# Landscape analysis of alternative splicing in kidney renal clear cell carcinoma and their clinical significance

## TL;DR

This study explores how alternative splicing in kidney cancer affects prognosis and identifies a new predictive model for patient survival.

## Contribution

The study provides a novel AS-based prognostic model for kidney renal clear cell carcinoma using TCGA data and experimental validation.

## Key findings

- 46,276 AS events from 10,577 genes were identified in KIRC patients.
- 5,864 prognostic-associated AS events were found using Cox regression analysis.
- A LASSO Cox regression model showed excellent prognostic accuracy for KIRC survival prediction.

## Abstract

A growing number of studies reveal that alternative splicing (AS) is associated with tumorigenesis, progression, and metastasis. Systematic analysis of alternative splicing signatures in renal cancer is lacking. In our study, we investigated the AS landscape of kidney renal clear cell carcinoma (KIRC) and identified AS predictive model to improve the prognostic prediction of KIRC. We obtained clinical data and gene expression profiles of KIRC patients from the TCGA database to evaluate AS events. The calculation results for seven types of AS events indicated that 46276 AS events from 10577 genes were identified. Next, we applied Cox regression analysis to identify 5864 prognostic-associated AS events. We used the Metascape database to verify the potential pathways of prognostic-associated AS. Moreover, we constructed KIRC prediction systems with prognostic-associated AS events by the LASSO Cox regression model. AUCs demonstrated that these prediction systems had excellent prognostic accuracy simultaneously. We identified 34 prognostic associated splicing factors (SFs) and constructed homologous regulatory networks. Furthermore, in vitro experiments were performed to validate the favorable effect of SFs FMR1 in KIRC. In conclusion, we overviewed AS events in KIRC and identified AS-based prognostic models to assist the survival prediction of KIRC patients. Our study may provide a novel predictive signature to improve the prognostic prediction of KIRC, which might facilitate KIRC patient counseling and individualized management.