Skip to main content
Fig. 1 | BMC Medical Genomics

Fig. 1

From: Gene biomarker prediction in glioma by integrating scRNA-seq data and gene regulatory network

Fig. 1

Overview of the computational framework. This framework consists of four steps: (1) Preprocessing of single-cell gene expression data. Split the original gene expression data according to the sample ID and then preprocess the data through data cleaning, feature selection and standardization. (2) Identification of tumor consensus genes. For each single-sample single-cell gene expression data, explore the gene expression patterns of all malignant cells. Then, tumor consensus genes were identified based on the overlapping degree of the differential genes among samples. (3) Identification of tumor cell types. Build a specific regulatory network based on tumor consensus genes and FFLs. Then, the single-cell specific regulatory expression matrix was constructed, and the cell types of glioma is obtained through a hybrid clustering method. (4) Identification of candidate tumor gene biomarkers. The tumor gene biomarkers were identified according to the degree of correlation between the candidate genes and the tumor eigenvector

Back to article page