MSFN: a multi-omics stacked fusion network for breast cancer survival prediction.

Introduction: Developing effective breast cancer survival prediction models is critical to breast cancer prognosis. With the widespread use of next-generation sequencing technologies, numerous studies have focused on survival prediction. However, previous methods predominantly relied on single-omics data, and survival prediction using multi-omics data remains a significant challenge. Methods: In this study, considering the similarity of patients and the relevance of multi-omics data, we propose a novel multi-omics stacked fusion network (MSFN) based on a stacking strategy to predict the survival of breast cancer patients. MSFN first constructs a patient similarity network (PSN) and employs a residual graph neural network (ResGCN) to obtain correlative prognostic information from PSN. Simultaneously, it employs convolutional neural networks (CNNs) to obtain specificity prognostic information from multi-omics data. Finally, MSFN stacks the prognostic information from these networks and feeds into AdaboostRF for survival prediction. Results: Experiments results demonstrated that our method outperformed several state-of-the-art methods, and biologically validated by Kaplan-Meier and t-SNE.

The characteristic VOCs of different parts of Artocarpus heterophyllus fruit based on HS-SPME-GC-MS and PTR-TOF-MS.

Insight investigation on both edible pulps and inedible parts involving inflorescence axis and shreds of Artocarpus heterophyllus Lam were carried out, a total of 98 VOCs and 201 masses were identified by the combination of HS-SPME-GC-MS and PTR-TOF-MS. Among them, according to the consistency of OAV and results of VIP > 1, p < 0.05, compounds methyl isovalerate (A2), 3-methylbutyl acetate (A5) and octanoic acid, ethyl ester (A21) were recognized as aroma markers to distinguish the pulps, shreds and inflorescence axis. Meanwhile, the inflorescence axis (IC50: 1.82 mg/mL) and shreds (IC50: 16.74 mg/mL) exhibited more excellent antioxidant potency than pulps (IC50: 17.43 mg/mL) in vitro. These findings validated the feasibility of coupling HS-SPME-GC-MS and PTR-TOF-MS for rapid detection of characteristic VOCs of this plant, and offered new prospect of fragrance utilization and waste management of the edible and inedible parts of A. heterophyllus fruit.

The auxin-responsive CsSPL9-CsGH3.4 module finely regulates auxin levels to suppress the development of adventitious roots in tea (Camellia sinensis).

The cutting technique is extensively used in tea breeding, with key emphasis on promoting the growth of adventitious roots (ARs). Despite its importance in tea cultivation, the mechanisms underlying AR development in tea remain unclear. In this study, we demonstrated the essential role of auxins in the initiation and progression of AR and established that the application of exogenous 1-naphthaleneacetic acid-enhanced AR formation in tissue-cultured seedlings and cuttings. Then, we found that the auxin-responsive transcription factor CsSPL9 acted as a negative regulator of AR development by reducing the levels of free indole-3-acetic acid (IAA) in tea plants. Furthermore, we identified CsGH3.4 as a downstream target of CsSPL9, which was activated by direct binding to its promoter. CsGH3.4 also inhibited AR development and maintained low levels of free IAA. Thus, these results revealed the inhibitory effect of the auxin-responsive CsSPL9-CsGH3.4 module on AR development by reducing free IAA levels in tea. These findings have significant theoretical and practical value for enhancing tea breeding practices.

STAT transcription factor is indispensable for oogenesis in silkworm.

Signal Transducer and Activator of Transcription (STAT) proteins represent a critical transcription factor family with multifaceted roles in diverse fundamental eukaryotic processes. In Drosophila, STAT exerts a pivotal regulatory influence on oogenesis, governing the early differentiation of follicular cells and ensuring proper encapsulation of germ-line cells. However, the role of STAT in egg development in silkworms remains unknown. In the present study, using CRISPR/Cas9 technology, we successfully generated a strain of silkworms with targeted deletion of the STAT-L gene, which resulted in significant reproductive abnormalities observed in female moths, including shortened fallopian tubes and reduced egg production. The ovaries dissected from STAT-L knockout silkworms during the pupal stage of silkworm exhibited varying degrees of fusion among egg chambers. Additionally, paraffin sections of prepupal ovaries also revealed evidence of egg chambers fusion. To elucidate the molecular mechanism underlying the role of the STAT-L gene regulation on egg development in silkworm, we performed ovarian transcriptomic analysis following STAT-L knockout. Our findings indicated that STAT-L gene can modulate Notch signaling pathway by down-regulating APH-1 gene expression. These results suggest that STAT-L gene plays a crucial role in normal egg chamber formation in silkworms, potentially through its influence on Notch signaling pathway expression.

