3. Results
For the epiGBS libraries, we recovered 85.8 million reads after demultiplexing. The number of reads per sample averaged 2.5 million and ranged between 0.6-4.5 million. Sequenced reads were evenly distributed across all four populations (21.4 ± 1.7 million reads per population), and across all 12 groups (7.2 ± 2.1 million reads per group) (Table S1). For the RNAseq libraries, we recovered 101.8 million reads after quality trimming. The number of reads per sample averaged 8.5 million and ranged between 6.9-10.2 million (Table S2). As for the epiGBS dataset, the sequencing power was evenly distributed across populations (51.0 and 50.8 million reads for Sc3 and Sc4 respectively), and the four groups (25.5 ± 1.2 million reads per group) (Table S2).
The overall read mapping rate to the de novo transcriptome assembly ranged between 83-86% across samples (Table S3). We found a total of 390 complete orthologues, out of which 73% were complete and single copy. In total, 26,638 transcripts were successfully annotated. A summary of the main GO terms in the annotated transcriptome is shown in Table S3.