Back Article

Figure 1. 

Principle and pipeline overview of resolveS. (a) Schematic representation of RNA-seq library strandedness in paired-end sequencing strategies. The known original RNA transcript orientation is used for strandedness inference. In fr-secondstrand libraries, reads from R1.fastq align predominantly in the same orientation as the original RNA transcript, whereas in fr-firststrand libraries they align predominantly in the opposite orientation. Unstranded libraries exhibit no consistent orientation bias. Natural antisense transcription may contribute a minor fraction of discordant signals. (b) Main application areas of resolveS. resolveS supports multiple applications in transcriptomic data analysis, including RNA-seq data quality control, verification of stranded library information, annotation of public datasets, integration of data across studies, large-scale transcriptomic analyses, and automated pipeline construction. (c) Overview of the resolveS pipeline. A compact universal rRNA reference database derived from SortMeRNA references is indexed with Bowtie2 and used for genome-independent strandedness inference. Paired-end RNA-seq reads (R1 and R2) are sequentially processed using a fixed-depth strategy (e.g., one million read pairs), enabling runtime scalability independent of total sequencing depth. During alignment, both R1 and R2 reads are utilized to improve mapping accuracy and pairing confidence within conserved rRNA regions.