An Analysis Pipeline for Identification of RNA Modification, Alternative Splicing and Polyadenylation Using Third Generation Sequencing

Background

Transcriptome provides an in-depth perspective for understanding the regulation in gene expression and growth development. With the rapid development of the third-generation sequencing technology, the data based on long-read sequencing presented great advantage in revealing post-transcriptional regulation (Smith et al., 2019). Post-transcription regulation has been known as a complex biological process, playing an important part in influencing the splicing, export, localization and translation of RNA (Keene, 2007). For example, alternative polyadenylation (APA) plays a vital role in maintaining the structure and stability of RNA (Di Giammartino et al., 2011, Tian and Manley, 2017). Alternative splicing (AS) is implicated in the regulation of rhizome-associated development (Wang et al., 2017) and hormone response (Zhang et al., 2018). A variety of AS events could be found in key genes related to biosynthesis pathways (Chen et al., 2020). PacBio Iso-Seq can capture the full-length mRNA in isoform-wide to analyze the post-transcriptional processing events such as AS and APA (Rhoads and Au, 2015). Accuracy and reliability tools are important to identify post-transcriptional events. AS based on short-read platform can be identified by several excellent software, such as rMATS (Shen et al., 2014) and SUPPA2 (Trincado et al., 2018). However short-read platforms show disadvantage in detecting full-length isoforms due to the limitation of transcriptome assembly (Rhoads and Au, 2015). TAPIS (Abdel-Ghany et al., 2016) and FLAIR (Tang et al., 2020) have been developed to identify AS events based on long-read sequencing. However, TAPIS does not provide quantitative analysis of post-transcriptional events (Abdel-Ghany et al., 2016). FLAIR can detect isoforms, but lacks APA detecting module (Tang et al., 2020). Hence, we provide a comprehensive and user-friendly pipeline (PRAPI) for the identification and visualization of AS and APA (Gao et al., 2018). Here we provide detail procedure for data processing, visualization, identification of AS and APA using PRAPI based on Iso-Seq isoforms and short read data.

In addition to PacBio platform, Oxford Nanopore Technologies (ONT) platform applies a unique measurement method to keep the information of RNA molecular passing through a special pore as an electrical signal (Garalde et al., 2018), which can identify RNA modifications including N⁶-methyladenosine (m⁶A) (Linder et al., 2015; Zhang et al., 2019), pseudouridine (Ψ) and 2′-O-methylation (Nm) (Begik et al., 2021) in single-nucleotide-resolution based on native RNA. Moreover, Nanopore DRS is capable of determining RNA secondary structure (Aw et al., 2021), and detecting amino acid from native unfolding protein sequence (Hu et al., 2021). Methods for detecting RNA modification based on ONT data have been reported in several software packages including Epinano (Liu et al., 2019), xPore (Pratanwanich et al., 2021), MINES (Lorenz et al., 2020), NanoDoc (Ueda, 2021), Nanom6A (Gao et al., 2021), Tombo (Stoiber et al., 2017). Epinano, xPore, Nanom6A, and MINES can achieve accuracy in single-base resolution. However, Epinano cannot distinguish m⁶A from other modified bases (such as m¹A) due to depending on base-calling errors (Liu et al., 2019). Moreover, xPore requires m⁶A writer knockout lines as comparison and may be biased due to its strong enrichment in DRACH motif (Pratanwanich et al., 2021). Finally, MINES only detects four RRACH motifs (AGACT, GGACA, GGACC, and GGACT) and overlooked other 8 RRACH motifs (Lorenz et al., 2020). Nanom6A provides method using Nanopore direct RNA reads for identification of both qualitative and quantitative m⁶A modification in single-base resolution based on XGBoost algorithm, providing a highly precise transcriptome-wide identification, quantification and sequence contexts (RRACH) of m⁶A modification (Gao et al., 2021). In this study, we provide detail procedure for high accuracy of quantitation of RNA modification in single-base resolution. In brief, we provide comprehensive procedure on the identification of AS, APA and m⁶A based on DRS or PacBio Iso-Seq data. The Linux Bash Shells and scripts for Nanom6A and PRAPI analysis are now available in https://github.com/GuInNGS/NanoPrapi and https://github.com/Bio-protocol/Pipeline_for_Identification_m6A_AS_and_APA_with_Long_Read.