Senegalese Sole RNA-seq Pre-processing and Analysis

A total of 111 Illumina RNA-seq libraries of $2\times 75$ nt per read ( $>1800$ million reads) and 5,663,225 GS-FLX RNA-seq single-end reads of Senegalese sole were available from BioProjects PRJNA255461, PRJNA241068 and PRJNA261151^{7 (link)}. These datasets comprised nine experiments that included negative control and drug treatment in triplicate libraries, or even time-course responses. To optimize the number of reads to assemble, only one out of each biological replicate was randomly selected by experiment, and when a time series was available, only the first and the last sampling points were chosen (Supplementary File 1). Libraries containing GS-FLX reads were pre-processed using SeqTrimNext (based on SeqTrim^{67 (link)}), while SeqTrimBB (based on BBmap suite⁶⁸ ) was used for Illumina reads. Default parameters were applied in both cases. Illumina reads shorter than 60 bp were discarded, whereas the threshold for GS-FLX reads was set in 90.
Pre-processed reads (Supplementary File 1) were mapped onto the S. senegalensis genome draft^{33 (link)} using Bowtie2^{69 (link)} with the –no-mixed parameter to discard reads from unpaired alignments. SAMtools^{70 (link)} was used for mapping manipulation and read counts.