结直肠癌 (CRC) 分子亚型的鉴定对患者具有预后和潜在的诊断价值，但临床上仍然无法获得可靠的亚型。目前结直肠癌的共识分子亚型 (CMS) 分类是基于在基因水平量化的复杂 RNA 表达模式。然而，由于单个样本分类和相关成本的高度不确定性，这些方法的临床应用具有挑战性。选择性剪接（AS）对转录组多样性有很大贡献，但很少用于组织类型分类。在这里，我们提出了一个基于 AS 的 CRC 亚型框架，对差异外显子使用敏感，可适用于临床应用。无监督聚类用于测量不同类别的 AS 和 CMS 之间的关联强度。为了构建分类器，CMS 标签的基本事实源自基因水平量化的表达数据。特征选择是通过引导和 L1 惩罚估计来实现的。所得的特征空间用于构建适用于单个和多个样本的子类型预测框架。对来自两个独立来源（Indivumed，n=129；TCGA，n=99）的看不见的 CRC 评估模型的性能。我们基于 29 个外显子跳跃 (ES) 事件开发了结直肠癌亚型标识符 (CRCi)，该标识符准确地与基于基因表达的分类器相比，它对看不见的肿瘤进行分类，并能够更精确地区分具有不同生物学和预后特征的亚型。在这里，我们证明少量的 ES 事件可以以合适的方式使用个体患者样本可靠地对结直肠癌亚型进行分类临床应用。版权所有 © 2024 AGA Institute。由爱思唯尔公司出版。保留所有权利。

The identification of colorectal cancer (CRC) molecular subtypes has prognostic and potentially diagnostic value for patients, yet reliable subtyping remains unavailable in the clinic. The current consensus molecular subtype (CMS) classification in colorectal cancers is based on complex RNA expression patterns quantified at gene level. The clinical application of these methods, however, is challenging due to high uncertainty of single sample classification and associated costs. Alternative splicing (AS), which strongly contributes to transcriptome diversity, has rarely been utilized for tissue type classification. Here, we present an AS-based CRC subtyping framework sensitive to differential exon usage that can be adapted for clinical application.Unsupervised clustering was used to measure the strength of association between different categories of AS and CMS. To build a classifier, the ground-truth for CMS labels was derived from expression data quantified at gene-level. Feature selection was achieved through bootstrapping and L1-penalized estimation. The resulting feature space was used to construct a subtype prediction framework applicable to single and multiple samples. The performance of the models was evaluated on unseen CRCs from two independent sources (Indivumed, n=129; TCGA, n=99).We developed a colorectal cancer subtype identifier (CRCi) based on 29 exon-skipping (ES) events that accurately classifies unseen tumors and enables more precise differentiation of subtypes characterized by distinct biological and prognostic features as compared to classifiers based on gene expression.Here we demonstrate that a small number of ES events can reliably classify colorectal cancer subtypes using individual patient specimen in a manner suitable to clinical application.Copyright © 2024 AGA Institute. Published by Elsevier Inc. All rights reserved.