Cargando…
MDMF: Predicting miRNA–Disease Association Based on Matrix Factorization with Disease Similarity Constraint
MicroRNAs (miRNAs) have drawn enormous attention owing to their significant roles in various biological processes, as well as in the pathogenesis of human diseases. Therefore, predicting miRNA–disease associations is a pivotal task for the early diagnosis and better understanding of disease pathogen...
Autor principal: | |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9224864/ https://www.ncbi.nlm.nih.gov/pubmed/35743670 http://dx.doi.org/10.3390/jpm12060885 |
_version_ | 1784733473515241472 |
---|---|
author | Ha, Jihwan |
author_facet | Ha, Jihwan |
author_sort | Ha, Jihwan |
collection | PubMed |
description | MicroRNAs (miRNAs) have drawn enormous attention owing to their significant roles in various biological processes, as well as in the pathogenesis of human diseases. Therefore, predicting miRNA–disease associations is a pivotal task for the early diagnosis and better understanding of disease pathogenesis. To date, numerous computational frameworks have been proposed to identify potential miRNA–disease associations without escalating the costs and time required for clinical experiments. In this regard, I propose a novel computational framework (MDMF) for identifying potential miRNA–disease associations using matrix factorization with a disease similarity constraint. To evaluate the performance of MDMF, I calculated the area under the ROC curve (AUCs) in the framework of global and local leave-one-out cross-validation (LOOCV). In conclusion, MDMF achieved reliable AUC values of 0.9147 and 0.8905 for global and local LOOCV, respectively, which was a significant improvement upon the previous methods. Additionally, case studies were conducted on two major human cancers (breast cancer and lung cancer) to validate the effectiveness of MDMF. Comprehensive experimental results demonstrate that MDMF not only discovers miRNA–disease associations efficiently but also deciphers the underlying roles of miRNAs in the pathogenesis of diseases at a system level. |
format | Online Article Text |
id | pubmed-9224864 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-92248642022-06-24 MDMF: Predicting miRNA–Disease Association Based on Matrix Factorization with Disease Similarity Constraint Ha, Jihwan J Pers Med Article MicroRNAs (miRNAs) have drawn enormous attention owing to their significant roles in various biological processes, as well as in the pathogenesis of human diseases. Therefore, predicting miRNA–disease associations is a pivotal task for the early diagnosis and better understanding of disease pathogenesis. To date, numerous computational frameworks have been proposed to identify potential miRNA–disease associations without escalating the costs and time required for clinical experiments. In this regard, I propose a novel computational framework (MDMF) for identifying potential miRNA–disease associations using matrix factorization with a disease similarity constraint. To evaluate the performance of MDMF, I calculated the area under the ROC curve (AUCs) in the framework of global and local leave-one-out cross-validation (LOOCV). In conclusion, MDMF achieved reliable AUC values of 0.9147 and 0.8905 for global and local LOOCV, respectively, which was a significant improvement upon the previous methods. Additionally, case studies were conducted on two major human cancers (breast cancer and lung cancer) to validate the effectiveness of MDMF. Comprehensive experimental results demonstrate that MDMF not only discovers miRNA–disease associations efficiently but also deciphers the underlying roles of miRNAs in the pathogenesis of diseases at a system level. MDPI 2022-05-27 /pmc/articles/PMC9224864/ /pubmed/35743670 http://dx.doi.org/10.3390/jpm12060885 Text en © 2022 by the author. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Ha, Jihwan MDMF: Predicting miRNA–Disease Association Based on Matrix Factorization with Disease Similarity Constraint |
title | MDMF: Predicting miRNA–Disease Association Based on Matrix Factorization with Disease Similarity Constraint |
title_full | MDMF: Predicting miRNA–Disease Association Based on Matrix Factorization with Disease Similarity Constraint |
title_fullStr | MDMF: Predicting miRNA–Disease Association Based on Matrix Factorization with Disease Similarity Constraint |
title_full_unstemmed | MDMF: Predicting miRNA–Disease Association Based on Matrix Factorization with Disease Similarity Constraint |
title_short | MDMF: Predicting miRNA–Disease Association Based on Matrix Factorization with Disease Similarity Constraint |
title_sort | mdmf: predicting mirna–disease association based on matrix factorization with disease similarity constraint |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9224864/ https://www.ncbi.nlm.nih.gov/pubmed/35743670 http://dx.doi.org/10.3390/jpm12060885 |
work_keys_str_mv | AT hajihwan mdmfpredictingmirnadiseaseassociationbasedonmatrixfactorizationwithdiseasesimilarityconstraint |