RNA-seq expression quantity

The transcriptome analysis workflow, encapsulated by Zhanglab graduate Feng Huadong (details available on GitHub and accessible via Docker Hub), was employed in this study.

For data quality control and preprocessing, FastQC and Fastp were utilized. In the first reference-based pipeline, HISAT2 was selected for alignment, StringTie for transcriptome assembly and quantitative analysis, and the R package DESeq2 for differential gene expression analysis. The second pipeline adopted STAR for alignment, HTSeq-count for quantitative analysis, and edgeR for differential gene expression analysis. In the reference-free pipeline, Bowtie2 was used for alignment, Trinity for transcriptome assembly, RSEM for quantitative analysis, DESeq2 for differential gene expression analysis, and TransDecoder and Trinotate for open reading frame (ORF) prediction and functional annotation of the assembled transcripts.

This study effectively isolated the analysis pipelines, resolving issues related to server environment redundancy, software installation, and configuration. This approach greatly facilitates cross-server transplantation and saves researchers' time.

Our project is publicly accessible on Docker Hub for open use. The encapsulated pipeline images can be easily transferred between servers, enabling even researchers without a bioinformatics background to conduct automated transcriptome data analysis in accordance with the user manual. They can obtain gene and transcript expression levels, differentially expressed genes, as well as volcano plots, heatmaps, and other visualizations of differential genes from raw data. Moreover, the reference-free pipeline includes ORF prediction and functional annotation for the assembled transcripts.