Accurate prediction of stomach adenocarcinomas of poorest and best prognosis with a combination of gene expression and clinical signatures
Lingyu Qiu , Huayu Kang , Jielin Yang , Yidong Zheng , Aiyue Chen , Chunlin Chen , Xinlong Wang , Qiongfang Fang , Wei-Guo Zhu , Ou Sha , Yejun Wang
Genome Instability & Disease ›› 2022, Vol. 3 ›› Issue (5) : 227 -237.
Accurate prediction of stomach adenocarcinomas of poorest and best prognosis with a combination of gene expression and clinical signatures
The genetic heterogeneity hampers the identification of biomarkers for Stomach adenocarcinoma (STAD). Here, we proposed a new pipeline to screen the cross-cohort reliable prognostic RNA signatures, and to develop practically applicable models predicting the gastric cancers with shortest and longest overall survival. A strategy of bi-end stratification combined with supervised comparison and LASSO regression was proposed, which could largely improve the statistical power in identification of prognostic gene signatures. By the strategy, we identified 12 signature genes whose expression was associated with the poorest prognosis, and other 12 genes associated with the best prognosis of STAD. Stably expressed genes (SEGs) among different cell types of normal and diseased stomachs were identified using single-cell RNA-seq data, expression levels of the signature genes were normalized with these SEGs, and multi-gene Cox models were built with the normalized expression of the signature genes with the training cohort. The models could well predict the STAD cases of poorest and best prognosis. Combination of stage information further improved the prediction performance, with the accuracy reaching 0.69–0.84 for poorest prognosis and 0.75–0.81 for best prognosis prediction in independent STAD cohorts. The study identified cross-cohort stable STAD prognosis-associated genes, and developed two practically applicable multi-gene models identifying STAD cases of poorest and best prognosis effectively.
High-level University Construction Project of Shenzhen University
Youth Innovation Team of Shenzhen University(no. 406/0000080805)
Natural Science Fund of Shenzhen(JCYJ20190808165205582)
National Natural ScieNational Natural Science Foundation of Chinance Foundation of China(NFSC81773939)
the Cultivation of Guangdong College Students’ Scientific and Technological Innovation, Climbing Program(pdjh2021b0431)
Guangdong Provincial Undergraduate Training Program of China for Innovation and Entrepreneurship(S202010590062)
/
| 〈 |
|
〉 |
PDF
