9 results about "Molecular function" patented technology

This invention belongs to the fields of genetic engineering and plant cultivation technology, specifically relating to a method for inducing flavonoid synthesis in kale through selenium induction. BroMYB34 Genes and their applications. Specifically, this invention is the first to identify and clone a selenium-induced gene in kale. BroMYB34 The gene was used to elucidate its molecular function as a transcription factor regulating flavonoid synthesis through overexpression. BroMYB34 Genes can significantly increase the flavonoid content of kale and enhance its antioxidant capacity, providing an efficient and targeted genetic engineering approach for quality improvement of kale; at the same time, BroMYB34 Genes in cruciferous plants have potential functional conservation and can be extended to the quality improvement of various vegetables. This is of great significance for promoting the breeding and industrialization of functional vegetables and therefore has good practical application value.

An ornithine decarboxylase gene related to Auricularia auricula-judae degradation and application thereof belong to the field of bioengineering technology. In order to solve the technical problem that there is lack of key functional genes and molecular targets which can be used for early identification, anti-degradation molecular breeding and strain rejuvenation due to continuous subculture leading to degradation in Auricularia auricula-judae industry, based on the G1, G10 and G20 mycelium transcriptome difference analysis of Auricularia auricula-judae strain Heiwei S1908, the ornithine decarboxylase gene g4162 significantly differentially expressed between normal mycelium and G20 degraded mycelium is screened, the molecular function of the gene in the degradation of Auricularia auricula-judae subculture is determined, and the system biology information prediction of the gene coding protein is completed, which provides core gene resources and theoretical basis for the degradation mechanism research, molecular detection kit development, anti-degradation germplasm screening and gene breeding of Auricularia auricula-judae.

The application belongs to the field of plant molecular biology and genetic engineering, and specifically discloses a negative regulation Gossypium hirsutum fruit branch length trait gene Gh_A08G254200, which is located on A08 chromosome 122390509~122395098, the gene is composed of two exons and one intron, there is a SNP site at A08:122392954, which is a non-synonymous mutation, the gene is named as GhCSLD3, the CDS sequence length is 3432 bp, the nucleotide sequence of the gene comprises SEQ ID No.1; the amino acid sequence of the encoded protein comprises SEQ ID No.2. In the extreme Gossypium hirsutum non-zero type cotton variety, the gene is silenced, and it is found that the Gossypium hirsutum fruit branch length is elongated, so the gene has a negative regulation effect on the Gossypium hirsutum fruit branch elongation. The verification of the molecular function of the gene has important significance for improving the breeding of Gossypium hirsutum plant type change.

The application discloses a new application of m6A modified gene GAS6 and its receptor MERTK in rheumatoid arthritis. Multi-omics integrated analysis of synovial tissue and peripheral blood mononuclear cells of RA patients reveals common dysregulation of transcriptomics and epitranscriptomics, highlighting genes with both differential expression and m6A modification, which are enriched in processes such as phagocytosis, Th17 differentiation and cell aging. Among these genes, GAS6 shows the most significant m6A hypermethylation and expression up-regulation, and is verified as a key effector molecule interacting with MERTK / AXL receptor. Functional experiments show that GAS6 and MERTK synergistically promote the malignant phenotype of RA fibroblast-like synoviocytes, enhance their proliferation, migration, inflammatory cytokine secretion and anti-apoptotic ability.

This invention relates to the field of microelectronic device technology, and more particularly to a single-molecule memristor-rectifier device based on supramolecular assembly and its fabrication method. The fabrication method involves covalently linking a guest molecule B to the ends of a graphene electrode pair, ensuring a stable and tight interfacial bond between the supramolecular functional molecule and the electrode. Subsequently, the graphene electrode pair connected to the guest molecule B is immersed in a solution containing a host molecule A and a guest molecule C, forming a supramolecular functional molecule between the graphene electrode pairs. The guest molecule C in the supramolecular functional molecule possesses intrinsic redox activity; its molecular structure itself can undergo reversible electron gain and loss processes, thereby achieving controllable and reversible adjustment of the intramolecular charge distribution. Under the influence of an electric field, the reversible redox reaction of the guest molecule C triggers significant changes in the intramolecular charge distribution and electron transport characteristics, ultimately achieving reliable switching between high and low resistance states of the device.

This invention discloses a multimodal cervical cancer immune subtype prediction method based on deep learning, including the following steps: (1) Data acquisition: acquiring pathological whole slide image data and transcriptome sequencing data of cervical cancer patients; (2) Data preprocessing: performing tissue segmentation, multi-scale block division and normalization processing on the whole slide images, and performing protein-coding gene screening and standardization processing on the transcriptome data; (3) Model construction: constructing an image feature extraction branch based on attention mechanism convolutional neural network, a gene feature extraction branch based on deep fully connected network, and a multimodal fusion module based on feature splicing or decision weighting; (4) Model training: using weighted cross-entropy loss function and cosine annealing strategy to train the model and optimize parameters to solve the problem of sample class imbalance; (5) Model prediction: this invention effectively overcomes the information bottleneck of single-modal data by synergistically fusing the spatial morphological features of pathological images and the molecular functional features of gene expression, and significantly improves the accuracy and robustness of cervical cancer C1 / C2 immune subtype prediction.

The application discloses a plasma exosome ncRNA biomarker of recurrent depression and application thereof, and the biomarker comprises miR-618 and / or miR-223-3p. It is verified by RT-qPCR that the average expression levels of miR-618 and miR-223-3p of patients with recurrent depression are significantly higher than those of healthy controls, ROC curve analysis shows that miR-618 and miR-223-3p both show good diagnostic performance, both have the potential of being biomarkers of recurrent depression, and have good application prospect. In addition, the molecular function of miR-618 is verified through in-vivo experiments, and the result shows that miR-618 regulates PI3K-Akt and other signal pathways, and then causes depressive behaviors.

This invention relates to an intelligent system for precise typing of tumor-infiltrating immune cells. The system includes a pathological data acquisition module, a data preprocessing module, a multimodal interaction module, a typing prediction model construction module, and an immune cell typing prediction module. The invention obtains raw data through data acquisition; employs data preprocessing methods including coarse data processing, data standardization, preliminary feature extraction, and dataset segmentation; utilizes a multimodal fusion model to integrate multimodal features, enabling deep alignment and interaction of data from different modalities under the guidance of prior knowledge, thereby revealing the intrinsic relationship between immune cell morphology and molecular function and enhancing the complementarity of feature representations; and employs a multi-scale deep network model as the typing prediction model, simultaneously capturing immune cell morphological patterns and the overall biological context, enabling the analysis of multi-scale features of immune cell typing, thus improving the accuracy and generalization of typing.