Generation and analysis of expressed sequence tags in the extreme large genomes Lilium and Tulipa

Abstract

BackgroundBulbous flowers such as lily and tulip (Liliaceae family) are monocot perennial herbs that are economically very important ornamental plants worldwide. However, there are hardly any genetic studies performed and genomic resources are lacking. To build genomic resources and develop tools to speed up the breeding in both crops, next generation sequencing was implemented. We sequenced and assembled transcriptomes of four lily and five tulip genotypes using 454 pyro-sequencing technology.ResultsSuccessfully, we developed the first set of 81,791 contigs with an average length of 514 bp for tulip, and enriched the very limited number of 3,329 available ESTs (Expressed Sequence Tags) for lily with 52,172 contigs with an average length of 555 bp. The contigs together with singletons covered on average 37% of lily and 39% of tulip estimated transcriptome. Mining lily and tulip sequence data for SSRs (Simple Sequence Repeats) showed that di-nucleotide repeats were twice more abundant in UTRs (UnTranslated Regions) compared to coding regions, while tri-nucleotide repeats were equally spread over coding and UTR regions. Two sets of single nucleotide polymorphism (SNP) markers suitable for high throughput genotyping were developed. In the first set, no SNPs flanking the target SNP (50 bp on either side) were allowed. In the second set, one SNP in the flanking regions was allowed, which resulted in a 2 to 3 fold increase in SNP marker numbers compared with the first set. Orthologous groups between the two flower bulbs: lily and tulip (12,017 groups) and among the three monocot species: lily, tulip, and rice (6,900 groups) were determined using OrthoMCL. Orthologous groups were screened for common SNP markers and EST-SSRs to study synteny between lily and tulip, which resulted in 113 common SNP markers and 292 common EST-SSR. Lily and tulip contigs generated were annotated and described according to Gene Ontology terminology.ConclusionsTwo transcriptome sets were built that are valuable resources for marker development, comparative genomic studies and candidate gene approaches. Next generation sequencing of leaf transcriptome is very effective; however, deeper sequencing and using more tissues and stages is advisable for extended comparative studies.

Mechanisms of vegetative propagation in bulbs : a molecular approach

“…To get a better impression of the quality and completeness of the functional annotated datasets, we compared our transcriptomes and annotation with previously published transcriptomes of tulip and lily (Shahin et al 2012b). Initially, we performed a BLAST search at the nucleotide level to determine how well we covered the transcripts present in these publicly available datasets.…”

“…The number of transcripts and predicted genes in our non-filtered transcriptomes may be highly over-estimated taking into account that there are only 27,024 protein coding gene models in the recently sequenced monocot genome of pineapple (Ming et al 2015), 39,045 genes reported for rice-the monocot model species-(Sequencing ProjectInternational Rice 2005) and 81,791 for tulip, based on a previous transcriptome sequencing effort (Shahin et al 2012b). …”

“…Although various reasons can be brought forward for this, the large genome sizes for bulbous plants (Tulip ~25000 Mbp; Lily ~36000 Mbp) and technical difficulties in isolating e.g. RNA from bulb scales have been particularly decisive in this (Shahin et al 2012a). …”

“…Tulip and lily (Tulipa sp and Lilium sp) are ornamental geophytes with an estimated genome size of 25 and 36 GB, respectively (Shahin et al 2012b). One of the first studies of a transcriptome characterization for both species was done by Shahin et al in 2012 using 454 pyro-sequencing technology of messenger RNA (mRNA) from leaves (Shahin et al 2012b). They obtained 81,791 unigenes for tulip with an average length of 514 bp and 52,172 unigenes for lily with an average length of 555 bp.…”

“…The Illumina HiSeq platform was used to sequence a pooled sample for lily and for tulip, each made up of a mixture of equal amounts of poly adenylated mRNA obtained from flowers, stem, leaves, bulb and bulblets. Even though short reads are generated with the Illumina HiSeq platform, a tremendous throughput can be reached, resulting in an improved coverage of rare transcripts in comparison to the other platforms used in some of the previous transcriptome studies of bulbous species (Shahin et al 2012b;Kamenetsky et al 2015). The generated data was used to assemble reference transcriptomes for tulip and lily.…”