CN-121975923-A - Method for identifying specific cells and characteristic genes of rheumatoid arthritis based on peripheral blood single cell transcriptome sequencing and gene combination

CN121975923ACN 121975923 ACN121975923 ACN 121975923ACN-121975923-A

Abstract

The invention discloses a method for screening specific cells and characteristic genes of rheumatoid arthritis based on peripheral blood single cell transcriptome sequencing and a gene combination. The method comprises the steps of carrying out single-cell transcriptome sequencing on Peripheral Blood Mononuclear Cells (PBMC) of a rheumatoid arthritis patient, obtaining a disease-related cell population through multidimensional analysis, further obtaining characteristic expression genes of the cell population, comprehensively evaluating single-cell transcriptome sequencing data and whole transcriptome sequencing data of various autoimmune diseases PBMC, screening out specific genes CREB5, GLIPR1 and ISGF6 of the rheumatoid arthritis, constructing a line diagram model by using the obtained gene combination, and verifying that the model has higher accuracy for diagnosing the rheumatoid arthritis. The invention provides a screening and identifying method of specific cells and characteristic genes with higher biological significance, which is favorable for obtaining diagnostic markers or therapeutic targets with higher reliability and provides a new idea for diagnosis and treatment of autoimmune diseases.

Inventors

LONG MIN
Zeng Xinru
YIN XIUJUAN
Yi Yuanxue

Assignees

重庆博艾生物医学研究院(集团)有限公司

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (6)

1. A method for identifying rheumatoid arthritis specific cells and characteristic genes based on peripheral blood single cell transcriptome sequencing, comprising the steps of: s1, obtaining peripheral blood of RA patients and healthy volunteers; S2, separating peripheral blood mononuclear cells to prepare qualified single-cell suspension; s3, sequencing a single cell transcriptome; S4, obtaining peripheral blood mononuclear cell scRNA-seq data of SLE and pSS; S5, peripheral blood mononuclear cell scRNA-seq data integration analysis of RA, SLE and pSS; S6, cell composition distribution analysis discovers cell subgroups with increased or decreased RA specificity, and explores functions and disease related mechanisms thereof; S7, cell communication analysis discovers cell groups related to the specific cell interaction mode of RA, and searches the correlation between the gene expression mode and diseases; s8, screening RA characteristic genes based on scRNA-seq data; S9, obtaining peripheral blood mononuclear cell bulk RNA-seq data of RA, SLE and pSS; S10, comprehensively evaluating the expression modes of candidate gene sets in the peripheral blood mononuclear cell scRNA-seq data and bulk RNA-seq data of RA, SLE and pSS, and further defining specific marker genes; s11, evaluating the distinction degree of candidate genes on RA through ROC; s12, constructing a nomogram model, and verifying the diagnosis capability of the nomogram model on RA.
2. The method for identifying rheumatoid arthritis-specific cells and characteristic genes based on peripheral blood single cell transcriptome sequencing of claim 1, wherein in step S6, four major peripheral blood cell populations of T, B, NK and mononuclear cells are automatically released and then subjected to subpopulation analysis when cell composition distribution analysis is performed using scRNA-seq data, and cell subpopulations are manually annotated according to Top 15-specific genes of each subpopulation.
3. The method for identifying rheumatoid arthritis-specific cells and characteristic genes based on peripheral blood single cell transcriptome sequencing of claim 1, wherein in step S7, before cell communication analysis using scRNA-seq data, the cell population is re-annotated in advance, that is, annotation is made according to the function imparted to the cell population by the Top 15-specific genes of the clustered cell population, the signal generator and the signal receiver are appropriately divided, and cell signal pattern difference analysis is performed, and the RA-specific cell population is confirmed from the difference in cell interactions.
4. The method for identifying rheumatoid arthritis-specific cells and characteristic genes based on peripheral blood single cell transcriptome sequencing of claim 1, wherein in step S8, the RA-characteristic gene screening criteria are that the differential expression of other cell populations is p.adj <0.05& log2FC >1& pct >0.3 and the differential expression of RA group and control group is p.adj <0.05&|log2FC| >1.
5. The method for identifying rheumatoid arthritis-specific cells and characteristic genes based on peripheral blood single cell transcriptome sequencing of claim 1, wherein in step S10, candidate gene expression patterns are evaluated using bulk RNA-seq data, only if differences are statistically significant and trends are consistent with scRNA-seq data, no threshold is set for log2 FC.
6. The method for identifying rheumatoid arthritis-specific cells and characteristic genes based on peripheral blood single cell transcriptome sequencing according to claim 1, wherein the rheumatoid arthritis-specific cell population and characteristic genes obtained by the method are examined for SLE and ps data expression, excluding common features of SLE and ps and RA.

