41 results about "Biomarker discovery" patented technology

The invention discloses a protein mass spectrum coding method based on self-supervised learning, and the method comprises the steps: guaranteeing the consistency of model input through the preprocessing steps of data enhancement, normalization, sequence length standardization and the like; by constructing a self-supervised learning task and utilizing a multi-head attention mechanism of a Transform architecture, efficient extraction of global and local features of mass spectrum data is realized; through an autoregressive encoder training framework, the model can automatically learn the internal structure of mass spectrum data without a large amount of labeled data, and high-dimensional coding representation with robustness and generalization ability is generated. According to the method, the workload of manual feature design is remarkably reduced, and the generated coding representation can be applied to downstream tasks such as mass spectrum data quality evaluation, protein identification and quantification, post-translational modification identification and biomarker discovery, and has wide applicability and high efficiency; and an innovative intelligent solution is provided for proteomics research.

A system and method are provided for identifying a multimodal biomarker of a prognostic prediction, using a deep learning framework trained to analyze different modality data, including radiomic image data, pathology image data, and molecular image data to obtain unimodal embedding predictions from those modality data and generate multimodal embedding predictions, through application of a loss minimization and attention-based fusion processes.

The present disclosure relates generally to biomarker discovery and patient stratification, and more specifically to machine learning techniques for discovering relevant biomarkers using data collected as part of the standard-of-care (SoC), which can be used to identify a relevant patient population for a therapeutic with a known mechanism of action (MoA). An exemplary method for predicting activity of a molecular analyte of a patient comprises: training a first module of a machine learning model based on a plurality of medical images of a first cohort; training a second module of the machine learning model based on one or more molecular analyte data sets obtained from a second cohort; receiving a medical image from the patient; and predicting, using the trained first and second modules of the machine learning model, the activity of the molecular analyte from the medical image of the patient.

The invention discloses a multi-omics data integration and classification method, system and device based on hierarchical attention, and is applied to the field of precise medical big data analysis. The method comprises the following steps: firstly, generating feature embedding and feature importance scores through a plurality of parallel feature-level attention modules; then, embedding and inputting all the characteristics of the omics into a unified omics-level attention module, and generating omics embedding and omics importance scores; and finally, a classification prediction task is executed based on omics embedding, and a classification result is output for disease classification. The invention completely abandons a traditional dependency graph convolutional network and an integration normal form of variants of the dependency graph convolutional network, and provides a universal hierarchical attention integration architecture. The framework supports classification tasks of any complex diseases, is not limited by omics data types and combination modes, not only is remarkably superior to a traditional integration normal form in classification performance, but also shows a unique negative generalization distance, and proves that the framework has excellent generalization ability. Meanwhile, features and omics importance scores automatically output by the model provide a powerful analysis tool for biomarker discovery and precise diagnosis and treatment of complex diseases.

The present disclosure relates generally to biomarker discovery and patient stratification, and more specifically to machine learning techniques for discovering relevant biomarkers using data collected as part of the standard-of-care (SoC), which can be used to identify a relevant patient population for a therapeutic with a known mechanism of action (MoA). An exemplary method for predicting activity of a molecular analyte of a patient comprises: training a first module of a machine learning model based on a plurality of medical images of a first cohort; training a second module of the machine learning model based on one or more molecular analyte data sets obtained from a second cohort; receiving a medical image from the patient; and predicting, using the trained first and second modules of the machine learning model, the activity of the molecular analyte from the medical image of the patient.

The invention discloses an infantile autism spectrum disorder identification system based on a graph neural network, and the system specifically comprises a data processing module which is used for extracting time sequence information from fMRI data and constructing a brain function connection graph; the model training module is configured with an unsupervised graph auto-encoder GAE and comprises a three-layer graph convolutional network GCN as an encoder and an inner product decoder; the biomarker recognition module is used for executing a replacement test and screening brain regions with significant differences; and the performance evaluation module is used for evaluating the diagnostic performance of the system and the effectiveness of biomarker discovery on the ABIDE data set. The invention provides an ASD identification solution which does not need manual annotation, is high in interpretability and excellent in performance, and has important significance for promoting the development of ASD early diagnosis and personalized treatment.

