Advancing genomics through the Global Invertebrate Genomics Alliance (GIGA)

Additional keywords: biodiversity, evolution, genetics, Metazoa.

Genomic research will likely exhibit exponential growth for years to come (Stephens et al. 2015). In anticipation of this, the Global Invertebrate Genomics Alliance (GIGA) formed to address the multiple needs and opportunities of the expanding invertebrate genomics community (GIGA COS 2014). GIGA complements taxonomically similar but differently focused efforts, such as the arthropod-focused Genome 10K (Genome 10K COS 2009; Koepfli et al. 2015) and the Arthropod Genomics Consortium (i5k) (Robinson et al. 2011), and interacts with other initiatives, such as the Genomic Observatories Network (Davies et al. 2014), the Global Genome Initiative (GGI, http://naturalhistory.si.edu/ggi/), and the Ocean Genome Legacy (OGL, http://www.northeastern.edu/cos/marinescience/ogl/). These latter initiatives were created to archive diverse arrays of specimens, tissues, nucleic acid samples, voucher material and supporting data in a global effort to help coordinate and disseminate these data and materials across the broader scientific community.

GIGA has adopted a ‘bottom up’ philosophy to advance its community goals, with voluntary and vigorous contributions of members building a consensus approach to avoiding duplication of research efforts through effective coordination within the community (see Fig. S1 for the distribution of member countries, available as Supplementary material). GIGA was launched with an initial workshop in 2013 and an inaugural white paper (GIGA COS 2014). The second GIGA workshop (March 2015 at the Ludwig-Maximilians-Universität, Munich, Germany) reviewed the progress of this growing initiative. More than 80 scientists from 23 countries participated in the 3-day event, which explored aspects of project implementation and comparative analyses across invertebrates as well as training and funding efforts and opportunities. We briefly report here on the major outcomes of these discussions.

The first GIGA white paper (GIGA COS 2014) and two workshops identified many major hurdles hindering large-scale genome research, which have started to be addressed by concerted efforts from the community. For example, GIGA is tallying the growing number of completed invertebrate genomes from individual laboratories (Dunn and Ryan 2015), tracking new advances in sequencing and bioinformatics and monitoring the investment of specific research consortia. To keep abreast of these ongoing efforts, the updated GIGA homepage (www.giga-cos.org) features improved access to member projects and relevant publications, and now provides phylogenetically sorted links to organismal genomes and curated transcriptome resources (see Box 1). In addition, the GIGA listserver (https://lists.lrz.de/mailman/listinfo/giga) was set up to facilitate information exchange within the community.

While GIGA supports the focused efforts of single laboratories, the consortium acknowledges the need to collectively target priority species for sequencing due to their economic, evolutionary and/or conservation significance (see Box 2). These priority species (not exhaustive) should occupy a distinct space in the GIGA framework, hopefully becoming the focus of future concerted community efforts (Table 1).

Table 1.  Proposed priority species for whole genome sequencing
N/A, not available; TBD, to be determined. We are limited in the number of taxa we can list, so this table only shows a small cross section of priority taxa

The GIGA II workshop identified ‘divide-and-conquer’ as the best strategy for the distribution of efforts concerning the coordination of genome and transcriptome sequencing and knowledge exchange in different taxonomic and phylogenetic groups and scientific communities. For example, clade-specific ‘ambassadors’ (see http://giga-cos.org/index.php/members) were chosen to recruit within their own smaller communities and discuss candidate taxa for future genomics. The ambassador framework will also effectively distribute the efforts to provide a complete and comprehensive overview of current and future invertebrate genomic and transcriptomic resources, and GIGA members will have direct ‘go-to’ experts for the organisms of interest.

High-throughput sequencing technologies have boosted the field of comparative genomics (Green et al. 2015; Koepfli et al. 2015). The volume of sequence data available through public databases such as the International Nucleotide Sequence Database Collaboration (INSDC) portal and its members, GenBank (at the National Center for Biotechnology Information, NCBI), European Nucleotide Archive (ENA) (at European Bioinformatics Institute) and DNA Data Bank of Japan (DDBJ), has increased exponentially (Baxevanis 2011; Stephens et al. 2015). As overall sequencing costs continue to decrease, individual laboratories and community efforts can generate ever larger amounts of novel genomic and transcriptomic data from non-model organisms. Aggregative databases such as Compagen (http://compagen.org) (Hemmrich & Bosch 2008) or Reefgenomics (http://reefgenomics.org) (Liew et al. 2016) coordinate data presentation and analyses for particular species or ecosystem groups. While the development of a single database covering data from a wide variety of invertebrate species would likely have wide appeal, the programming, curatorial and funding burden of producing and maintaining a centralised resource goes beyond the reach of GIGA and similar consortia at present. Many researchers only need simple queries (e.g. BLAST-based sequence similarity searches), while a minority needs more complex, multivariate queries (e.g. synteny-based orthology inference). We propose an approach that will cater to both needs by providing: (1) a central sequence database that focuses on simple queries accessing a large range of data, in combination with (2) more focused, de-centralised species databases that satisfy complex queries and more specific requirements (Parkhill et al. 2010). The GIGA homepage will serve as a hub from which to access these platforms.

