity, and composition of the Bitahai peatlands at 0-100 cm soil depths were analyzed by sequencing of 16S rRNA genes (Illumina, MiSeq). Soil pH, total C, N, and P concentrations and stoichiometric ratios were also estimated. The results revealed that total C and total N contents, as well as C:P and N:P ratios, significantly increased with increasing peatland soil depths, while total P decreased. The top three dominant phyla were Proteobacteria (39.64%), Acidobacteria (12.93%), and Chloroflexi (12.81%) in bacterial communities, and were Crenarchaeota (58.67%), Thaumarchaeota (14.34%), and Euryarchaeota (10.82%) in archaeal communities in the Bitahai peatland, respectively. The total relative abundance of methanogenic groups and ammonia-oxidizing microorganisms all significantly decreased with soil depth. Both bacterial and archaeal diversities were significantly affected by the soil depth. Soil C, N, and P concentrations and stoichiometric ratios markedly impacted the community structure and diversity in archaea, but not in bacteria. Therefore, these results highlighted that the microbial community structure and diversity depended on soil depth for the Bitahai peatlands, and the factors affecting bacteria and archaea in the Bitahai peatlands were different.

A genome-wide association study reveals that the glucosyltransferase OsIAGLU regulates root growth in rice.

Good root growth in the early post-germination stages is an important trait for direct seeding in rice, but its genetic control is poorly understood. In this study, we examined the genetic architecture of variation in primary root length using a diverse panel of 178 accessions. Four QTLs for root length (qRL3, qRL6, qRL7, and qRL11) were identified using genome-wide association studies. One candidate gene was validated for the major QTL qRL11, namely the glucosyltransferase OsIAGLU. Disruption of this gene in Osiaglu mutants reduced the primary root length and the numbers of lateral and crown roots. The natural allelic variations of OsIAGLU contributing to root growth were identified. Functional analysis revealed that OsIAGLU regulates root growth mainly via modulating multiple hormones in the roots, including levels of auxin, jasmonic acid, abscisic acid, and cytokinin. OsIAGLU also influences the expression of multiple hormone-related genes associated with root growth. The regulation of root growth through multiple hormone pathways by OsIAGLU makes it a potential target for future rice breeding for crop improvement.

Soil water content and pH drive archaeal distribution patterns in sediment and soils of water-level-fluctuating zones in the East Dongting Lake wetland, China.

Archaea play a vital role in Earth's geochemical cycles, but the factors that drive their distribution between sediments and water-level-fluctuating zones in the East Dongting Lake (EDL) wetland are poorly understood. Here, we used Illumina MiSeq to investigate the variation in the soil archaeal community structure and diversity among sediments and four water-level-fluctuating zones (mudflat, sedge, sedge-Phragmites, and Phragmites) in the EDL wetland. Diverse archaeal assemblages were found in our study, Crenarchaeota, Euryarchaeota, and ammonia-oxidizing and methanogenic subset were the dominant groups, and all their abundances shifted from sediment to water-level-fluctuating zones. The principal coordinates analysis and cluster analysis showed that the overall archaeal community structure was separated into two clusters: cluster I contained nine samples from sediment, mudflat, and sedge zones, whereas cluster II contained six samples from sedge-Phragmites and Phragmites zones. Archaeal diversity was significantly highest in sediment and lowest in Phragmites zone soils. The Mantel test showed that the variation in archaeal community structure was significantly positively correlated with soil water content and pH. The relative abundances of Crenarchaeota and Nitrososphaerales decreased with soil water content, while Euryarchaeota and Methanomicrobiales increased with soil water content. The relative abundance of Methanomicrobiales significantly decreased with pH (R2 = 0.34-0.48). Chao 1, observed operational taxonomic units, Shannon index, and Simpson index all correlated significantly positively with water content (R2 = 0.40-0.60), while Shannon and Simpson indexes both correlated significantly negatively with pH (R2 = 0.20-0.37). Our results demonstrated that the variations in the archaeal community structure were markedly driven by soil water content and pH in the EDL wetland. Our findings suggested that archaeal communities shifted among sediment and four water-level-fluctuating zones, highlighting that the spatiotemporal heterogeneity of greenhouse gas flux in small scale should be taken into account for accurate prediction of greenhouse gas emissions in the Dongting Lake area, especially on the background of climate change and human activities.

Suspension array-based deafness genetic screening in 53,033 Chinese newborns identifies high prevalence of 109 G>A in GJB2.

