DiCleave: a deep learning model for predicting human Dicer cleavage sites
Abstract Background MicroRNAs (miRNAs) are a class of non-coding RNAs that play a pivotal role as gene expression regulators. These miRNAs are typically approximately 20 to 25 nucleotides long. The maturation of miRNAs requires Dicer cleavage at specific sites within the precursor miRNAs (pre-miRNAs...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
BMC
2024-01-01
|
Series: | BMC Bioinformatics |
Subjects: | |
Online Access: | https://doi.org/10.1186/s12859-024-05638-4 |
_version_ | 1797355470595293184 |
---|---|
author | Lixuan Mu Jiangning Song Tatsuya Akutsu Tomoya Mori |
author_facet | Lixuan Mu Jiangning Song Tatsuya Akutsu Tomoya Mori |
author_sort | Lixuan Mu |
collection | DOAJ |
description | Abstract Background MicroRNAs (miRNAs) are a class of non-coding RNAs that play a pivotal role as gene expression regulators. These miRNAs are typically approximately 20 to 25 nucleotides long. The maturation of miRNAs requires Dicer cleavage at specific sites within the precursor miRNAs (pre-miRNAs). Recent advances in machine learning-based approaches for cleavage site prediction, such as PHDcleav and LBSizeCleav, have been reported. ReCGBM, a gradient boosting-based model, demonstrates superior performance compared with existing methods. Nonetheless, ReCGBM operates solely as a binary classifier despite the presence of two cleavage sites in a typical pre-miRNA. Previous approaches have focused on utilizing only a fraction of the structural information in pre-miRNAs, often overlooking comprehensive secondary structure information. There is a compelling need for the development of a novel model to address these limitations. Results In this study, we developed a deep learning model for predicting the presence of a Dicer cleavage site within a pre-miRNA segment. This model was enhanced by an autoencoder that learned the secondary structure embeddings of pre-miRNA. Benchmarking experiments demonstrated that the performance of our model was comparable to that of ReCGBM in the binary classification tasks. In addition, our model excelled in multi-class classification tasks, making it a more versatile and practical solution than ReCGBM. Conclusions Our proposed model exhibited superior performance compared with the current state-of-the-art model, underscoring the effectiveness of a deep learning approach in predicting Dicer cleavage sites. Furthermore, our model could be trained using only sequence and secondary structure information. Its capacity to accommodate multi-class classification tasks has enhanced the practical utility of our model. |
first_indexed | 2024-03-08T14:12:36Z |
format | Article |
id | doaj.art-da523028fc604264a044ac6d7099ff0c |
institution | Directory Open Access Journal |
issn | 1471-2105 |
language | English |
last_indexed | 2024-03-08T14:12:36Z |
publishDate | 2024-01-01 |
publisher | BMC |
record_format | Article |
series | BMC Bioinformatics |
spelling | doaj.art-da523028fc604264a044ac6d7099ff0c2024-01-14T12:38:48ZengBMCBMC Bioinformatics1471-21052024-01-0125111510.1186/s12859-024-05638-4DiCleave: a deep learning model for predicting human Dicer cleavage sitesLixuan Mu0Jiangning Song1Tatsuya Akutsu2Tomoya Mori3Bioinformatics Center, Institute for Chemical Research, Kyoto UniversityMonash Biomedicine Discovery Institute and Department of Biochemistry and Molecular Biology, Monash UniversityBioinformatics Center, Institute for Chemical Research, Kyoto UniversityBioinformatics Center, Institute for Chemical Research, Kyoto UniversityAbstract Background MicroRNAs (miRNAs) are a class of non-coding RNAs that play a pivotal role as gene expression regulators. These miRNAs are typically approximately 20 to 25 nucleotides long. The maturation of miRNAs requires Dicer cleavage at specific sites within the precursor miRNAs (pre-miRNAs). Recent advances in machine learning-based approaches for cleavage site prediction, such as PHDcleav and LBSizeCleav, have been reported. ReCGBM, a gradient boosting-based model, demonstrates superior performance compared with existing methods. Nonetheless, ReCGBM operates solely as a binary classifier despite the presence of two cleavage sites in a typical pre-miRNA. Previous approaches have focused on utilizing only a fraction of the structural information in pre-miRNAs, often overlooking comprehensive secondary structure information. There is a compelling need for the development of a novel model to address these limitations. Results In this study, we developed a deep learning model for predicting the presence of a Dicer cleavage site within a pre-miRNA segment. This model was enhanced by an autoencoder that learned the secondary structure embeddings of pre-miRNA. Benchmarking experiments demonstrated that the performance of our model was comparable to that of ReCGBM in the binary classification tasks. In addition, our model excelled in multi-class classification tasks, making it a more versatile and practical solution than ReCGBM. Conclusions Our proposed model exhibited superior performance compared with the current state-of-the-art model, underscoring the effectiveness of a deep learning approach in predicting Dicer cleavage sites. Furthermore, our model could be trained using only sequence and secondary structure information. Its capacity to accommodate multi-class classification tasks has enhanced the practical utility of our model.https://doi.org/10.1186/s12859-024-05638-4miRNADicer cleavage site predictionDeep learningAutoencoder |
spellingShingle | Lixuan Mu Jiangning Song Tatsuya Akutsu Tomoya Mori DiCleave: a deep learning model for predicting human Dicer cleavage sites BMC Bioinformatics miRNA Dicer cleavage site prediction Deep learning Autoencoder |
title | DiCleave: a deep learning model for predicting human Dicer cleavage sites |
title_full | DiCleave: a deep learning model for predicting human Dicer cleavage sites |
title_fullStr | DiCleave: a deep learning model for predicting human Dicer cleavage sites |
title_full_unstemmed | DiCleave: a deep learning model for predicting human Dicer cleavage sites |
title_short | DiCleave: a deep learning model for predicting human Dicer cleavage sites |
title_sort | dicleave a deep learning model for predicting human dicer cleavage sites |
topic | miRNA Dicer cleavage site prediction Deep learning Autoencoder |
url | https://doi.org/10.1186/s12859-024-05638-4 |
work_keys_str_mv | AT lixuanmu dicleaveadeeplearningmodelforpredictinghumandicercleavagesites AT jiangningsong dicleaveadeeplearningmodelforpredictinghumandicercleavagesites AT tatsuyaakutsu dicleaveadeeplearningmodelforpredictinghumandicercleavagesites AT tomoyamori dicleaveadeeplearningmodelforpredictinghumandicercleavagesites |