A systematic benchmark of Nanopore long-read RNA sequencing for transcript-level analysis in human cell lines

Area of Science:

• Genomics
• Molecular Biology
• Transcriptomics

Background:

• The human genome encodes over 200,000 RNA transcripts.
• Alternative splicing generates numerous similar RNA isoforms from single genes, complicating quantification.
• Accurate RNA isoform quantification is crucial for understanding gene expression.

Purpose of the Study:

• To evaluate and compare different RNA sequencing (RNA-seq) protocols for studying RNA transcript expression.
• To assess the ability of various RNA-seq methods to identify and quantify RNA isoforms.
• To establish a comprehensive resource for benchmarking computational methods for transcriptomic analysis.

Main Methods:

• Profiling of seven human cell lines using five distinct RNA-sequencing protocols.
• Inclusion of short-read cDNA, Nanopore long-read direct RNA, amplification-free direct cDNA, PCR-amplified cDNA sequencing, and PacBio IsoSeq.
• Utilization of spike-in controls and transcriptome-wide N⁶-methyladenosine (m⁶A) profiling data.

Main Results:

• Significant differences observed in read length, coverage, throughput, and transcript expression across protocols.
• Long-read RNA sequencing protocols demonstrated superior ability in robustly identifying major RNA isoforms.
• The SG-NEx dataset enabled identification of alternative isoforms, novel transcripts, fusion transcripts, and m⁶A modifications.

Conclusions:

• Long-read RNA sequencing offers enhanced resolution for isoform-level transcript profiling.
• The SG-NEx data resource is valuable for developing and validating computational tools for complex transcriptomic analysis.
• This study provides a foundation for improved understanding of alternative splicing and RNA modifications.