The present disclosure relates generally to biomarker discovery and patient stratification, and more specifically to machine learning techniques for discovering relevant biomarkers using data collected as part of the standard-of-care (SoC), which can be used to identify a relevant patient population for a therapeutic with a known mechanism of action (MoA). An exemplary method for predicting activity of a molecular analyte of a patient comprises: training a first module of a machine learning model based on a plurality of medical images of a first cohort; training a second module of the machine learning model based on one or more molecular analyte data sets obtained from a second cohort; receiving a medical image from the patient; and predicting, using the trained first and second modules of the machine learning model, the activity of the molecular analyte from the medical image of the patient.

The invention discloses an early pancreatic cancer prediction and risk stratification system based on artificial intelligence, and belongs to the technical field of medical health data analysis and artificial intelligence. The system comprises a multi-omics data adaptive fusion module, a longitudinal health trajectory coding module, a biomarker combination discovery module, a risk prediction and dynamic layering module and a closed-loop feedback optimization module, and a data confidence index generated by the multi-omics fusion module directly affects the attention weight of longitudinal trajectory coding. Longitudinal track coding adopts a bidirectional long-short-term memory network to extract time sequence characteristics, a biomarker discovery module recognizes a synergistic marker combination through a Transform mechanism, a closed-loop feedback module dynamically adjusts parameters of each module according to a prediction result, and clinical verification shows that the prediction accuracy of the system reaches 85%, the I-stage diagnosis rate is improved by 60%, diseases are discovered 8-12 months in advance, and the diagnosis efficiency is improved. The method is obviously superior to the prior art.

The application belongs to the technical field of biotechnology, and particularly relates to a biomarker for differential diagnosis of severe pulmonary tuberculosis. It is found that the XO content in blood of a severe pulmonary tuberculosis patient is significantly lower than that of a non-severe subject, the area (AUC) under the ROC curve in a screening test is 0.898, and the area (AUC) under the ROC curve in a verification test is 0.904. The application is helpful for early assessment of the condition of a pulmonary tuberculosis patient, and provides a basis for early intervention and comprehensive treatment of severe patients.

The invention belongs to the technical field of biomedicine, and particularly relates to a nucleic acid combination product for detecting SIAH2, a diagnostic kit for osteoarthritis and application of the diagnostic kit. According to the application, the models are established by using machine learning to analyze various cell death modes, and the application finds that in the models of the four cell death modes, the ferroptosis model has the highest diagnosis efficiency on the OA and identifies the corresponding biomarkers. It is found that SIAH2 is used as a biomarker to regulate occurrence and development of OA through cartilage cells, and a new diagnosis and treatment strategy is provided for OA.

The present invention relates to an aptamer specifically binding to cell-derived extracellular vesicles and a method for discovering an extracellular vesicle-specific biomarker using same. Specifically, the present invention relates to an aptamer specifically binding to cancer cell-derived extracellular vesicles, and a method for discovering an extracellular vesicle-specific biomarker using same. Unlike prior art requiring immobilization, a method for selecting an aptamer specifically binding to extracellular vesicles according to the present invention can select an aptamer specifically binding to extracellular vesicles that maintain a three-dimensional structure in an actual clinical environment. A biomarker discovery method using the aptamer selected through the selection method can be used to quickly and precisely identify unknown target proteins present in extracellular vesicles. The present invention can be efficiently applied to the fields of novel biomarker discovery, target protein verification for diagnosis, and aptamer-protein interaction-based drug development.

The invention relates to the field of material analytical chemistry and biochemistry, and provides a continuous batch pretreatment device and method for proteomics, glycoproteomics and phosphorylated proteomics samples. According to the continuous batch pretreatment device for the proteomics, glycoproteomics and phosphorylated proteomics samples, under the assistance of the filler, the whole process of protein denaturation, aggregation capture, enzymolysis and complete glycopeptide and phosphorylated peptide enrichment in biological samples can be achieved. Compared with a free solution method conventionally used in proteomics, the sample pretreatment method based on the device has the advantages that the sample treatment time can be effectively shortened, and continuous analysis of mass proteome, glycoprotein group and phosphorylated proteome of trace protein samples can be realized; the method is expected to be widely applied to the aspects of disease occurrence mechanisms, drug action mechanisms, clinical detection, biomarker discovery and the like.

