OikoBase: a genomics and developmental transcriptomics resource for the urochordate Oikopleura dioica

Abstract

We report the development of OikoBase (http://oikoarrays.biology.uiowa.edu/Oiko/), a tiling array-based genome browser resource for Oikopleura dioica, a metazoan belonging to the urochordates, the closest extant group to vertebrates. OikoBase facilitates retrieval and mining of a variety of useful genomics information. First, it includes a genome browser which interrogates 1260 genomic sequence scaffolds and features gene, transcript and CDS annotation tracks. Second, we annotated gene models with gene ontology (GO) terms and InterPro domains which are directly accessible in the browser with links to their entries in the GO (http://www.geneontology.org/) and InterPro (http://www.ebi.ac.uk/interpro/) databases, and we provide transcript and peptide links for sequence downloads. Third, we introduce the transcriptomics of a comprehensive set of developmental stages of O. dioica at high resolution and provide downloadable gene expression data for all developmental stages. Fourth, we incorporate a BLAST tool to identify homologs of genes and proteins. Finally, we include a tutorial that describes how to use OikoBase as well as a link to detailed methods, explaining the data generation and analysis pipeline. OikoBase will provide a valuable resource for research in chordate development, genome evolution and plasticity and the molecular ecology of this important marine planktonic organism.

Figure 1.

Complex developmental gene expression traffic in the compact O. dioica genome as visualized in the OikoBase genome browser. In addition to the normal browser output, the developmental stages are indicated in larger font in red type for clarity. Flanking the browser expression data, vertical timelines are included to illustrate key cell cycle (grey-scale, left-hand side) and organismal (blue-scale, right-hand side) processes and transitions covered by our transcriptomics analysis. The illustrated region includes testes-specific gene models (GSOIDG00004881001 and GSOIDG00004882001, green shading) that are also more weakly expressed in Day5 animals, a sub-portion of which contain testes at an earlier developmental stage. Immediately to the left of these is a ubiquitously expressed gene model (GSOIDG00004880001, purple shading). Notably, gene models GSOIDG00004883001 and GSOIDG00004885001 (pink shading) are expressed predominantly from tailbud to Day3, and RACE analysis indicated that they are part of the same gene (unpublished data; the red ‘I’ indicates their connectedness), expressed from a promoter left of GSOIDG00004883001. However, the short 5′-exon (GSOIDG00004883001) is not expressed in the ovary at which point GSOIDG00004885001 expression is driven by an internal bidirectional promoter also driving GSOIDG0004884001 (yellow shading), as confirmed by RACE and qRT–PCR analysis (unpublished data). This high diversity of developmentally regulated expression occurs in a region measuring <15 kb and indicates that our transcriptome analysis at high temporal and spatial resolution can improve gene model annotation. The Y-axis scale is exponential (log₂).

Figure 2.

Data mining in OikoBase. (A) Browser window of a 50-kb segment of scaffold 1 showing several genes, with the tailshift expression track selected as an example of a time point. (B) By placing the pointer over a gene model (1) (here, the sixth gene from the left) a balloon appears that specifies the gene name, gata-3 (assigned by Blast2GO) and its genomic coordinates (2). (C) By clicking on the gene model, a different balloon (3) is opened in its place which allows links to both the GO website (http://www.geneontology.org/) (4) and the InterPro website (http://www.ebi.ac.uk/interpro/). By clicking on either link, the browser is redirected to the relevant GO or InterPro terms associated with this gene which catalogue information regarding biological processes, molecular functions and cellular components as well as evolutionarily conserved protein domains. Options to directly download nucleotide (TRANSCRIPT SEQ) and protein (PEPTIDE SEQ) are also provided in this balloon.

Figure 3.

Obtaining developmental expression profiles for O. dioica homologs of genes-of-interest from other organisms. To provide an example of how to use this pipeline, the human cytoplasmic actin protein has been selected as the query sequence. First, the ‘Programs available for the BLAST search’ page (1) is accessed by clicking the Facepage BLAST link. The desired Program should then be matched up with the appropriate Database using the dropdown menus (2). Note that selecting ‘blastp’ (as in this example) or ‘blastx’ will automatically pair these programs with the ‘O. dioica peptides reference’ database. After executing the search (3), the desired putative O. dioica homologs are displayed on the BLAST search results page where unique GSOID identifiers and corresponding gene names are provided (4). By clicking on the desired live link GSOIDP identifier the user is taken to the location of the putative homolog (5). Subsequently, any desired additional annotation or transcript tracks can be activated in the browser. Here, the expression of an O. dioica cytoplasmic actin homolog is shown at the tailshift stage with the view zoomed to 2 kb.