Overview
object name: Multicenter analysis of genomic and metabolic data of neonatal genetic diseases.
goal of study:(1) Gene sequencing data (138 genes related to 133 common genetic diseases) and tandem mass spectrometry metabolomics data (11 amino acids and 28 acylcarnitines) of about 40,000 newborns from the South China Neonatal Genetic Screening Alliance participating units were collected and collated to complete the database construction of genes and mass spectrometry.
(2) Explore the use of genome and metabolome big data and machine learning algorithms such as Random forest, Support Vector Machine, Elastic net, Multilayer Perceptron to construct prediction models for common genetic diseases, and strive to achieve accurate diagnosis and prediction of common genetic diseases using simple tandem mass spectrometry metabolome data, and expand the application range of tandem mass spectrometry technology for disease detection.
research design:retrospective observational study Research period:September 2022 to December 2025 Participating units:South China Neonatal genetic screening Alliance (including cooperation units of 123 hospitals) research object:Gene screening data of 40,000 newborns ( 138 genes related to 133 common genetic diseases ) and tandem mass spectrometry data ( 11 amino acids and 28 acylcarnitines ).
Inclusion criteria:( 1 ) Newborns who underwent genetic screening and tandem mass spectrometry at the same time. ( 2 ) Age : 0-28 days, gestational age 37-42 weeks.
Excluded criteria:Data that meets any of the following conditions need to be eliminated : ( 1 ) Neonatal data with unclear clinical basic information ; ( 2 ) Lack of traceability core information data ; ( 3 ) The data that the test results cannot be analyzed and interpreted.
data collection:( 1 ) Basic information : gender, age, sample type, subject traceability number / ID number, etc. ( 2 ) Clinical symptoms, biochemical and imaging data of positive samples. ( 3 ) Gene detection results and tandem mass spectrometry results. ( 4 ) Date of test data, instrument model, reagent type, etc.
Description
Research Design: This study is a multi-center cooperative study of the South China Neonatal Genetic Screening Alliance. The principal investigator ( PI ) and project leader of this study are Hao Hu, chief physician of pediatrics of the Sixth Affiliated Hospital of Sun Yat-sen University, who plans to include 123 cooperative units of the South China Neonatal Genetic Screening Alliance. In this study, 40,000 neonatal genetic screening data and MS / MS data were retrospectively analyzed through multi-center cooperation. The collection date was from January 2019 to August 2022.
Through the statistical analysis of neonatal genetic screening data ( 138 genes related to 133 common genetic diseases ), the incidence of common genetic diseases in newborns in China, the carrying rate of pathogenic variation and the high-frequency variation sites of the population were clarified, and the epidemiological characteristics of newborns in China were studied.
Through the statistical analysis of neonatal genetic screening data and MS / MS metabolomics data ( 11 amino acids and 28 acylcarnitines ), the correlation between gene and metabolism will be explored, and the pathogenicity of high-frequency VUS mutation sites will be identified by using protein function artificial intelligence analysis platform and tandem mass spectrometry metabolite data.
The prediction model of common genetic diseases is constructed by using machine learning algorithms such as random forest, support vector machine, elastic network and multi-layer perceptron, so as to realize the accurate diagnosis of common genetic diseases through tandem mass spectrometry metabolomics data, and expand 2-3 kinds of diseases that can be detected by MS / MS technology.
Sample size: This study plans to collect genetic screening data ( 138 genes related to 133 common genetic diseases ) and tandem mass spectrometry metabolomics data ( 11 amino acids and 28 acylcarnitines ) of about 40,000 newborns from January 2019 to August 2022 in 123 cooperative units of the South China Neonatal Genetic Screening Alliance.
Data source: The gene sequencing data and MS / MS metabolic data of 40,000 newborns were from 123 cooperative units of the South China Neonatal Gene Screening Alliance.In this study, the data table established by Microsoft Excel was used. The neonatal gene data and tandem mass spectrometry of the multi-center cooperative units were transmitted through the Excel data table. Effective measures will be taken to strictly record, clean and check the data. Multi-centers ensure the authenticity, accuracy and completeness of the neonatal gene sequencing data and MS / MS metabolic data provided, and all data and test reports can be traced. In addition, the data management, pay attention to the confidentiality of the data, to ensure the privacy of patients and their families.
Informed consent: This study is a retrospective study. Subjects have signed informed consent from parents or guardians when doing neonatal MS / MS metabolic disease screening or genetic screening. The informed consent form clearly states that the test data can be used for scientific research after removing personal privacy information. Therefore, the application for exemption from informed consent.
Benefits of participating in research: There is no direct economic benefit for all the test subjects included in this study, but for the positive children included in this study, free first-generation sequencing verification is provided, and professional genetic counseling and clinical treatment advice are provided to parents by pediatric clinicians with the consent of the parents of the children.
Privacy protection measures: All the data of the subjects during the study period will be entered into the computer for confidential storage and analysis. If necessary, the relevant institutions may review the records to confirm the authenticity, accuracy and integrity of the data. The data obtained from the study may also be published in academic journals, but the names of the subjects will not be published, and the privacy of the subjects will be kept confidential.
All selected populations do not involve special populations, and patient privacy information is strictly protected.
Eligibility
Inclusion Criteria:
- Age 1-28 days
- gestational age 37-42 weeks
Exclusion Criteria:
- Neonatal data with unclear clinical basic information
- Lack of traceability core information data
- The data that the test results cannot be analyzed and interpreted