AUTHOR=Xie Yiluo , Chen Huili , Tian Mei , Wang Ziqang , Wang Luyao , Zhang Jing , Wang Xiaojing , Lian Chaoqun TITLE=Integrating multi-omics and machine learning survival frameworks to build a prognostic model based on immune function and cell death patterns in a lung adenocarcinoma cohort JOURNAL=Frontiers in Immunology VOLUME=Volume 15 - 2024 YEAR=2024 URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2024.1460547 DOI=10.3389/fimmu.2024.1460547 ISSN=1664-3224 ABSTRACT=We used 10 clustering algorithms and multi-omics data to categorize Lung adenocarcinoma (LUAD) patients into three subtypes. patients with the Cluster3 (CS3) subtype had the best prognosis, whereas patients with the Cluster1 (CS1) and Cluster2 (CS2) subtypes had a poorer prognosis. Immune-related programmed cell death risk score (PIGRS), a combination of 15 highimpact genes, showed strong prognostic performance for LUAD patients. PIGRS had very strong predictive efficacy in comparison with our collected models, PIGRS has a very strong prognostic efficacy compared to our collection. In conclusion, we found that proteasome activator subunit 3 (PSME3), which is rare in lung adenocarcinoma, may be a novel prognostic factor in lung adenocarcinoma, and that it may affect apoptosis of lung adenocarcinoma cells through the PI3K/AKT/Bcl-2 signalling pathway. Our findings contribute to precision medicine and inform the development of rational clinical strategies for targeted and immunotherapy.