Idiopathic pulmonary fibrosis plasma protein marker and application thereof in preparation of detection reagent or diagnostic tool
A technology for pulmonary fibrosis and plasma protein, applied in the direction of biological testing, material inspection products, etc., can solve the problem that there are no biomarkers for pulmonary fibrosis disease
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0028] Mass spectrometry analysis of plasma samples from normal population and patients with pulmonary fibrosis
[0029] 1. Experimental Materials
[0030] Plasma samples from patients with pulmonary fibrosis, sourced from a hospital, include whole blood collected from 30 normal people and whole blood collected from 30 patients with pulmonary fibrosis, a total of 60 whole blood samples. According to American Thoracic Society (ATS), European Respiratory Society (ERS), Japanese Respiratory Society (JRS), Association of Latin American Thorax (ALAT) The 2018 edition of the clinical practical guidelines for the diagnosis of IPF mainly implements diagnostic measures from two aspects of chest high-resolution CT and lung biopsy histopathological examination: (1) Chest CT: high-resolution CT features of patients with pulmonary fibrosis include honeycomb, stretch Bronchial or bronchiectasis (mainly in ground-glass opacities and fine-grid opacities), and high-resolution CT findings in a...
Embodiment 2
[0037] Machine Learning Screens for Pulmonary Fibrosis Markers
[0038] 1. Experimental Materials
[0039] 30 plasma samples (see Example 1) mass spectrometry data
[0040] 2. Experimental procedure
[0041] (1) Screening of differential proteins, the absolute value of protein fold change (FC) in the control group and the IPF group was greater than 1.5 (|log2FC|>1.5), and the difference was significantly less than 0.05 (P<0.05).
[0042](2) Randomly select no more than 4 proteins from the differential proteins to form a potential optimal marker combination (OBC), the initial weight value of each protein is set to 1, and 1000 OBC candidates are set.
[0043] (3) For each candidate OBC, we randomly divide the training dataset and the test dataset from the original dataset with a ratio of 4:1. The test dataset is only used to evaluate performance, not for training. Using machine learning algorithm penalized logistic regression (PLR), iteratively optimizes the weight values ...
Embodiment 3
[0047] Verifying OBC Prediction Accuracy
[0048] (1) A pulmonary fibrosis plasma mass spectrometry dataset was obtained from a public database, including 19 normal people and 17 pulmonary fibrosis patients.
[0049] (2) Plasma mass spectrometry results such as Figure 5 shown.
[0050] (3) Four proteins as OBC combinations: amino protease inhibitor family G1 (Serpin Family GMember 1, gene name SERPING1), kininogen 1 (Kininogen 1, gene name KNG1), ficolin 3 (Ficolin 3) , gene name FCN3), transthyretin (gene name TTR) expression in the mass spectrometry data such as Image 6 shown.
[0051] (4) The logistic regression model constructed by the above four proteins SERPING1, KNG1, FCN3, and TTR was used to verify the data set, and the confusion matrix and receiver operating characteristic (ROC) curve were obtained, and the ROC was calculated. The area under the line (AUC) value, the results are as follows Figure 7 , Figure 8 shown.
[0052] (5) The AUC of the OBC combinat...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com