Enhancing drug property prediction with dual-channel transfer learning based on molecular fragment.

BACKGROUND: Accurate prediction of molecular property holds significance in contemporary drug discovery and medical research. Recent advances in AI-driven molecular property prediction have shown promising results. Due to the costly annotation of in vitro and in vivo experiments, transfer learning paradigm has been gaining momentum in extracting general self-supervised information to facilitate neural network learning. However, prior pretraining strategies have overlooked the necessity of explicitly incorporating domain knowledge, especially the molecular fragments, into model design, resulting in the under-exploration of the molecular semantic space. RESULTS: We propose an effective model with FRagment-based dual-channEL pretraining (FREL). Equipped with molecular fragments, FREL comprehensively employs masked autoencoder and contrastive learning to learn intra- and inter-molecule agreement, respectively. We further conduct extensive experiments on ten public datasets to demonstrate its superiority over state-of-the-art models. Further investigations and interpretations manifest the underlying relationship between molecular representations and molecular properties. CONCLUSIONS: Our proposed model FREL achieves state-of-the-art performance on the benchmark datasets, emphasizing the importance of incorporating molecular fragments into model design. The expressiveness of learned molecular representations is also investigated by visualization and correlation analysis. Case studies indicate that the learned molecular representations better capture the drug property variation and fragment semantics.

Distribution and Interactive Effects of Heavy Metals in Soil-Maize (Zea Mays L.) System in the Mercury Mining Area, Southwestern China.

The concentrations and interactive effects of beneficial elements (i.e., Se, Mo, and Zn) and heavy metals (As, Cd, Hg, and Pb) of maize (Zea mays L.) grown on lime soil and/or soil with mercury tailing were investigated in this study. The results show that the concentrations of heavy metals (i.e., As, Hg, and Pb) in soil with tailing were higher than those in lime soil. The concentrations of beneficial elements (i.e., Mo and Zn) in maize grown on soil with tailing were higher than those of maize grown on lime soil. The mean concentrations of Se, Mo, and Zn in maize grown on soil with tailing were 3.67 mg/kg, 0.530 mg/kg, and 27.4 mg/kg. The pH and an antagonistic effect played an important role in the concentrations of Mo and Zn in maize. The Se concentration in maize was controlled by the planting media.

Distribution, accumulation, and potential risks of heavy metals in soil and tea leaves from geologically different plantations.

Risk assessment regarding heavy metals in tea is crucial to ensure the health of tea customers. However, the effects of geological difference on distribution of heavy metals in soils and their accumulation in tea leaves remain unclear. This study aimed to estimate the impacts of geological difference on distribution of cadmium (Cd), lead (Pb), thallium (Tl), mercury (Hg), arsenic (As), antimony (Sb), chromium (Cr), nickel (Ni), and manganese (Mn) in soils and their accumulation in tea leaves, and further evaluate their health risks. 22 soils and corresponding young tea leaves (YTL) and old tea leaves (OTL), from geologically different plantations, were sampled and analyzed. Results showed that heavy metals concentrations in soils, derived from Permian limestone and Cambrian weakly mineralized dolomite, were obviously greater than those from Silurian clastic rock. The geological difference controlled the distribution of soil heavy metals to a large extent. Contents of Cd, Tl, and Mn in tea leaves mainly depended on their contents in soils. Soil Hg, Pb, As, and Sb contents may not be the only influencing factors for their respective accumulation in tea leaves. More attentions should be paid to soil acidification of tea plantations to ensure the tea quality security. Target hazard quotients (THQ) of Cd, Pb, Tl, Hg, As, Sb, Cr, and Ni and hazard index (HI) via tea intake were below one, indicating no human health risk. The non-mineralized Silurian area was less at risk of heavy metals accumulation in tea leaves than the Cambrian metallogenic belt and the Permian Cd-enriched zone. This study could provide an important basis to understand and mitigate the potential risks of heavy metals in tea.

Use of mineral multi-elemental analysis to authenticate geographical origin of different cultivars of tea in Guizhou, China.

BACKGROUND: The geographical origin of tea (Camellia sinensis) can be traced using mineral elements in its leaves as fingerprints. However, the role that could be played by soil mineral elements in the geographical authentication of tea leaves has been unclear. In this study, 22 mineral elements in 73 pairs of tea leaves and soils from three regions (Pu'an, Duyun, and Liping) in Guizhou, China, were determined using inductively coupled plasma mass spectrometry (ICP-MS) and inductively coupled plasma atomic emission spectrometry (ICP-AES). The mineral element concentrations were processed by multivariate statistical analysis, including one-way analysis of variance (ANOVA), correlation analysis, principal component analysis (PCA), and stepwise linear discriminant analysis (S-LDA). RESULTS: Based on a one-way ANOVA, tea leaves and soils with different origins possessed unique mineral element fingerprints. Sixteen mineral element concentrations in tea leaves were significantly correlated with those in soils (P < 0.05). The geographical origins of tea leaves were effectively differentiated using the 16 correlated mineral elements combined with PCA. The S-LDA model offered a 100% differentiation rate, and six indicative elements (phosphorus, Sr, U, Pb, Cd, and Cr) were selected as important fingerprinting markers for the geographic traceability of tea leaves. The accurate discrimination rate of geographical origin was unaffected by the cultivars of tea in the S-LDA model. CONCLUSIONS: Mineral elements in soils played an important role in the geographical authentication of tea leaves. Mineral elemental concentrations with significant correlations between tea leaves and soils could be robust, and could be used to trace the geographical origins of tea leaves. © 2020 Society of Chemical Industry.

Studies on geochemical characteristics and biomineralization of Cambrian phosphorites, Zhijin, Guizhou Province, China.

