Multiomic machine learning on lactylation for molecular typing and prognosis of lung adenocarcinoma
Abstract To integrate machine learning and multiomic data on lactylation-related genes (LRGs) for molecular typing and prognosis prediction in lung adenocarcinoma (LUAD). LRG mRNA and long non-coding RNA transcriptomes, epigenetic methylation data, and somatic mutation data from The Cancer Genome At...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-87419-4 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585852835856384 |
|---|---|
| author | Mengmeng Hua Tao Li |
| author_facet | Mengmeng Hua Tao Li |
| author_sort | Mengmeng Hua |
| collection | DOAJ |
| description | Abstract To integrate machine learning and multiomic data on lactylation-related genes (LRGs) for molecular typing and prognosis prediction in lung adenocarcinoma (LUAD). LRG mRNA and long non-coding RNA transcriptomes, epigenetic methylation data, and somatic mutation data from The Cancer Genome Atlas LUAD cohort were analyzed to identify lactylation cancer subtypes (CSs) using 10 multiomics ensemble clustering techniques. The findings were then validated using the GSE31210 and GSE13213 LUAD cohorts. A prognosis model for LUAD was developed using the identified hub LRGs to divide patients into high- and low-risk groups. The effectiveness of this model was validated. We identified two lactylation CSs, which were validated in the GSE31210 and GSE13213 LUAD cohorts. Nine hub LRGs, namely HNRNPC, PPIA, BZW1, GAPDH, H2AFZ, RAN, KIF2C, RACGAP1, and WBP11, were used to construct the prognosis model. In the subsequent prognosis validation, the high-risk group included more patients with stage T3 + 4, N1 + 2 + 3, M1, and III + IV cancer; higher recurrence/metastasis rates; and lower 1, 3, and 5 year overall survival rates. In the oncogenic pathway analysis, most of the oncogenic mutations were detected in the high-risk group. The tumor microenvironment analysis illustrated that immune activity was notably elevated in low-risk patients, indicating they might more strongly respond to immunotherapy than high-risk patients. Further, oncoPredict analysis revealed that low-risk patients have increased sensitivity to chemotherapeutics. Overall, we developed a model that combines multiomic analysis and machine learning for LUAD prognosis. Our findings represent a valuable reference for further understanding the important function of lactylation modification pathways in LUAD progression. |
| format | Article |
| id | doaj-art-806cd07b834b4327a00fd010ed2311ce |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-806cd07b834b4327a00fd010ed2311ce2025-01-26T12:25:23ZengNature PortfolioScientific Reports2045-23222025-01-0115111710.1038/s41598-025-87419-4Multiomic machine learning on lactylation for molecular typing and prognosis of lung adenocarcinomaMengmeng Hua0Tao Li1Department of Oral and Maxillofacial Surgery, Qilu Hospital of Shandong UniversityDepartment of Respiratory Diseases, Qilu Hospital of Shandong UniversityAbstract To integrate machine learning and multiomic data on lactylation-related genes (LRGs) for molecular typing and prognosis prediction in lung adenocarcinoma (LUAD). LRG mRNA and long non-coding RNA transcriptomes, epigenetic methylation data, and somatic mutation data from The Cancer Genome Atlas LUAD cohort were analyzed to identify lactylation cancer subtypes (CSs) using 10 multiomics ensemble clustering techniques. The findings were then validated using the GSE31210 and GSE13213 LUAD cohorts. A prognosis model for LUAD was developed using the identified hub LRGs to divide patients into high- and low-risk groups. The effectiveness of this model was validated. We identified two lactylation CSs, which were validated in the GSE31210 and GSE13213 LUAD cohorts. Nine hub LRGs, namely HNRNPC, PPIA, BZW1, GAPDH, H2AFZ, RAN, KIF2C, RACGAP1, and WBP11, were used to construct the prognosis model. In the subsequent prognosis validation, the high-risk group included more patients with stage T3 + 4, N1 + 2 + 3, M1, and III + IV cancer; higher recurrence/metastasis rates; and lower 1, 3, and 5 year overall survival rates. In the oncogenic pathway analysis, most of the oncogenic mutations were detected in the high-risk group. The tumor microenvironment analysis illustrated that immune activity was notably elevated in low-risk patients, indicating they might more strongly respond to immunotherapy than high-risk patients. Further, oncoPredict analysis revealed that low-risk patients have increased sensitivity to chemotherapeutics. Overall, we developed a model that combines multiomic analysis and machine learning for LUAD prognosis. Our findings represent a valuable reference for further understanding the important function of lactylation modification pathways in LUAD progression.https://doi.org/10.1038/s41598-025-87419-4Lung adenocarcinomaLactylationMachine learningMultiomicsPrognosis |
| spellingShingle | Mengmeng Hua Tao Li Multiomic machine learning on lactylation for molecular typing and prognosis of lung adenocarcinoma Scientific Reports Lung adenocarcinoma Lactylation Machine learning Multiomics Prognosis |
| title | Multiomic machine learning on lactylation for molecular typing and prognosis of lung adenocarcinoma |
| title_full | Multiomic machine learning on lactylation for molecular typing and prognosis of lung adenocarcinoma |
| title_fullStr | Multiomic machine learning on lactylation for molecular typing and prognosis of lung adenocarcinoma |
| title_full_unstemmed | Multiomic machine learning on lactylation for molecular typing and prognosis of lung adenocarcinoma |
| title_short | Multiomic machine learning on lactylation for molecular typing and prognosis of lung adenocarcinoma |
| title_sort | multiomic machine learning on lactylation for molecular typing and prognosis of lung adenocarcinoma |
| topic | Lung adenocarcinoma Lactylation Machine learning Multiomics Prognosis |
| url | https://doi.org/10.1038/s41598-025-87419-4 |
| work_keys_str_mv | AT mengmenghua multiomicmachinelearningonlactylationformoleculartypingandprognosisoflungadenocarcinoma AT taoli multiomicmachinelearningonlactylationformoleculartypingandprognosisoflungadenocarcinoma |