LASSO–MOGAT: a multi-omics graph attention framework for cancer classification
The application of machine learning (ML) methods to analyze changes in gene expression patterns has recently emerged as a powerful approach in cancer research, enhancing our understanding of the molecular mechanisms underpinning cancer development and progression. Combining gene expressio...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Academia.edu Journals
2024-08-01
|
| Series: | Academia Biology |
| Online Access: | https://www.academia.edu/123385504/LASSO_MOGAT_a_multi_omics_graph_attention_framework_for_cancer_classification |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823859540080197632 |
|---|---|
| author | Fadi Alharbi Aleksandar Vakanski Murtada K. Elbashir Mohanad Mohammed |
| author_facet | Fadi Alharbi Aleksandar Vakanski Murtada K. Elbashir Mohanad Mohammed |
| author_sort | Fadi Alharbi |
| collection | DOAJ |
| description |
The application of machine learning (ML) methods to analyze changes in gene expression patterns has recently emerged as a powerful approach in cancer research, enhancing our understanding of the molecular mechanisms underpinning cancer development and progression. Combining gene expression data with other types of omics data has been reported by numerous works to improve cancer classification outcomes. Despite these advances, effectively integrating high-dimensional multi-omics data and capturing the complex relationships across different biological layers remain challenging. This article introduces Least Absolute Shrinkage and Selection Operator–Multi-omics Gated Attention (LASSO–MOGAT), a novel graph-based deep learning framework that integrates messenger RNA, microRNA, and DNA methylation data to classify 31 cancer types. By utilizing differential expression analysis (DEG) with Linear Models for Microarray (LIMMA) and LASSO regression for feature selection and leveraging graph attention networks (GATs) to incorporate protein–protein interaction (PPI) networks, LASSO–MOGAT effectively captures intricate relationships within multi-omics data. Experimental validation using fivefold cross-validation demonstrates the method’s precision, reliability, and capacity to provide comprehensive insights into cancer molecular mechanisms. The computation of attention coefficients for the edges in the graph, facilitated by the proposed graph attention architecture based on PPIs, proved beneficial for identifying synergies in multi-omics data for cancer classification. |
| format | Article |
| id | doaj-art-634bd204b3014ca7a7ddbd01e20e0679 |
| institution | Kabale University |
| issn | 2837-4010 |
| language | English |
| publishDate | 2024-08-01 |
| publisher | Academia.edu Journals |
| record_format | Article |
| series | Academia Biology |
| spelling | doaj-art-634bd204b3014ca7a7ddbd01e20e06792025-02-11T00:44:19ZengAcademia.edu JournalsAcademia Biology2837-40102024-08-012310.20935/AcadBiol7325LASSO–MOGAT: a multi-omics graph attention framework for cancer classificationFadi Alharbi0Aleksandar Vakanski1Murtada K. Elbashir2Mohanad Mohammed3Department of Computer Science, College of Engineering, University of Idaho, Moscow, ID 83844, USA.Department of Computer Science, College of Engineering, University of Idaho, Moscow, ID 83844, USA.Department of Information Systems, College of Computer and Information Sciences, Jouf University, Sakaka, Aljouf 72441, Saudi Arabia.School of Mathematics, Statistics and Computer Science, University of KwaZulu-Natal, Pietermaritzburg, Scottsville 3209, South Africa. The application of machine learning (ML) methods to analyze changes in gene expression patterns has recently emerged as a powerful approach in cancer research, enhancing our understanding of the molecular mechanisms underpinning cancer development and progression. Combining gene expression data with other types of omics data has been reported by numerous works to improve cancer classification outcomes. Despite these advances, effectively integrating high-dimensional multi-omics data and capturing the complex relationships across different biological layers remain challenging. This article introduces Least Absolute Shrinkage and Selection Operator–Multi-omics Gated Attention (LASSO–MOGAT), a novel graph-based deep learning framework that integrates messenger RNA, microRNA, and DNA methylation data to classify 31 cancer types. By utilizing differential expression analysis (DEG) with Linear Models for Microarray (LIMMA) and LASSO regression for feature selection and leveraging graph attention networks (GATs) to incorporate protein–protein interaction (PPI) networks, LASSO–MOGAT effectively captures intricate relationships within multi-omics data. Experimental validation using fivefold cross-validation demonstrates the method’s precision, reliability, and capacity to provide comprehensive insights into cancer molecular mechanisms. The computation of attention coefficients for the edges in the graph, facilitated by the proposed graph attention architecture based on PPIs, proved beneficial for identifying synergies in multi-omics data for cancer classification.https://www.academia.edu/123385504/LASSO_MOGAT_a_multi_omics_graph_attention_framework_for_cancer_classification |
| spellingShingle | Fadi Alharbi Aleksandar Vakanski Murtada K. Elbashir Mohanad Mohammed LASSO–MOGAT: a multi-omics graph attention framework for cancer classification Academia Biology |
| title | LASSO–MOGAT: a multi-omics graph attention framework for cancer classification |
| title_full | LASSO–MOGAT: a multi-omics graph attention framework for cancer classification |
| title_fullStr | LASSO–MOGAT: a multi-omics graph attention framework for cancer classification |
| title_full_unstemmed | LASSO–MOGAT: a multi-omics graph attention framework for cancer classification |
| title_short | LASSO–MOGAT: a multi-omics graph attention framework for cancer classification |
| title_sort | lasso mogat a multi omics graph attention framework for cancer classification |
| url | https://www.academia.edu/123385504/LASSO_MOGAT_a_multi_omics_graph_attention_framework_for_cancer_classification |
| work_keys_str_mv | AT fadialharbi lassomogatamultiomicsgraphattentionframeworkforcancerclassification AT aleksandarvakanski lassomogatamultiomicsgraphattentionframeworkforcancerclassification AT murtadakelbashir lassomogatamultiomicsgraphattentionframeworkforcancerclassification AT mohanadmohammed lassomogatamultiomicsgraphattentionframeworkforcancerclassification |