Transcriptome Assembly and Analysis

The raw reads were assessed using CASAVA (v1.8, Illumina Inc.) and filtered to produce clean reads. First, the adaptors were removed. Second, the reads with an N ratio of more than 10% were removed. Finally, low-quality reads (Q_phred value of <20 and >50% bases) were trimmed, with a Q_phred cut-off value of −10 log₁₀(e). Short reads obtained by sequencing were assembled using Trinity software (Grabherr et al., 2011 (link)) and the unigenes used for subsequent analysis. TransDecoder was used to identify candidate coding regions based on the following criteria: (1) An ORF that meets the minimum limit length (200 bp) can be found in the transcript sequence. (2) The logarithm likelihood score was greater than 0. (3) The logarithm likelihood score of the first open reading frame (ORF) is the maximum compared with the other 5. (4) If the candidate ORF was completely involved in other ORFs, we only report the longest ORF.