A method for predicting pregnancy-related diseases based on high-throughput sequencing of peripheral blood cell-free DNA

A prediction method and high-throughput technology, applied in biochemical equipment and methods, genomics, sequence analysis, etc., can solve problems such as low positive rate, high extraction cost, and easy degradation of samples

CN110580934BActive Publication Date: 2022-05-10SOUTHERN MEDICAL UNIVERSITY
7 Cites 1 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2022-05-10

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a pregnancy-related disease prediction model based on high-throughput sequencing of free peripheral blood DNA. The study of the present invention found that the distribution of free DNA in the gene transcription start site region of pregnant women can reflect the physiological state of pregnant women and fetuses. There are significant differences between healthy pregnant women and can effectively predict the onset of pregnancy-related diseases. Based on this, the present invention constructs a pregnancy-related disease screening and prediction model based on peripheral blood cell-free DNA detection, which can predict the onset of pregnancy-related diseases before the clinical symptoms of pregnancy-related diseases appear, and is a non-invasive, economical, convenient, The method for accurate early prediction of pregnancy-related diseases has a good application prospect in the development of predictive screening products for pregnancy-related diseases.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of disease detection products. More specifically, it relates to a method for predicting pregnancy-related diseases based on high-throughput sequencing of peripheral blood cell-free DNA. Background technique

[0002] Diseases related to pregnancy can affect a series of physiological processes of pregnancy, harm the health of pregnant women and fetuses, and even endanger the lives of pregnant women and fetuses in severe cases. Some pregnancy-related diseases have a high incidence frequency in the population, including preeclampsia (incidence 3-8%), gestational diabetes (5-10%), fetal growth restriction (5-10%), macrosomia children (7 ~ 12%) and so on.

[0003] At present, according to the clear diagnostic indicators of these pregnancy-related diseases, the diagnosis can be made when the fetus is approaching maturity or has been delivered. The diagnosis time of fetal growth restriction was after the birth of...

Examples

Embodiment 1

[0058] 1. Experimental method

[0059] (1) Collect free peripheral blood DNA, peripheral blood leukocyte DNA and RNA from the same sample. Micrococcal Nuclease (MNase) was used to treat peripheral blood leukocyte DNA to obtain the DNA sequence bound to nucleosomes. Perform high-throughput sequencing of free peripheral blood DNA, peripheral blood leukocyte RNA, and peripheral blood leukocyte nucleosome-bound DNA, compare the sequencing results with the genome sequence map, and calculate the transcription start sites from all genes in the same sample The number of regional DNA fragments, the number of nucleosome-bound DNA fragments, and the number of RNA fragments of the gene to be tested.

[0060] This step is specifically to determine where the DNA or RNA fragment comes from on the chromosome: after DNA double-end sequencing (alternatively, single-end sequencing can also be used), the sequences at both ends can be compared with the human genome standard sequence 37.1( http: / ...

Embodiment 2

[0075] Based on the above research results of genes, the present invention uses machine learning algorithms to provide a relatively non-invasive and economical and convenient early prediction method for pregnancy-related diseases through the optimal combination of different differential genes, which can be used for pre-eclampsia, gestational diabetes, fetal growth disorders Early detection and screening of pregnancy-related diseases such as limited and macrosomia.

[0076] Specifically, a pregnancy-related disease prediction model based on high-throughput sequencing of peripheral blood cell-free DNA includes three modules:

[0077] (1) High-throughput sequencing and analysis module for free peripheral blood DNA of samples to be tested:

[0078] Perform high-throughput sequencing of free peripheral blood DNA of the sample to be tested, compare the sequencing results with the genome sequence map, and calculate the number of DNA fragments from the transcription start site region of...

Embodiment 3

[0101] Embodiment 3 detection example

[0102] 1. Example of gestational diabetes detection

[0103] Operate according to the method of Example 2. In step 2, the total number of aligned sequences of the sample is counted. In this example, sample 1 and sample 2 are 69479 and 57037, respectively. Calculate the number of DNA fragments in the region of the transcription start site of the gene to be tested in the same sample, and use Formula 1 to correct the abundance of DNA fragments. Table 1 is an example of the calculation of the abundance of DNA fragments in the region of the transcription start site of the gene to be tested in two samples:

[0104] Table 1

[0105]

[0106] Use Equation 2 to calculate the risk of developing gestational diabetes mellitus. An example calculation is as follows:

[0107] Sample 1 (pre-onset sample with confirmed gestational diabetes):

[0108] logit(Y)=0.957+0.565×CC2D2B–1.060×NAT10–1.070×SIPA1–0.620×ZNF565–0.805×ZNF552–0.367×WDR35+0.559×M...