DeepEA:a Galaxy-based Framework For Exploring Epitranscriptome Sequencing Data

Eukaryotic RNA can carry more than 160 different types of chemical modifications(CMR),which is a newly discovered layer of gene regulation with roles in stress responses,RNA folding,and mRNA translation,among other functions.Recently,with the development of several HTS(High Throughput Sequencing)technologies,the large-scale sequencing of CMR in the whole transcriptome generates a large number of epitranscriptome data,which provides abundant data resources for analyzing the post-transcriptional regulation mechanism of CMR.However,despite the advances in epitranscriptome analysis,several challenges still exist,such as the lack of standardized and comprehensive bioinformatics analysis process of epitranscriptome data,lack of bioinformatics software to achieve epitranscriptome data denoising,CMR precisely localizes as well as transcriptome-wide prediction.Therefore,the development of analytical method and software for epitranscriptome would contribute greatly to decipher the mechanism of CMR.The main work of this study is summarized as follows:Based on machine learning technology,the study constructs a bioinformatics model for denoising and accurate CMR prediction of epitranscriptome data.Further,a bioinformatics pipeline for the integrated analysis of epitranscriptome data is established through the integration and development of CMR analysis tools,including sequencing data preprocessing,quality control,CMR identification and analysis,CMR annotation,CMR prediction,data visualization and so forth functions.We developed DeepEA,an integrated analysis platform for epitranscriptome data.DeepEA integrates the bioinformatics software and dependencies required in the epitranscriptome data analysis process to achieve a functional set from sequencing data preprocessing to CMR analysis visualization via the Galaxy framework.DeepEA is packaged with Docker technology and can be easily deployed on high-performance servers,that provides a new tool for large-scale analysis and function mining of epitranscriptome data.