Evaluation and development of deep neural networks for RNA 5-Methyluridine classifications using autoBioSeqpy
Post-transcriptionally RNA modifications, also known as the epitranscriptome, play crucial roles in the regulation of gene expression during development. Recently, deep learning (DL) has been employed for RNA modification site prediction and has shown promising results. However, due to the lack of r...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2023-05-01
|
Series: | Frontiers in Microbiology |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fmicb.2023.1175925/full |
_version_ | 1797824140185436160 |
---|---|
author | Lezheng Yu Yonglin Zhang Li Xue Fengjuan Liu Runyu Jing Jiesi Luo Jiesi Luo |
author_facet | Lezheng Yu Yonglin Zhang Li Xue Fengjuan Liu Runyu Jing Jiesi Luo Jiesi Luo |
author_sort | Lezheng Yu |
collection | DOAJ |
description | Post-transcriptionally RNA modifications, also known as the epitranscriptome, play crucial roles in the regulation of gene expression during development. Recently, deep learning (DL) has been employed for RNA modification site prediction and has shown promising results. However, due to the lack of relevant studies, it is unclear which DL architecture is best suited for some pyrimidine modifications, such as 5-methyluridine (m5U). To fill this knowledge gap, we first performed a comparative evaluation of various commonly used DL models for epigenetic studies with the help of autoBioSeqpy. We identified optimal architectural variations for m5U site classification, optimizing the layer depth and neuron width. Second, we used this knowledge to develop Deepm5U, an improved convolutional-recurrent neural network that accurately predicts m5U sites from RNA sequences. We successfully applied Deepm5U to transcriptomewide m5U profiling data across different sequencing technologies and cell types. Third, we showed that the techniques for interpreting deep neural networks, including LayerUMAP and DeepSHAP, can provide important insights into the internal operation and behavior of models. Overall, we offered practical guidance for the development, benchmark, and analysis of deep learning models when designing new algorithms for RNA modifications. |
first_indexed | 2024-03-13T10:34:34Z |
format | Article |
id | doaj.art-f4259417139148eab9743d2d66870551 |
institution | Directory Open Access Journal |
issn | 1664-302X |
language | English |
last_indexed | 2024-03-13T10:34:34Z |
publishDate | 2023-05-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Microbiology |
spelling | doaj.art-f4259417139148eab9743d2d668705512023-05-18T07:21:02ZengFrontiers Media S.A.Frontiers in Microbiology1664-302X2023-05-011410.3389/fmicb.2023.11759251175925Evaluation and development of deep neural networks for RNA 5-Methyluridine classifications using autoBioSeqpyLezheng Yu0Yonglin Zhang1Li Xue2Fengjuan Liu3Runyu Jing4Jiesi Luo5Jiesi Luo6School of Chemistry and Materials Science, Guizhou Education University, Guiyang, ChinaDepartment of Pharmacy, Affiliated Hospital of North Sichuan Medical College, Nanchong, ChinaSchool of Public Health, Southwest Medical University, Luzhou, ChinaSchool of Geography and Resources, Guizhou Education University, Guiyang, ChinaSchool of Cyber Science and Engineering, Sichuan University, Chengdu, ChinaBasic Medical College, Southwest Medical University, Luzhou, ChinaSichuan Key Medical Laboratory of New Drug Discovery and Druggability Evaluation, Luzhou Key Laboratory of Activity Screening and Druggability Evaluation for Chinese Materia Medica, Southwest Medical University, Luzhou, ChinaPost-transcriptionally RNA modifications, also known as the epitranscriptome, play crucial roles in the regulation of gene expression during development. Recently, deep learning (DL) has been employed for RNA modification site prediction and has shown promising results. However, due to the lack of relevant studies, it is unclear which DL architecture is best suited for some pyrimidine modifications, such as 5-methyluridine (m5U). To fill this knowledge gap, we first performed a comparative evaluation of various commonly used DL models for epigenetic studies with the help of autoBioSeqpy. We identified optimal architectural variations for m5U site classification, optimizing the layer depth and neuron width. Second, we used this knowledge to develop Deepm5U, an improved convolutional-recurrent neural network that accurately predicts m5U sites from RNA sequences. We successfully applied Deepm5U to transcriptomewide m5U profiling data across different sequencing technologies and cell types. Third, we showed that the techniques for interpreting deep neural networks, including LayerUMAP and DeepSHAP, can provide important insights into the internal operation and behavior of models. Overall, we offered practical guidance for the development, benchmark, and analysis of deep learning models when designing new algorithms for RNA modifications.https://www.frontiersin.org/articles/10.3389/fmicb.2023.1175925/fullRNA 5-methyluridinedeep learningautoBioSeqpyUMAPSHAP |
spellingShingle | Lezheng Yu Yonglin Zhang Li Xue Fengjuan Liu Runyu Jing Jiesi Luo Jiesi Luo Evaluation and development of deep neural networks for RNA 5-Methyluridine classifications using autoBioSeqpy Frontiers in Microbiology RNA 5-methyluridine deep learning autoBioSeqpy UMAP SHAP |
title | Evaluation and development of deep neural networks for RNA 5-Methyluridine classifications using autoBioSeqpy |
title_full | Evaluation and development of deep neural networks for RNA 5-Methyluridine classifications using autoBioSeqpy |
title_fullStr | Evaluation and development of deep neural networks for RNA 5-Methyluridine classifications using autoBioSeqpy |
title_full_unstemmed | Evaluation and development of deep neural networks for RNA 5-Methyluridine classifications using autoBioSeqpy |
title_short | Evaluation and development of deep neural networks for RNA 5-Methyluridine classifications using autoBioSeqpy |
title_sort | evaluation and development of deep neural networks for rna 5 methyluridine classifications using autobioseqpy |
topic | RNA 5-methyluridine deep learning autoBioSeqpy UMAP SHAP |
url | https://www.frontiersin.org/articles/10.3389/fmicb.2023.1175925/full |
work_keys_str_mv | AT lezhengyu evaluationanddevelopmentofdeepneuralnetworksforrna5methyluridineclassificationsusingautobioseqpy AT yonglinzhang evaluationanddevelopmentofdeepneuralnetworksforrna5methyluridineclassificationsusingautobioseqpy AT lixue evaluationanddevelopmentofdeepneuralnetworksforrna5methyluridineclassificationsusingautobioseqpy AT fengjuanliu evaluationanddevelopmentofdeepneuralnetworksforrna5methyluridineclassificationsusingautobioseqpy AT runyujing evaluationanddevelopmentofdeepneuralnetworksforrna5methyluridineclassificationsusingautobioseqpy AT jiesiluo evaluationanddevelopmentofdeepneuralnetworksforrna5methyluridineclassificationsusingautobioseqpy AT jiesiluo evaluationanddevelopmentofdeepneuralnetworksforrna5methyluridineclassificationsusingautobioseqpy |