Integrative pan-cancer and COAD-focused multi-omics analysis of non-histone acetylation substrate-related transcriptional landscapes
TL;DR
Findings indicate that non-histone acetylation substrate-related transcriptional programs in COAD are linked to poor prognosis and spatial immune heterogeneity, while pathomics provides supportive morphology-linked evidence requiring further validation.
Abstract
Non-histone acetylation regulates protein stability, transcription, and immune signaling, yet its substrate-specific roles in colon adenocarcinoma (COAD) remain unclear. We performed an integrative multi-omics analysis to characterize acetylation substrate-related transcriptional signatures and their clinical relevance. Expression patterns and prognostic impacts of 33 acetylation-related enzymes were evaluated pan-cancer, and substrate-related networks were computationally inferred. A 15-gene substrate expression-based acetylation substrate-related imbalance score (AIScore) stratified COAD patients into prognostically distinct groups. Single-cell RNA sequencing and spatial transcriptomics revealed AIScore-high malignant cells preferentially interacting with SPP1-positive macrophages, forming immunosuppressive niches with reduced predicted immunotherapy response. Exploratory deep learning-based pathomics of whole-slide images, including ResNet-50 feature extraction, PCA, clustering, and Grad-CAM visualization, highlighted AIScore-associated morphologic hotspots at tumor–immune interfaces. These findings indicate that non-histone acetylation substrate-related transcriptional programs in COAD are linked to poor prognosis and spatial immune heterogeneity, while pathomics provides supportive morphology-linked evidence requiring further validation.