The present application relates to the technical field of in vitro detection, in particular to a matrix material, a preparation method and application thereof. The present application provides a new matrix material for MALDI detection, which comprises MXene material and gold nanowires attached to the surface thereof. By using the high-sensitivity nanomaterial as the matrix material for MALDI detection, metabolites related to diseases (such as X-linked Alport syndrome) can be identified, which opens up a new way for biomarker discovery of diseases and may have important potential in future clinical applications.

The present disclosure relates generally to biomarker discovery and patient stratification, and more particularly to machine learning techniques for discovering related biomarkers using data collected as part of standard care (SoC), which may be used to identify related patient populations with therapeutic agents of known mechanisms of action (MoA). An exemplary method for predicting patient molecular analyte activity includes: training a first module of a machine learning model based on a plurality of medical images of a first queue; a second module to train the machine learning model based on one or more molecular analyte data sets obtained from a second queue; receiving a medical image from the patient; and predicting the activity of the molecular analyte from the medical image of the patient using the trained first and second modules of the machine learning model.

A method and application for deep proteomics analysis of trace samples, which is used for high-depth and high-sensitivity proteomics analysis of the retinal pigment epithelium / choroid complex, and is also applicable to the deep proteomics testing of other trace samples, with wide applicability and flexibility. By systematically optimizing the LC-FAIMS-MS / MS parameters, the present invention overcomes the technical bottlenecks of low coverage of trace tissue proteome and poor quantitative stability, providing a highly sensitive solution for disease mechanism research and biomarker discovery. This method increases the number of protein identifications by 28%, expands the dynamic range to six orders of magnitude, improves the quantitative repeatability, and reduces the proportion of missing values to 6%; it can detect AMD-related low-abundance proteins and 10 cell type markers, helping with disease mechanism research.

The present disclosure provides a method of EEG-based biomarker discovery and commercialization for personalized diagnosis and treatment of brain disorders. Further, the method may include receiving, using a communication device, a brain-wave data from a diagnostic device. Further, the brain-wave data corresponds to a graphical representation of an electrical activity of a brain of an individual. Further, the individual may be receiving at least one therapeutic for at least one brain disorder. Further, the method may include analyzing, using a processing device, the brain-wave data based on an artificial intelligence (AI) model. Further, the method may include generating, using the processing device, at least one output data based on the analyzing. Further, the method may include transmitting, using the communication device, the at least one output data to at least one device.

The present invention provides a SPARK-seq high-throughput platform for the identification and kinetics analysis of nucleic acid aptamers and target proteins thereof. By combining cell screening, CRISPR gene perturbation and single-cell multi-omics sequencing technology, the platform enables systematic identification of the interaction between thousands of nucleic acid aptamers and target proteins thereof in a single experiment; in addition, high-throughput screening of high-stability nucleic acid aptamers having "slow dissociation" characteristics is achieved on the basis of dissociation kinetics. The method provided by the present invention overcomes the limitations of the traditional technology of low throughput, difficulty in identifying targets having relatively low abundance, and inability to perform efficiently screening in natural cell environments, not only achieves large-scale and unbiased discovery of aptamer targets, but also can accurately screen high-stability aptamers having slow dissociation rates, providing more powerful molecular tools and new biomarker discovery pathways for tumor diagnosis, targeted therapy, and accurate medical treatment.

The present invention provides methods and systems using gas-phase separation with mass spectrometry analysis instead of liquid chromatography, thereby enabling faster peptide, proteome, and multi-omic analysis. Also provided are improved methods and software for data independent acquisition. One embodiment referred to as Direct Infusion-Shotgun Proteome Analysis (DI-SPA) used with data-independent acquisition mass spectrometry (DIA-MS), resulted in targeted quantification of over 500 proteins within minutes of MS data collection (˜3.5 proteins / second). Enabling fast, unbiased protein and proteome quantification without liquid chromatography, DI-SPA offers a new approach to boosting throughput critical to drug and biomarker discovery studies that require analysis of thousands of proteomes. This invention is also able to perform complex multi-omic analysis of proteomes, lipidomes, and metabolomes on a single platform.

The invention described herein provides a method to identify predictive response biomarkers (PRBs) for a treatment regimen. The PRBs can be used to identify suitable or unsuitable) patient population for the treatment regimen.

