Multi-omics Data Mining Based On Important Secondary Metabolites Of Ginkgo Biloba

As one of the most ancient seed plants currently,Ginkgo biloba is widely regarded as the‘living fossil' in plants?Since its unique morphology and physiology characteristics,together with its debates on the phylogenetic status in seed plants,G.biloba plays a very important role in evolutionary studies.Apart from this,G.biloba as well has very high medical values,including its very important secondary metabolisms-flavonoid and terpenoid which are widely used.In this study,transcriptome data of G.biloba kernels across different developmental periods,together with the transcriptome/metabolome data of G.biloba mature leaves and the newly published G.biloba draft genome were analyzed.On the aspects of transcriptome,genome and metabolome,and we took the important secondary metabolites as the breakthrough point,majorly studying the following issues.For instance,different gene expression across developmental periods in G.biloba kernels,the alternative splicing in G.biliba genome and genes related with secondary metabolics,codon usage bias in G.biloba nuclear genome,the establishment of SNP calling in tree species,and the changes of metabolics between different time periods in G.biloba leaves.We wish our results could be beneficial to the studies on the synthesis and regulation mechanisms that were involved in secondary metabolics.Our main results were listed below:1.For the first time,the transcriptome of G.biloba kernels across five periods was sequenced using Illumina paired-end sequencing platform,and comparison results across different periods revealed that there were in total of 3,869 differential expressed unigenes,including 66 unigenes involved in terpenoid lactone synthesis and 161 unigenes involved in flavonoid synthesis.According to the result of real-time PCR,we assumed that the synthesis of bilobalide might interfere with the synthesis of Ginkgolide.Based on the draft genome of G.biloba,alternative splicing events were analyzed in G.biloba genome,and the results revealed that on average there were 33.78% unigenes involved in alternative splicing.The frequencies of alternative splicing increased with gene expression and time.In addition to that,genes annotated with terpenoid lactones were specially analyzed with AS events.2.Amount to 17,579 unigenes with the length longer that 300 bp were selected for codon usage bias analysis among 68,547 functional unigenes.The results revealed that A/U-ending codons were more preferred in G.biloba.Besides,there seemed to be a tendency from A/U ending to G/C ending,ranging from gymnosperms to dicots,and then to monocots.The comparison between high-expressed and low-expressed genes revealed that high-expressed genes seemed to be more bias to using G/C-ending codons.Then these unigenes were classified into different categories,involving the genes with flavonoid and terpenoid lactones,and the codon usage bias analysis upon them revealed that genes involved in environment adaptation were more bias to using G/C-ending codons.The codon usage bias in G.biloba was dominated by natural selection,and there was a tendency from A/U-ending to G/C-end codons.Codon adaptation index(CAI)which was used for extimating gene expression in many studies was discussed as well,and we think this index was not suitable for the estimation of gene expression in higher eukaryotes.3.SNP is the important molecular marker when studying the relationship between genotype and phenotype,while the large-scale SNP calling work in tree species was still at the starting edge.In our study,three widely used SNP calling programs were evaluated,including SAMtools,GATK and Freebayes,and two popular sequence alignment programs – BWA and Bowtie 2 were analyzed as well.The results revealed that these SNP calling programs showed very low accordance,which may easily resulting in the false positive results upon downstream experimental progress.Among all six combinations,BWA+SAMtools were most suitable for the large-scale SNP prediction in tree species.Besides,457,594 potential SNPs were predicted,and eight unigenes involved in flavonoid metabolism were experimentally validated.4.To further study the detailed differences on metabolism and gene function across different periods,G.biloba mature leaves were collected from three time points when have the most significant metabolic variations(June,August and October),and their metabolites were identified respectively.Amount to 780 metabolites were successfully isolated with UPLC-MS/MS method,including 133 flavonoids(69 flavones,41 flavonols and 23 flavanones),5 terpenes together with72 organic acids and their derivatives.Among all detected 133 flavonoids,83 of them were not contained in KEGG database,and these new identified metabolites would be a good supplement for the studies on secondary metabolites.The results revealed that though the major componentflavonoid was not essential for G.biloba,its symbiotic metabolites-stilbene could play an important role in resisting environment change and plant diseases.Combination analysis between differential unigenes and metabolites showed that a total of 2,523 unigenes and 203 metabolites were strongly connected,and this information would be beneficial to the following molecular mechanisms on G.biloba secondary metabolites.