Thanks for your scrupulous reviewing regarding our data analysis. I am Chungling Zhang , one of the coauthor in this manuscript and I am willing to answer your raised question. Here are few things need to be declared firstly. Because there are moso bamboo genomic resources as the reference, so after the original image data was transferred into sequence data as raw data or raw reads via base calling, the clean reads were selected via sequencing analysis, quality control and assessment by using Illumina PIPILINE Software, and then they were mapped to the reference genome and genes. However, ‘Sequence data analysis and assessment’ may be more accurate than ‘sequence data analysis and assembly’ in original describing this section. Thanks again for your consideration.

Thanks for your scrupulous reviewing regarding our data analysis. I am Chungling Zhang , one of the coauthor in this manuscript and I am willing to answer your raised question. Here are few things need to be declared firstly. Since there are moso bamboo genomic resources as the reference, the original image data was transferred into sequence data as raw data or raw reads via base calling after the libraries had been sequenced. Then, the clean reads were selected via analysis and assessment of base composition and quality, and filtering of raw reads by using Illumina PIPILINE software. Since the algorithms used in transcriptome construction of the reads provided by the Illumina platform may be severely inhibited by sequencing errors, a stringent cDNA sequence filtering process was employed to select clean reads. Firstly, reads with adaptors were removed. Secondly, reads in which unknown bases were more than 5% were removed. Finally, low quality reads which the percentage of low quality bases (base quality≤20) was over 30% were removed. Directly following next process, the clean reads were mapped to the reference genome and genes. Therefore, ‘Sequence data quality control and filtering’ may be more accurate than ‘sequence data analysis and assembly’ in original describing this section. When there were no genomic resources as the reference in transcriptome, sequence data was analyzed and assembled after the libraries had been sequenced. Firstly, the original image data was transferred into sequence data as raw data or raw reads via base calling. Secondly, the clean reads were selected after removing reads containing sequencing adapters, reads in which unknown bases were more than 5% were removed and reads of low quality which the percentage of low quality bases (base quality≤20) was over 30%. Thirdly, the reads were assembled with short reads assembling program–Trinity (Grabherr M.G., Haas B.J., et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nature Biotechnology. 2011, doi: 10.1038 /nbt.1883.) which combines three independent software modules: Inchworm, Chrysalis, and Butterfly, applied sequentially to process large volumes of reads. Finally, non-redundant unigenes were acquired and used for further analysis, after the result sequences of assembly were taken into further process of sequence splicing and redundancy removing with sequence clustering software. Thanks again for your consideration.