Comprehensive Analysis of Splicing Factor and Alternative Splicing Event to Construct Subtype-Specific Prognosis-Predicting Models for Breast Cancer

• Published in: Frontiers in genetics Vol. 12; p. 736423
• Main Authors: Zhang, He, Han, Baoai, Han, Xingxing, Zhu, Yuying, Liu, Hui, Wang, Zhiyong, Cui, Yanfen, Tian, Ran, Gao, Zicong, Tian, Ruinan, Ren, Sixin, Zuo, Xiaoyan, Tian, Jianfei, Zhang, Fei, Niu, Ruifang
• Format: Journal Article
• Language: English
• Published: Frontiers Media S.A 24.09.2021
• Subjects: alternative splicing; breast cancer; Genetics; prognosis; splicing factor; TCGA database
• ISSN: 1664-8021; 1664-8021
• DOI: 10.3389/fgene.2021.736423

Recent evidence suggests that splicing factors (SFs) and alternative splicing (AS) play important roles in cancer progression. We constructed four SF-risk-models using 12 survival-related SFs. In Luminal-A, Luminal-B, Her-2, and Basal-Like BRCA, SF-risk-models for three genes ( PAXBP1 , NKAP , and NCBP2 ), four genes ( RBM15B , PNN , ACIN1 , and SRSF8 ), three genes ( LSM3 , SNRNP200 , and SNU13 ), and three genes ( SRPK3 , PUF60 , and PNN ) were constructed. These models have a promising prognosis-predicting power. The co-expression and protein-protein interaction analysis suggest that the 12 SFs are highly functional-connected. Pathway analysis and gene set enrichment analysis suggests that the functional role of the selected 12 SFs is highly context-dependent among different BRCA subtypes. We further constructed four AS-risk-models with good prognosis predicting ability in four BRCA subtypes by integrating the four SF-risk-models and 21 survival-related AS-events. This study proposed that SFs and ASs were potential multidimensional biomarkers for the diagnosis, prognosis, and treatment of BRCA.