DP-SSLoRA: A privacy-preserving medical classification model combining differential privacy with self-supervised low-rank adaptation.

BACKGROUND AND OBJECTIVE: Concerns about patient privacy issues have limited the application of medical deep learning models in certain real-world scenarios. Differential privacy (DP) can alleviate this problem by injecting random noise into the model. However, naively applying DP to medical models will not achieve a satisfactory balance between privacy and utility due to the high dimensionality of medical models and the limited labeled samples. METHODS: This work proposed the DP-SSLoRA model, a privacy-preserving classification model for medical images combining differential privacy with self-supervised low-rank adaptation. In this work, a self-supervised pre-training method is used to obtain enhanced representations from unlabeled publicly available medical data. Then, a low-rank decomposition method is employed to mitigate the impact of differentially private noise and combined with pre-trained features to conduct the classification task on private datasets. RESULTS: In the classification experiments using three real chest-X ray datasets, DP-SSLoRA achieves good performance with strong privacy guarantees. Under the premise of ɛ=2, with the AUC of 0.942 in RSNA, the AUC of 0.9658 in Covid-QU-mini, and the AUC of 0.9886 in Chest X-ray 15k. CONCLUSION: Extensive experiments on real chest X-ray datasets show that DP-SSLoRA can achieve satisfactory performance with stronger privacy guarantees. This study provides guidance for studying privacy-preserving in the medical field. Source code is publicly available online. https://github.com/oneheartforone/DP-SSLoRA.

MPASL: multi-perspective learning knowledge graph attention network for synthetic lethality prediction in human cancer.

Synthetic lethality (SL) is widely used to discover the anti-cancer drug targets. However, the identification of SL interactions through wet experiments is costly and inefficient. Hence, the development of efficient and high-accuracy computational methods for SL interactions prediction is of great significance. In this study, we propose MPASL, a multi-perspective learning knowledge graph attention network to enhance synthetic lethality prediction. MPASL utilizes knowledge graph hierarchy propagation to explore multi-source neighbor nodes related to genes. The knowledge graph ripple propagation expands gene representations through existing gene SL preference sets. MPASL can learn the gene representations from both gene-entity perspective and entity-entity perspective. Specifically, based on the aggregation method, we learn to obtain gene-oriented entity embeddings. Then, the gene representations are refined by comparing the various layer-wise neighborhood features of entities using the discrepancy contrastive technique. Finally, the learned gene representation is applied in SL prediction. Experimental results demonstrated that MPASL outperforms several state-of-the-art methods. Additionally, case studies have validated the effectiveness of MPASL in identifying SL interactions between genes.

Rapid detection of markers in green coffee beans with different primary processing treatments of Coffea arabica L. from Yunnan.

The characteristic flavor of Coffea arabica from Yunnan is largely attributed to the primary processing treatments through affecting the VOCs accumulation. Therefore, a rapid and comprehensive detection technique is needed to accurately recognize VOCs in green coffee beans with different pretreatment methods. Hence, we conducted volatile profiles and identified nine markers of three different primary processed green coffee beans from the major production areas in Yunnan with the combined of HS-SPME-GC-MS and PTR-TOF-MS. The relationships between the chemical composition and the content of VOCs in green coffee beans were elucidated. Among the markers, palmitic acid (F3), linoleic acid (F6), α-ethylidene phenylacetaldehyde (T4), and phytane (T8) contributed to the antioxidant activity of sun-exposed green coffee beans. In conclusion, the analytical technology presented here provided a general tool for an overall and rapid understanding of a detailed volatile profiles of green coffee beans in Yunnan.

Maternal tobacco exposure during pregnancy and atopic dermatitis in offspring: A systematic review and meta-analysis.