Provided are methods of phospho-tau aggregation-based biomarker discovery, and new utilities for discovered biomarkers in Alzheimer's disease (AD) diagnosis, differentiation, treatment, and identification of the presence of pretangles in a subject. Novel p-tau sites, p-tau198, p-tauS356, p-tau396, and p-tau422, identified through such methods showed comparable or superior characteristics with established p-tau biomarkers, and identified biomarkers were capable of differentiating AD or mild cognitive impairment (MCI) from cognitively normal controls.

The invention relates to a multi-modal feature importance guide network based on deep learning. The multi-modal feature importance guide network comprises a feature reconstruction module, a feature selection module and a classifier, the feature reconstruction module simulates and repairs missing values in the input data through a mask automatic encoder; the input data is multi-modal clinical data of an AML patient; the feature selection module dynamically generates an importance coefficient of a reconstruction variable based on a self-attention mechanism, and weights the reconstruction variable according to the importance coefficient to obtain a weighted variable; and the classifier outputs a lifetime prediction result of the AML patient according to the weighted variable. The beneficial effects of the invention are that the invention provides a data processing method executed by a multi-modal feature importance guide network, and the method utilizes the powerful nonlinear representation capability of deep learning and integrates an interpretability mechanism of biomarker discovery.

The invention discloses a method for detecting metabolite containing phosphate radicals, which is characterized in that a mass spectrum label with stable and characteristic chlorine isotope distribution is introduced through a chlorination reagent derivatization strategy, so that a derivatization product can be subjected to rapid and high-specificity preliminary identification based on primary mass spectrum data. By combining with an automatic data mining program specially developed for this purpose, high-throughput and accurate screening and locking of potential phosphate radical-containing metabolites from massive and complex non-targeted metabonomics original data are realized, and the data analysis efficiency and reliability are greatly improved. The method is successfully applied to analysis of actual biological samples such as cholestasis disease models, efficient identification and accurate quantification of various key metabolites such as nucleotide and sugar phosphoric acid are achieved, disturbance of related metabolic pathways is disclosed, and a powerful innovative analysis tool is provided for disease mechanism research and biomarker discovery.

The invention discloses a feature selection and dimension reduction method and system for multi-omics data fusion, and the system comprises a multi-omics data collection and heterogeneous preprocessing module which is used for carrying out the omics specific preprocessing of genome, transcriptome, proteome and metabolome data; the cross-omics heterogeneous graph neural network fusion module is used for constructing a heterogeneous graph and carrying out information fusion by adopting a heterogeneous graph attention network; the feature selection module based on multi-task sparse learning is used for selecting a key feature subset from the fused features; a non-linear dimension reduction module based on an adversarial variational auto-encoder and used for performing dimension reduction on the key feature subset to a low-dimensional submerged space; and a result visualization and biological interpretation module. According to the method, efficient fusion and dimension reduction of heterogeneous and high-dimensional multi-omics data are realized, the accuracy of feature selection and the interpretability of results are improved, and an effective calculation tool is provided for disease typing and biomarker discovery.

The invention relates to the technical field of tumor molecular typing and biomarkers, and discloses a wide-period SCLC biomarker exploration method based on molecular typing, which comprises the following steps: carrying out molecular typing detection on tumor tissues of a patient, and collecting a peripheral blood sample; performing high-throughput immunochromatogram data acquisition on the peripheral blood sample; carrying out bioinformatics analysis on the high-dimensional data of the immune spectrum, and locking and quantifying a specific CD8 + T cell subset; calculating the frequency dynamic change of the specific CD8 + T cell subpopulation, and analyzing statistical relevance; and a set of biomarker standards and methodology is established. According to the invention, a set of complete biomarker standard and methodology is established by locking the specific CD8 + T cell subpopulation, calculating the dynamic frequency change of the specific CD8 + T cell subpopulation and analyzing the statistical relevance between the dynamic frequency change characteristics and clinical curative effect data of the patient, and can be used for judging or predicting the clinical curative effect of the wide-period SCLC patient.

The present disclosure belongs to the field of artificial intelligence and biomedicine, and relates to a method and device for determining a biomarker, a storage medium, and an electronic device. The method includes: obtaining original omics data corresponding to a target tumor, and inputting the original omics data as training data into a biomarker discovery model to train the biomarker discovery model and determine a first mapping relationship between the target tumor and the biomarker; obtaining test omics data corresponding to the target tumor, and inputting the test omics data into the biomarker discovery model; and determining potential biomarkers in the test omics data based on the first mapping relationship. In the present disclosure, potential biomarkers are determined according to a biomarker discovery model that has a corresponding relationship with the target tumor, which avoids the situation in the prior art where mutant genes with high mutation frequencies are used as biomarkers, and improves the accuracy of the determined biomarkers.