Compagen.org, originally designed to cater to the early branching (non-bilaterian) animal community, will be extended to serve as our central sequence database providing access to processed datasets such as assembled transcriptomes/genomes and predicted peptides, thereby complementing sequence datasets available from public domains (e.g. NCBI Sequence Read Archive (SRA), NCBI dbEST (GenBank’s sequence data on cDNA sequences or expressed sequence tags) and ENA trace archives). Compagen.org provides a critical extension to public databases due to the disconnect between what is submitted in the form of primary sequence data and what is available in the form of assembled and annotated data (Fig. 1). Having compagen.org as a central repository will clearly facilitate provision of assembled invertebrate sequence data as well as retrieval of data for analyses. This present choice does not preclude the additional development of other (taxon or feature specific) portals to GIGA-related sequences in the future.

Fig. 1.  Disconnect between primary (Sequence Read Archive, SRA) and assembled (Transcriptome Shotgun Assembly, TSA) high throughput sequence data available for invertebrates at the National Center for Biotechnology Information (NCBI). The plot shows a strong difference between the number of species for which primary transcriptomic sequence data are available (SRA) in comparison with the number of species for which assembled transcriptomes are available (TSA). One aim of GIGA is to bridge this gap by providing a hub for assembled transcriptome and genome data currently not accessible from public databases.

For taxon groups studied by larger research communities, it is realistic (and tractable) to develop or coordinate independent, researcher or laboratory-curated, taxon-specific databases, such as http://www.spongebase.net for sponges or http://reefgenomics.org for coral reef invertebrates. These databases can accommodate the needs of the specific research groups focusing on specific taxa or groups of taxa, emphasising genomics, transcriptomics, variation or multi-species interaction as required. In connection with those, several single-species databases already exist, including the Aiptasia genome platform (http://aiptasia.reefgenomics.org/) (Baumgarten et al. 2015) and the Hypsibius dujardini genome server (http://www.tardigrades.org) (Koutsovoulos et al. 2016). From these efforts, cross-species databases can arise, such as http://comparative.reefgenomics.org, which provides a comparative platform for coral transcriptomes focusing on orthologous genes that can be used to interrogate coral species-specific genes (Bhattacharya et al. 2016). Community-managed data repositories should be extended to link genomic efforts to physical specimens deposited in natural history collections or to genetic resources deposited in genetic repositories in natural history museums or in other institutions. Overall, the accessibility of novel sequence data enhanced with metadata will open multiple doors to new hypotheses and crossover collaborations, thereby propelling broad scientific, technological and societal advances (i5K Consortium 2013; Dubilier et al. 2015; Green et al. 2015; Koepfli et al. 2015).

The GIGA II workshop highlighted the common goals shared among i5k and GIGA initiatives, primarily with regards to help in structuring and advising the community for handling large genome project initiatives. Other key shared goals include funding, expert genome annotation and dissemination of results. There will be additional opportunities for collaboration between research groups involved in these and other large invertebrate genomics and transcriptomics consortia (such as 1Kite or the nematode.net) due to the shared methodological and bioinformatic approaches and training opportunities.

Achieving the broad goals of GIGA will require new commitment and investment in training and educating the next generation of genome scientists (Fig. 2). Computational biology skills are needed to allow students to navigate, customise and populate increasingly complex databases, but additionally, training in aspects of organismal and evolutionary biology, ecology and biodiversity legislation that regulates the exchange of genetic resources across national boundaries needs to be provided. Besides a sweeping view and comparison of various Big Data issues, recent reviews provide a compelling argument for the genomics community to prepare an educational foundation to handle the deluge of genomic data and its associated metadata (Stephens et al. 2015; Dougherty et al. 2016). In general, current genomics and bioinformatics curricula appear relatively staid, while independent bioinformatics workshops (e.g. evomics.org, and those listed at http://meetings.cshl.edu/courseshome.aspx) can provide cutting edge approaches and showcase innovation; however, scheduling may be more uncertain and irregular for the latter and those specialist courses do not cover all aspects needed for the holistic training needed to cater for future job markets in (invertebrate and/or biomedical) genomics.

Fig. 2.  Proposed GIGA deliverables and funding opportunities. Funding opportunities intersect with GIGA activities. Various activities have cumulative and synergistic effect. Specific areas for funding opportunities and requests are indicated by stars, with the number of stars proportional to estimated relative costs (e.g. workshops and conferences cost less than current sample collection or bioinformatics training – see Stephens et al. 2015)). ‘At Large’ genomics consortia can represent new and old groups with similar or complementary goals (Genome 10K 2009; Koepfli et al. 2015).

Currently, targeted funding for invertebrate genomes pales in comparison to the investments being made in vertebrate genomics (Couzin-Frankel 2012; AAAS 2015), so we compiled a limited selection of some of the agencies that we thought might contribute funding to further GIGA efforts (Table 2). Realising these will eventually expire, individual scientists, small collaborative cohorts or larger groups will have to forge imagination, fortitude and the scientific foundations and infrastructure discussed herein to persuade funding entities on the values of each particular project. However, during GIGA II the current dearth of international, multilateral funding programs that allow researchers in different continents (e.g. EU, Asia and Americas) to collaborate became strikingly obvious.

Table 2.  Potential funding agencies/opportunities
JGI, Joint Genomics Institute; NSF, National Science Foundation

Many benefits derived from single or multi-species genome sequencing projects have been described, realised and projected in recent years (GIGA COS 2014). Discoveries will continue to accrue with further innovations. Efforts will be further enhanced through the GIGA framework, which coordinates and promotes the concerted contributions of diverse experts. In the words of Immanuel Kant by way of Durant (1938), ‘Science is organized knowledge; wisdom is organized life’. In a similar spirit, GIGA will facilitate scientific organisation, and consequently the collective progress (wisdom) of humanity.