A newly-discovered tea population variety processed Bai Mu Dan white tea: Flavor characteristics and chemical basis.

The quality of white tea (WT) is impacted by selected tea cultivars. To explore the organoleptic quality of a recently-discovered WT ("Caicha", CC), HS-SPME/GC-MS and UPLC were employed to identify volatile and non-volatile compounds in tea samples. Multiple statistical methods demonstrated the distinctions between CC and four mainstream WT varieties from main producing areas. CC exhibited abundant volatile alcohol, terpenoids, ketone, aldehyde and ester, as well as non-volatile lignans and coumarins, phenolic acids and low-molecular carbohydrates. These substances combinedly contributed to the flavor attributes of CC, characterized by an intense herbal/citrus-like cleanness and flower/fruit-like sweetness, scarce in existing commercial WT varieties. Sensory evaluation corroborated these findings. In conclusion, we have processed a new tea variety (CC) with WT manufacturing technology, and discovered the unique cleanness and sweetness of it. This study enriches the raw material database for WT production and blending, and boosts the development of more premium WT varieties.

Aroma Identification and Classification in 18 Kinds of Teas (Camellia sinensis) by Sensory Evaluation, HS-SPME-GC-IMS/GC × GC-MS, and Chemometrics.

Tea (Camellia sinensis) is one of the most popular beverages worldwide. Many types of tea products continuously emerge in an endless stream; so, the classification of tea becomes more difficult. Aroma is a vital indicator of tea quality. The present study deals with the identification of aroma compounds in 18 different kinds of tea belonging to three typical tea varieties, including green tea, oolong tea, and black tea, using GC-IMS and GC × GC-O-MS. Moreover, the clustering of all 18 tea samples and the in depth correlation analysis between sensory evaluation and instrumental data were performed using the PCA and OPLS-DA. The results revealed that in all 18 kinds of tea, a total of 85 aroma compounds were detected by GC-IMS, whereas 318 were detected by GC × GC-O-MS. The PCA result revealed that green tea, oolong tea, and black tea could be clearly separated based on their peak areas. The OPLS-DA result showed that a total of 49 aroma compounds with VIP value > 1.0 could be considered as the potential indicators to quickly classify or verify tea types. This study not only compared the aroma differences across different types of teas, but also provided ideas for the rapid monitoring of tea quality and variety.

Production regions discrimination of Huangguanyin oolong tea by using the content of chemical components and rare earth elements.

Oolong tea is one of the most popular tea beverages in China. Tea cultivars, processing technology and origin of production affect the quality and price of oolong teas. To investigate the differences in Huangguanyin oolong tea from different production regions, the chemical components, mineral elements and rare earth elements of Huangguanyin oolong tea produced in Yunxiao (YX) and Wuyishan (WY) were analyzed by using spectrophotometry methods, targeted metabolomics and inductive plasma coupled mass spectrometry (ICP-MS). The results of spectrophotometry methods revealed that there were significant differences in thearubigin, tea polyphenols and water extract between Huangguanyin oolong teas from different production regions. Targeted metabolomics identified a total of 31 chemical components in Huangguanyin oolong teas from the two production regions, of which 14 chemical components were significantly different and contributed to the regional differentiation of Huangguanyin oolong tea. Yunxiao Huangguanyin had relatively higher contents of (-)-Epigallocatechin-3-O-(3-O-methylgallate) (EGCG3″Me), ornithine (Orn) and histidine (His), while Wuyishan Huangguanyin had relatively higher contents of glutamic acid (Glu), γ-aminobutyric acid (GABA), ß-aminobutyric acid (ß-ABA) and other components. Moreover, ICP-MS identified a total of 15 mineral elements and 15 rare earth elements in Huangguanyin oolong tea from the two production regions, of which 15 elements were significantly different between YX and WY, and contributed to the regional differentiation of Huangguanyin oolong tea. K had a relatively higher content in Yunxiao Huangguanyin, while rare earth elements had relatively higher contents in Wuyishan Huangguanyin. The classification results by the production region showed that the discrimination rate of the support vector machine (SVM) model based on the 14 different chemical components reached 88.89%, while the SVM model based on the 15 elements reached 100%. Therefore, we used targeted metabolomics and ICP-MS techniques to screen and explore the chemical components, mineral elements and rare earth elements differences among two production regions, which indicated the feasibility of Huangguanyin oolong tea classification by production regions in the study. The results will provide some reference for the distinction between the two production regions of Huangguanyin oolong tea.

Attention-wise masked graph contrastive learning for predicting molecular property.