Description

Method for identifying specific cells and characteristic genes of rheumatoid arthritis based on peripheral blood single cell transcriptome sequencing and gene combination Technical Field The invention belongs to the technical field of biological medicines, and particularly relates to a method for identifying specific cells and characteristic genes of rheumatoid arthritis based on peripheral blood single cell transcriptome sequencing and a gene combination. Background Rheumatoid arthritis (Rheumatoid Arthritis, RA) is a complex disease characterized by chronic synovial inflammation, joint destruction and systemic inflammation, with female morbidity significantly higher than male, and the onset population is in a younger trend. The diagnosis of RA at present mainly refers to two RA classification standards of ACR release in 1987 and ACR/EULAR release in 2010, wherein the RA classification standards are mainly diagnosed according to the compliance of clinical symptoms (such as morning stiffness, arthritis of a plurality of joint areas, symmetrical arthritis and the like), the specificity is higher, and the RA classification standards are integrated diagnosis in various aspects of joint involvement, serological indexes (rheumatoid factors (RF), anti-cyclic citrullinated peptide antibodies (ACPA)) and the like, so that the sensitivity is better and early discovery is facilitated. To ensure adequate sensitivity and specificity, definitive diagnosis of RA is currently based on comprehensive assessment of clinical features, physical examination, serological examination and imaging examination. However, autoimmune diseases often have many common clinical, serological and immunological features, even at the genetic level, with similar genetic and epigenetic features, especially rheumatoid arthritis, systemic lupus erythematosus (Systemic Lupus Erythematosus, SLE) and primary Sj, gren's Syndrome, ps. It has been challenging to improve the sensitivity and specificity of RA diagnosis, as well as to improve early diagnosis. T Tuller and Daniel Toro et al propose that Peripheral Blood Mononuclear Cells (PBMC) are a common source of immune cell infiltration of specific target organs in autoimmune diseases, and research on common characteristics and specific characteristics of PBMC of various representative autoimmune diseases in terms of gene expression, protein interaction and the like, thus providing a new idea for diagnosis and understanding of specific autoimmune diseases. However, since the characterization of peripheral blood immune cells of autoimmune diseases is mainly derived from analysis of peripheral blood batch RNA sequencing (bulk RNA-seq), no practical progress has been made at present. In recent years, single cell transcriptome sequencing (scRNA-seq) technology has provided a new perspective for resolving immune features of RA, and scRNA-seq studies of multiple synovial tissues have identified a variety of potential pathogenic cell subsets, such as itga5+ synovial fibroblasts (highly expressed POSTN, COL3 A1), spp1+ macrophages and cxcl13+cd4+ T cells. However, the immune microenvironment of the synovium is too complex and changeable, and the disease course state and individual difference of the patient easily cause the change of immune cell composition, so that a consistent characteristic marker is not obtained at present, and the synovium biopsy is invasive and difficult to repeatedly sample to monitor the disease activity for the formulation and adjustment of a treatment scheme, so that the peripheral blood as an organ immune cell source can noninvasively reflect the advantage of the systemic immune state, thereby having more application value in research. Single-cell sequencing technology has been widely applied to research of various heterogeneous materials, and analysis methods have been standardized and flowered, but for advanced analysis of deeper biological significance, the analysis methods or flowerings still have no unified standard depending on experience or existing findings of researchers, so the reliability of analysis results still needs to be verified through a large number of subsequent experiments. In order to improve the interpretability of the analysis result, and also due to the high cost and low accessibility of single cell sequencing, researchers usually integrate single cell sequencing data with bulk RNA-seq data, for example, by analyzing the scRNA-seq data to confirm the relevant cell group of the disease and obtain the module genes thereof, then analyzing the clinical relevance of the disease by combining the bulk RNA-seq data, or obtaining the relevant genes of the disease by analyzing the bulk RNA-seq data, and then combining the scRNA-seq data to perform cell type localization and potential function analysis of the genes. However, these methods rely mostly on bulk RNA-seq data machine learning or other subjective means to narrow the candidate genes, do not take full advantage