| Cucumber is the largest protected cultivation of vegetable,it is also an important model plant for sexual development and vascular transport studies.Cucumber whole genome sequence has been published,and on this basis,the whole genome SSR marker has been completed,covering 3.3 million variation Loci group map,they are important platform and tools for cucumber genetic and functional genome research.Cucumber transcriptome studies are common,but co-expression network studies are rarely reported.In this experiment,we selected 10 different tissues from domestic cucumber 9930 then conducted massively parallel complementary DNA sequencing(RNA-Seq).After removing the bar code and the low quality reads,high-quality reads were mapped to the cucumber genome sequence by Tophat2.Cufflinks was used to calculate FPKM values of 24,274 genes of 10 tissues.In order to remove genes with low expression,if the gene’s max FPKM from 10 tissues is under 5,this gene is removed.At last 16,924 genes were obtained.The expression data obtained above was used to construct a co-expression network by WGCNA(weighted gene co-expression network analysis)package in R.Then,we received a total of 1134 modules.The genes in modules can be regarded as co-expressed genes,their expression similarity are relatively high and have similar expression patterns.the average value of the correlation between genes in module less than 0.9 modules are removed and finally obtain 839 modules,11,844 genes.Co-expression genes are have similar expression pattern,they may be specifically associated with tissues.In order to calculate the correlation between modules and tissues,first principal component analysis(principle component analysis,PCA)was applied in each module to get ME(module eigengene),ME may indicate the expression pattern of the genes in this module.By calculating the correlation between the MEs and the tissues,a total of 323 modules were found tissue-specific.Functional enrichment analysis by topGO found that these specific modules enriched functions associated with specific tissues.Co-expression genes often form a cluster on chromosome.Here,cluster is defined as genes whose physical location interval is less than 25kb.839 modules were searched and found a total of 220 clusters in 71 modules,the cluster generally contains 2 to 5 genes,clustered genes often have related function.Co-expression genes may be regulated by the same transcriptional factors.We extracted sequence before start codon 2kb from genes in 839 modules.The sequences were uploaded to the website(PLACE)for motif analysis.After enrichment analysis,a total of 367 motifs were found,among them,six motifs were reported to play a role in cucumber plants.Cucumber bitterness biosynthetic pathway studies have been published,11 genes were found.We found 4 modules,including 10 genes of 11 genes.Functional enrichment analysis of 4 modules also indicated that these modules associated with bitterness biosynthesis.Genes in these modules were found physically clustered in chromosome.The method described in this paper is a combination of transcriptome sequencing and network analysis,and we found gene co-expression modules in cucumber by this method.This paper provides a very important cornerstone and data analysis support for cucumber. |