MOTIVATION: Accurate and efficient prediction of the molecular property is one of the fundamental problems in drug research and development. Recent advancements in representation learning have been shown to greatly improve the performance of molecular property prediction. However, due to limited labeled data, supervised learning-based molecular representation algorithms can only search limited chemical space and suffer from poor generalizability. RESULTS: In this work, we proposed a self-supervised learning method, ATMOL, for molecular representation learning and properties prediction. We developed a novel molecular graph augmentation strategy, referred to as attention-wise graph masking, to generate challenging positive samples for contrastive learning. We adopted the graph attention network as the molecular graph encoder, and leveraged the learned attention weights as masking guidance to generate molecular augmentation graphs. By minimization of the contrastive loss between original graph and augmented graph, our model can capture important molecular structure and higher order semantic information. Extensive experiments showed that our attention-wise graph mask contrastive learning exhibited state-of-the-art performance in a couple of downstream molecular property prediction tasks. We also verified that our model pretrained on larger scale of unlabeled data improved the generalization of learned molecular representation. Moreover, visualization of the attention heatmaps showed meaningful patterns indicative of atoms and atomic groups important to specific molecular property.

Complete chloroplast genomes of five classical Wuyi tea varieties (Camellia sinensis, Synonym: Thea bohea L.), the most famous Oolong tea in China.

Wuyi tea (Camellia. sinensis, Synonym: Thea Bohea L.) is recognized as the most prestigious oolong tea in China. For germplasm identification and protection, the complete chloroplast genomes of five classical Wuyi tea varieties were determined by next-generation sequencing. These chloroplast genomes showed highly conserved structures and are 157,024-157,126 bp in length, consisting of a pair of reverse repeats (IR) regions of 25,944-26,095 bp, one large single-copy (LSC) region of 86,594-86,859 bp, and one small single copy (SSC) region of 18,276-18,291 bp. A total of 137 genes were observed and overall GC contents were all about 37.3%. Phylogenetic analysis revealed Wuyi tea varieties did not cluster together, suggesting that these Wuyi tea varieties might have diverged early in their evolutionary history and the complete chloroplast genome could be used as a super-barcode to identify these varieties. This study will be valuable for future studies of evolution and intraspecific identification in Wuyi tea.

Graph2MDA: a multi-modal variational graph embedding model for predicting microbe-drug associations.

MOTIVATION: Accumulated clinical studies show that microbes living in humans interact closely with human hosts, and get involved in modulating drug efficacy and drug toxicity. Microbes have become novel targets for the development of antibacterial agents. Therefore, screening of microbe-drug associations can benefit greatly drug research and development. With the increase of microbial genomic and pharmacological datasets, we are greatly motivated to develop an effective computational method to identify new microbe-drug associations. RESULTS: In this article, we proposed a novel method, Graph2MDA, to predict microbe-drug associations by using variational graph autoencoder (VGAE). We constructed multi-modal attributed graphs based on multiple features of microbes and drugs, such as molecular structures, microbe genetic sequences and function annotations. Taking as input the multi-modal attribute graphs, VGAE was trained to learn the informative and interpretable latent representations of each node and the whole graph, and then a deep neural network classifier was used to predict microbe-drug associations. The hyperparameter analysis and model ablation studies showed the sensitivity and robustness of our model. We evaluated our method on three independent datasets and the experimental results showed that our proposed method outperformed six existing state-of-the-art methods. We also explored the meaning of the learned latent representations of drugs and found that the drugs show obvious clustering patterns that are significantly consistent with drug ATC classification. Moreover, we conducted case studies on two microbes and two drugs and found 75-95% predicted associations have been reported in PubMed literature. Our extensive performance evaluations validated the effectiveness of our proposed method. AVAILABILITY AND IMPLEMENTATION: Source codes and preprocessed data are available at https://github.com/moen-hyb/Graph2MDA. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Randomised clinical trial: dose optimising strategy by NUDT15 genotyping reduces leucopenia during thiopurine treatment of Crohn's disease.

