4 results about "Gene classification" patented technology

The application discloses a landscape gene identification method based on machine learning, relates to the technical field of machine learning and data science, and comprises the following steps: constructing a multi-level landscape gene classification architecture, pre-training a machine learning model based on the multi-level landscape gene classification architecture, and obtaining a landscape feature identification model; then, acquiring landscape element data corresponding to a to-be-identified landscape, calling the landscape feature identification model to analyze the landscape element data, and obtaining a landscape index identification result; then, taking the landscape index identification result as the basis, acquiring a first landscape gene identification result by using a feature deconstruction method, acquiring a second landscape gene identification result by using a prototype-variation theory, and acquiring a third landscape gene identification result by using a digital analysis strategy; and finally, taking the three results as a target landscape gene identification result, so that the efficiency and accuracy of landscape gene text identification and extraction are improved, and the application of the landscape gene is facilitated.

The embodiment of the invention discloses a gene classification method and device based on gene expression data and electronic equipment, and belongs to the technical field of bioinformatics and the field of data processing. The method comprises the following steps: determining expression values of genes in different cell clusters to obtain an expression value data set corresponding to the genes; determining a target expression value threshold value of the gene according to the expression value distribution condition in the expression value data set; and classifying the genes in the cell cluster based on the target expression value threshold. According to the embodiment of the invention, the distribution rule of the gene expression values can be utilized, the objectivity and accuracy of biological classification can be improved, meanwhile, fluctuation of the gene expression level caused by the batch effect can be effectively handled, and the consistency of gene classification is ensured.

This invention provides a species gene classification system, method, apparatus, electronic device, and storage medium, belonging to the field of integrated circuit technology. It includes: a preprocessing module for converting a first serial current signal corresponding to a gene of a species to be tested into a hash vector representing that gene; a content-addressable memory (MAP) for performing vector-matrix multiplication on the hash vectors to obtain multiple first cumulative currents; the MAP is an RRAM array structure, and its conductance is positively correlated with a reference hash vector, where each reference hash vector represents a gene from a species gene pool; and a first comparison module for comparing the various first cumulative currents to determine the species to which the gene belongs. This invention achieves species gene classification through in-memory computation, reducing data movement and eliminating the need for complex algorithms to correct noise, thereby accelerating system operation, reducing sequencing time, and lowering system energy consumption.

The invention relates to the technical field of biology, and particularly discloses a typing and prognosis prediction model construction method based on HR < + > / HER2-LOW breast cancer gene classification. According to the application, clinical pathological characteristics and biological characteristics of the HR < + > / HER2-low breast cancer are comprehensively analyzed, an internal molecular typing system is established, specific molecular treatment targets of all subtypes are identified, a combined treatment scheme is optimized, and'classified treatment 'is developed to accurately treat the HR < + > / HER2-low breast cancer.