The main purpose of this review was to examine the evidence of the relationship between active smoking or passive smoking during pregnancy and atopic dermatitis in offspring. The protocol was written following the PRISMA Checklist and was registered in the PROSPERO database (registration number CRD42022381136). We implemented a comprehensive search in PubMed, Embase and Web of Science databases to identify all potentially related articles from inception through 1 December 2022. We assessed cohort studies and case-control studies using the Newcastle-Ottawa Scale (NOS), and the Joanna Briggs Institute (JBI) critical appraisal tool to assess the quality of cross-sectional studies. Heterogeneity was investigated by using Cochrane Q tests and I2 statistics. In addition, according to the research design, population source and population size, the reasons for the heterogeneity were analysed. A total of 15 observational studies were included in this analysis. Our meta-analysis suggests that atopic dermatitis in offspring is not associated with active smoking during pregnancy (pooled OR, 0.96 [95% CI 0.86-1.07]); however, it is related to passive smoking (OR, 1.52 [95% CI 1.36-1.70]). Passive smoking during pregnancy is associated with an increased risk of eczema development in offspring. More research is needed to explore the risk of active smoking and eczema development in offspring, especially the association between measurements of pregnancy cotinine levels in maternal body fluids and AD in offspring.

Drug-drug interactions prediction based on deep learning and knowledge graph: A review.

Drug-drug interactions (DDIs) can produce unpredictable pharmacological effects and lead to adverse events that have the potential to cause irreversible damage to the organism. Traditional methods to detect DDIs through biological or pharmacological analysis are time-consuming and expensive, therefore, there is an urgent need to develop computational methods to effectively predict drug-drug interactions. Currently, deep learning and knowledge graph techniques which can effectively extract features of entities have been widely utilized to develop DDI prediction methods. In this research, we aim to systematically review DDI prediction researches applying deep learning and graph knowledge. The available biomedical data and public databases related to drugs are firstly summarized in this review. Then, we discuss the existing drug-drug interactions prediction methods which have utilized deep learning and knowledge graph techniques and group them into three main classes: deep learning-based methods, knowledge graph-based methods, and methods that combine deep learning with knowledge graph. We comprehensively analyze the commonly used drug related data and various DDI prediction methods, and compare these prediction methods on benchmark datasets. Finally, we briefly discuss the challenges related to drug-drug interactions prediction, including asymmetric DDIs prediction and high-order DDI prediction.

Microbial community, pathogenic bacteria and high-risk anti-biotic resistance genes at two tropical coastal beaches adjacent to villages in Hainan, China.

OBJECTIVE: The aim of the study was to explore the correlation between characteristics of microbial community, pathogenic bacteria and high-risk antibiotic-resistant genes, between coastal beaches and a multi-warm-blooded host, as well as to determine potential species biomarkers for faecal source contamination on tropical coastal beaches in China. MATERIAL AND METHODS: The 'One-Health' approach was used in a microbiological study of beaches and warm-blooded hosts. The microbial.community was analyzed using 16S rRNA gene amplicons and shotgun metagenomics on Illumina NovaSeq. RESULTS: The Chao, Simpson, Shannon, and ACE indices of non-salt beach were greater than those of salt beaches at the genus and OTU levels (P < 0.001). Bacteroidota, Halanaerobiaeota, Cyanobacteria, and Firmicutes were abundant on salt beaches (P<0.01). Human-sourced microorganisms were more abundant on salt beaches, which accounted for 0.57%. Faecalibacterium prausnitzii and Eubacterium hallii were considered as reliable indicators for the contamination of human faeces. High-risk carbapenem-resistant Klebsiella pneumoniae and the genotypes KPC-14 and KPC-24 were observed on salt beaches. Tet(X3)/tet(X4) genes and four types of MCR genes co-occurred on beaches and humans; MCR9.1 accounted for the majority. Tet(X4) found among Cyanobacteria. Although rarely reported at Chinese beaches, pathogens, such as Vibrio vulnificus, Legionella pneumophila, and Helicobacter pylori, were observed. CONCLUSIONS: The low microbial community diversity, however, did not indicate a reduced risk. The transfer of high-risk ARGs to extreme coastal environments should be given sufficient attention.