INTRODUCTION: Thiopurine S-methyltransferase (TPTM) is a well known biomarker for thiopurine-induced leucopenia, which has limited value in Asia. Instead, NUDT15 C415T is a promising predictor in Asia. AIMS: To explore whether an optimised strategy based on NUDT15 C415T genotypes affects thiopurine-induced leucopenia, as well as efficacy in Chinese patients with Crohn's disease. METHODS: Patients with Crohn's disease and indications for thiopurines were included from two hospitals in China. They were randomly assigned to either the intervention or the control group. In the intervention group, those with genotype CC received a standard dose (control group), those with CT genotype received 50% of the standard dose, those with TT genotype received alternative drugs. The primary endpoint was thiopurine-induced leucopenia (<3.5 × 109 /L). Secondary outcomes were the incidence of other adverse events and the efficacy for maintaining steroid-free remission at week 36. RESULTS: The rate of thiopurine-induced leucopenia was lower in the intervention group (n = 52) than in the control group (n = 66) (23.7% vs 32.4%, P = 0.049, RR = 0.73, 95% CI 0.53-1.00). In CT subgroup, the incidence of leucopenia in the intervention group (n = 10) was significantly lower than in the control group (n = 28) (31.3% vs 65.1%, RR = 0.48, 95% CI 0.28-0.84). Neither other adverse events nor treatment efficacy was significantly different between the two groups during follow-up. CONCLUSIONS: Among Chinese patients with Crohn's disease, dose optimisation by NUDT15 C415T reduced the rate of thiopurine-induced leucopenia, without significant influence on efficacy. Using 50% dose reduction for heterozygotes, and alternative drugs for homozygotes, are practicable strategies. Clinical trial number: NCT02929706.

SMALF: miRNA-disease associations prediction based on stacked autoencoder and XGBoost.

BACKGROUND: Identifying miRNA and disease associations helps us understand disease mechanisms of action from the molecular level. However, it is usually blind, time-consuming, and small-scale based on biological experiments. Hence, developing computational methods to predict unknown miRNA and disease associations is becoming increasingly important. RESULTS: In this work, we develop a computational framework called SMALF to predict unknown miRNA-disease associations. SMALF first utilizes a stacked autoencoder to learn miRNA latent feature and disease latent feature from the original miRNA-disease association matrix. Then, SMALF obtains the feature vector of representing miRNA-disease by integrating miRNA functional similarity, miRNA latent feature, disease semantic similarity, and disease latent feature. Finally, XGBoost is utilized to predict unknown miRNA-disease associations. We implement cross-validation experiments. Compared with other state-of-the-art methods, SAMLF achieved the best AUC value. We also construct three case studies, including hepatocellular carcinoma, colon cancer, and breast cancer. The results show that 10, 10, and 9 out of the top ten predicted miRNAs are verified in MNDR v3.0 or miRCancer, respectively. CONCLUSION: The comprehensive experimental results demonstrate that SMALF is effective in identifying unknown miRNA-disease associations.

Combined Detection of NUDT15 Variants Could Highly Predict Thiopurine-induced Leukopenia in Chinese Patients with Inflammatory Bowel Disease: A Multicenter Analysis.

BACKGROUND: NUDT15 c.415C>T was a novel genetic marker confirmed in our center for thiopurine-induced leukopenia in Chinese inflammatory bowel disease (IBD). For validation, a large cohort study is needed. Meanwhile, the newly discovered NUDT15 coding variants (c.36_37insGGAGTC and c.52 G>A) have not been studied in patients with IBD. We aimed to further confirm the influence of 3 NUDT15 variants (c.415C>T, c.36_37insGGAGTC, and c.52G>A) on thiopurine-induced leukopenia in Chinese patients with IBD. METHODS: Patients prescribed on thiopurines for at least 2 weeks were recruited from 4 tertiary hospitals. Clinical data were collected. NUDT15 genotypes were determined with polymerase chain reaction-RFLP and sequencing. The interactions between variants and leukopenia were analyzed. RESULTS: A total of 732 patients were included, 177 (24.3%) of whom developed leukopenia. There were strong associations of NUDT15 c.415C>T, c.36_37insGGAGTC, and c.52G>A with thiopurine-induced leukopenia (P = 1.81 × 10, P = 4.74 × 10 and P = 0.04, respectively), whereas there was no relevance for thiopurine S-methyltransferase genotypes (P = 0.25). The predictive sensitivity of NUDT15 c.415C>T was 49.2%, whereas it increased to 55.4% when combined analysis with c.36_37insGGAGTC and c.52G>A. Notably, not only the homozygotes with NUDT15 c.415C>T but also the heterozygotes both carrying c.415C>T and c.52G>A developed early leukopenia. The median dosage for NUDT15 c.415C>T carriers was significantly lower than that for wild-type (P < 0.001). CONCLUSIONS: We confirmed that NUDT15 c.415C>T, c.36_37insGGAGTC, and c.52G>A variants were risk factors for thiopurine-induced leukopenia. Combined detection of the 3 variants could increase the predictive sensitivity of thiopurine-induced leukopenia and help to distinguish early leukopenia in heterozygote of c.415C>T in Chinese patients with IBD. Treatment monitoring by NUDT15 variants may be promising in individualized therapy.