This invention provides a SPARK-seq high-throughput platform for the identification and kinetic analysis of nucleic acid aptamers and their target proteins. This platform innovatively combines cell screening, CRISPR gene perturbation, and single-cell multi-omics sequencing technologies, enabling the systematic identification of thousands of nucleic acid aptamers and their target protein interactions in a single experiment. Simultaneously, it achieves, for the first time, high-throughput screening of highly stable nucleic acid aptamers with "slow dissociation" characteristics based on dissociation kinetics. The method provided by this invention overcomes the limitations of traditional techniques, such as low throughput, difficulty in identifying relatively low-abundance targets, and inability to perform efficient screening in natural cellular environments. It not only achieves large-scale, unbiased discovery of nucleic acid aptamer targets but also accurately screens for highly stable aptamers with slow dissociation rates, providing a more powerful molecular tool and a novel biomarker discovery pathway for tumor diagnosis, targeted therapy, and precision medicine.

The invention discloses discovery of a DM-TB patient early risk early warning biomarker as well as model construction and application of the DM-TB patient early risk early warning biomarker. The invention relates to the field of biomarkers, and particularly discloses a biomarker (IRF1, FPR1, LILRB3, SECTM1, hsa-miR-4726-5p, novel-miR-109, MSTRG.128052.1, MSTRG.4908.1, MSTRG.37670.90 gene, IFN-gamma, IL-6, CXCL10 and CXCL6 protein) and application of a detection substance of the biomarker in preparation of a product for early risk warning of type 2 diabetes mellitus complicated with tuberculosis. The biomarker and the model constructed by using the biomarker are verified to be accurate and reliable, and have good sensitivity and specificity. Through early risk early warning of type 2 diabetes mellitus complicated with tuberculosis, early dry treatment can be carried out on high-risk patients, and optimal disease management is realized.

A method to isolate proteins from biofluids, such as plasma or uterine lavage, for biomarker discovery or for clinical detection. One or more biofluid samples can be provided wherein the one or more biofluids are selected from plasma or uterine lavage samples. Extracellular vesicles (EV) can be selectively and reversibly binded from the biofluid sample to capture and enrich the EVs of the biofluid sample using Extracellular Vesicles total recovery and purification (EVtrap) or other suitable method. At least one compound in the biofluid samples can be detected. The least one compound can consists of plasma or uterine lavage EV proteins and any combination thereof enriched from the plasma or uterine lavage resulting in a greater than 95% recovery yield, greater than 99% purity and with less than 5% coefficient of variation of the at least one EV compound.

The proposed AI drug discovery platform represents a new approach to pharmaceutical research and development, integrating cutting-edge artificial intelligence and machine learning technologies across the entire drug discovery pipeline. This comprehensive system leverages multi-modal data integration, quantum-classical hybrid computing, environmental factor analysis, and digital twin simulations to address the complexities of drug discovery and development. By combining advanced predictive modeling, generative design, and virtual clinical trial capabilities, the platform aims to significantly accelerate the identification and optimization of novel therapeutic compounds while improving safety and efficacy predictions. The system's modular architecture incorporates state-of-the-art techniques in protein design, biomarker discovery, and personalized medicine, enabling a more holistic and precise approach to drug development.

The present invention provides methods and systems using gas-phase separation with mass spectrometry analysis instead of liquid chromatography, thereby enabling faster peptide, proteome, and multi-omic analysis. Also provided are improved methods and software for data independent acquisition. One embodiment referred to as Direct Infusion—Shotgun Proteome Analysis (DI-SPA) used with data-independent acquisition mass spectrometry (DIA-MS), resulted in targeted quantification of over 500 proteins within minutes of MS data collection (˜3.5 proteins / second). Enabling fast, unbiased protein and proteome quantification without liquid chromatography, DI-SPA offers a new approach to boosting throughput critical to drug and biomarker discovery studies that require analysis of thousands of proteomes. This invention is also able to perform complex multi-omic analysis of proteomes, lipidomes, and metabolomes on a single platform.