Phosphate rocks, an important ore resource in Guizhou Province, China, are mainly hosted within the Sinian Doushantuo Formation and the Cambrian Meishucun Formation. In addition, the phosphate rocks of the Cambrian Meishucun Formation are rich in biological fossils. Although numerous studies investigating the genesis of phosphate deposits have been performed, the relationship between biological activity and the formation of phosphate deposits in the lower Cambrian Meishucun Formation has not been convincingly explained. This study focuses on the biological fossil assemblage, the characteristics of phosphorus, and the relationship between biological and phosphorus enrichment of the lower Cambrian phosphorites. The primary objectives of our study are to analyze the role of organisms in the formation of phosphorites, restore the phosphorus-formation environment of the Cambrian Meishucun Formation, and construct a sedimentary model of the phosphorites in the Meishucun Formation. The results indicate that there is a significant positive correlation between biological activity and the deposition of phosphorites, that is, the higher the degree of biological enrichment and differentiation, the stronger the deposition. The geochemical analysis of several profiles in the Zhijin phosphorite block shows that the phosphorite block was deposited in an oxygen-rich environment and was affected by a high-temperature hydrothermal fluid. Upwelling ocean currents supplied abundant phosphorus and other nutrients, which provided the conditions for small shells and algae to flourish. Biochemical activity was a crucial factor in the deposition of the phosphorite.

METTL3 alleviates D-gal-induced renal tubular epithelial cellular senescence via promoting miR-181a maturation.

Methyltransferase-like protein 3 (METTL3) mediated N6-Methyladenosine (m6A) modification has been implicated in many physiological and pathological processes. However, its function and mechanism in kidney aging are not entirely clear. Here, we investigated changes in m6A levels of aging kidneys and the role of METTL3 in senescent renal tubular epithelial cells and its potential mechanisms. First, we used the naturally aged mouse model and the D-galactose (D-gal)-induced aged mouse model. Dot blot and m6A RNA methylation quantification showed significantly decreased m6A levels in both models. In addition, we observed that METTL3 was down-regulated in D-gal-induced senescent human renal tubular epithelial cell line (HK-2). METTL3 reduction was associated with senescence-related phenotypes of HK-2 cells. We also found that miR-181a-5p attenuated HK-2 senescence by targeting the NF-κB pathway. Moreover, METTL3 was able to promote the maturation of miR-181a-5p and then inhibited the expression of NF-κB and IL-1α. Taken together, we demonstrate that the METTL3/miR-181a-5p/NF-κB axis counteracts HK-2 senescence. Our results suggest that METTL3 may be a novel biomarker and a potential therapy target for kidney aging.

Distribution and dispersion of heavy metals in the rock-soil-moss system of the black shale areas in the southeast of Guizhou Province, China.

Black shales are easily exposed due to human activities such as mining, road construction, and shale gas development, which results in several environmental issues including heavy metal (HM) pollution, soil erosion, and the destruction of vegetation. Mosses are widely used to monitor metal pollution in the atmosphere, but few studies on the distribution and dispersion of HMs in the rock-soil-moss system are available. Here, mosses (Pohlia flexuosa Harv. in Hook), growing soils, and corresponding parent rocks were collected from black shale areas. After appropriate pretreatment, samples were analyzed for multiple elemental concentrations by ICP-AES and ICP-MS. The results show that black shale parent rocks have elevated HM concentration and act as a source of multiple metals. The overlying soil significantly inherits and accumulates heavy metals released from black shale. Significant positive correlations between HMs in P. flexuosa and the growing soils indicate that HMs are mainly originating from geological source rather than atmospheric deposition. Differential accumulation of HMs is observed between rhizoids and stems in our study. Moreover, P. flexuosa is able to cope with high concentrations of toxic metals without any visible negative effect on its growth and development. Finally, the bioconcentration factor (BCF) for all the HMs in P. flexuosa is less than 1, indicating that it has a tolerance and exclusion mechanism for these metals, especially for the non-essential elements As and Pb. Therefore, the luxuriant and spontaneous growth of P. flexuosa could be used as a phytostabilization pioneer plant in the black shale outcrop where vascular plants are rare.

The Role of Soil Mineral Multi-elements in Improving the Geographical Origin Discrimination of Tea (Camellia sinensis).

The combination of mineral multi-elements with chemometrics can effectively trace the geographical origin of tea (Camellia sinensis). However, the role of soil mineral multi-elements in discriminating the origin of tea was unknown. This study aimed to further validate whether the geographical origin of tea can be authenticated based on mineral multi-elements, the concentrations of which in tea leaves were significantly correlated with those in soil. Eighty-seven tea leaves samples and paired soils from Meitan and Fenggang (MTFG), Anshun, and Leishan in China were sampled, and 24 mineral elements were measured. The data were processed using one-way analysis of variance (ANOVA), Pearson correlation analysis, principal component analysis (PCA), and stepwise linear discriminant analysis (SLDA). Results indicated that tea and soil samples from different origins differed significantly (p < 0.05) in terms of most mineral multi-elemental concentrations. Conversely, the intra-regional differences of different cultivars of the same origin were relatively minor. Seventeen mineral elements in tea leaves were significantly correlated with those in soils. The SLDA model, based on the 17 aforementioned elements, produced a 98.85% accurate classification rate. In addition, the origin was also identified satisfactorily with 94.25% accuracy when considering the cultivar effect. In conclusion, the tea plant cultivars unaffected the accuracy of the discrimination rate. The geographical origin of tea could be authenticated based on the mineral multi-elements with significant correlation between tea leaves and soils. Soil mineral multi-elements played an important role in identifying the geographical origin of tea.