OBJECTIVES: More than 50% of congenital hearing loss is attributed to genetic factors. Data of gene mutation associated with hearing loss from large population studies in Chinese population are scarce. In this study, we conducted a comprehensive newborn genetic screening in China to establish the carrier frequency and mutation spectrum of deafness-associated genes. METHODS: A total of 53,033 newborns were screened for hearing defects associated mutations. Twenty hot spot mutations in GJB2, GJB3, SLC26A4 and mitochondria12S rRNA were examined using suspension array analysis. RESULTS: 14,185 newborns (26.75%) were identified with at least one mutated allele. 872 (1.64%) neonates carried homozygous mutations including 112 (0.21%) mitochondrial DNA homoplasmy, 228 (0.43%) were compound heterozygotes, and 11,985 (22.59%) were heterozygotes including 11 (0.02%) mitochondrial DNA heteroplasmy. Top five mutations included 109 G > A, 235 delC, 299-300 delAT in GJB2, IVS7-2 A > G in SLC26A4 and 1555 A > G in mitochondria12S rRNA. Notably, a total of 10,995 neonates (20.73%) carried 109 G > A in GJB2. Moreover, the allele frequencies of 109 G > A were detected 11.61% in Guangdong, 10.44% in Sichuan and 2.88% in Shandong, respectively, a significant difference in prevalence among these geographic regions (p<0.01). In addition, the high frequency of 109 G > A in GJB2 was confirmed by a TaqMan probe-based qPCR assay. Very recently, the ClinGen Hearing Loss Expert Panel reached a consensus and confirmed its pathogenic role in hearing impairment. CONCLUSION: We delineated the mutation profile of common deafness-causing genes in the Chinese population and highlighted the high prevalence of 109 G > A pathogenic mutation. Our study may facilitate early diagnosis/intervention and genetic counseling for hearing impairment in clinical practice.

[Application of suspension array technology for the genetic diagnosis of non-syndromic hearing loss].

OBJECTIVE: To assess the value of suspension array technology (SAT) for the genetic diagnosis of non-syndromic hearing loss (NSHL). METHODS: Three hundred and sixteen NSHL patients were simultaneously tested by SAT targeting 20 hotspot mutations within 4 common pathologic genes among the Chinese population as well as 9 deafness gene mutation detection kits. The results of the two approaches were validated by Sanger sequencing. RESULTS: Among the 316 patients, 161 were found to carry a mutation by SAT. Sixty five patients have carried homozygous or compound heterozygous mutations, which yielded a mutation rate of 50.9% and a diagnostic rate of 21.2%. Seventy three patients were found to be carriers by the 9 deafness gene mutation detection kits. These included 34 patients carrying homozygous or compound heterozygous mutations, which yielded a mutation rate of 23.1% and diagnostic rate of 11.4%. Above results were consistent with those of Sanger sequencing. CONCLUSION: SAT is a simple, rapid and accurate method featuring high detection rate for common mutations related to deafness among the Chinese population and has provided an effective means of genetic testing for hereditary deafness.

The peripheral blood transcriptome identifies dysregulation of inflammatory response genes in polycystic ovary syndrome.

Polycystic ovary syndrome (PCOS) is a common endocrine disorder in women, resulting in ovulation failure and other metabolic problems. However, the underlying mechanisms of it remain largely uncertain due to the complexity of clinical manifestations. This systemic disorder is involved in endocrine, metabolism, immune system and many organs, and few studies have explored peripheral blood transcriptome in patients with PCOS. We performed gene expression profiling of peripheral blood from 8 PCOS patients and eight healthy women with microarray. The significance analysis of microarray (SAM) software was employed to screen the differentially expressed genes (DEGs) and gene ontology (GO) was used for functional enrichment analysis. In total, 181 DEGs with fold-changes >2.0 and q-values <0.05 were identified between the two groups. Among them, 149 were up-regulated and 32 down-regulated in PCOS. Unsupervised clustering of expressed genes could readily differentiate PCOS from control. More importantly, inflammatory response pathway including 14 dysregulated genes was highly enriched in PCOS. Furthermore, 10 DEGs were validated using quantitative reverse-transcription PCR (qRT-PCR) assays. Our study provides independent evidence for the involvement of systemic inflammatory response in PCOS and it may facilitate a greater understanding of this complex disease.

The 'plant economic spectrum' in bryophytes, a comparative study in subalpine forest.

PREMISE OF THE STUDY: Tradeoffs among functional traits of vascular plants are starting to be better understood, but it is unclear whether bryophytes possess similar tradeoffs or how trait relationships, or the 'economic spectrum', differ between the two groups. METHODS: We determined functional-trait values [including shoot mass per area (SMA), light-saturated assimilation rate (Amass), dark respiration rate (Rdmass), N and P concentrations (Nmass and Pmass), and photosynthetic N and P use efficiency (PNUE and PPUE)] and their bivariate relationships for 28 bryophytes growing in a subalpine old-growth fir forest on the eastern Tibetan Plateau. Trait values and scaling relationships of these bryophytes were compared with data for vascular plant leaves from the Global Plant Trait Network (GLOPNET) dataset. KEY RESULTS: We found that the Amass, Nmass, N:P, PNUE and PPUE of bryophyte shoots were lower than those of vascular plant leaves. In contrast, bryophytes possessed higher Pmass and the two groups had similar values of SMA and Rdmass. The Nmass and Pmass were closely associated with Amass and Rdmass, and these traits were all significantly negatively related to SMA. Metabolic rates increased faster with nutrient concentrations in bryophytes than in vascular plants. CONCLUSIONS: Our research indicates that bryophytes have similar trait relationships as vascular plant leaves, although the slopes of the relationships differ for most trait combinations. This study confirms a functional-trait tradeoff in bryophytes, and reveals that bryophytes allocate greater proportions of N and P into the metabolic pools.