Ramping up biodiversity discovery via online quantum contributions

The pace of species discovery and documentation remains too slow on a human-altered planet in the midst of a massive extinction event. Increasing this pace requires altering conventional workflows. In this review, we propose that systematics needs to shift to a model of quantum contributions whereby species hypotheses are published as they are formulated and data as they are collected in web-based repositories and content-management systems. If our recommendation is followed, many species will make their first appearance on the Internet as candidate new species before documentation is complete. Acknowledging the changes that we describe may be controversial, we discuss problems that may be encountered along with possible solutions. The pace of species discovery and documentation remains too slow on a human-altered planet in the midst of a massive extinction event. Increasing this pace requires altering conventional workflows. In this review, we propose that systematics needs to shift to a model of quantum contributions whereby species hypotheses are published as they are formulated and data as they are collected in web-based repositories and content-management systems. If our recommendation is followed, many species will make their first appearance on the Internet as candidate new species before documentation is complete. Acknowledging the changes that we describe may be controversial, we discuss problems that may be encountered along with possible solutions. The discovery and documentation of biodiversity on Earth is proceeding at an inadequate pace, especially in the most diverse groups. Approximately two million species have been documented so far and an order of magnitude more may remain to be discovered (http://www.catalogueoflife.org/annual-checklist/2009) [1Chapman A.D. Numbers of Living Species in Australia and the World.2nd edn. Australian Biological Resources Study, 2009Google Scholar, 2Whitman W.B. The modern concept of the prokaryote.J. Bacteriol. 2009; 181: 2000-2009Crossref Scopus (36) Google Scholar, 3Mora C. How many species are there on earth and in the ocean?.PLoS Biol. 2011; 9: e1001127Crossref PubMed Scopus (1600) Google Scholar]. An estimated 6200–18 000 species of eukaryotes are described per year [3Mora C. How many species are there on earth and in the ocean?.PLoS Biol. 2011; 9: e1001127Crossref PubMed Scopus (1600) Google Scholar, 4International Institute for Species Exploration State of Observed Species 2010. International Institute for Species Exploration, 2010Google Scholar]. These estimates yield the frightening prospect of a millennium of basic discovery work at the present pace. In addition, anthropogenic impacts are dramatically altering the biota of the Earth, with extinction rates now as high as 27 000 known species per annum [5Wilson E.O. Diversity of Life. Harvard University Press, 1992Google Scholar, 6Sax D.F. Gains S.D. Species invasions and extinction: the future of native biodiversity on islands.Proc. Natl. Acad. Sci. U.S.A. 2008; 105: 11490-11497Crossref PubMed Scopus (514) Google Scholar]. Biologists need to embrace biodiversity discovery and documentation with increased urgency, commitment and innovation: if the status quo is maintained, it will take too long. Many improvements can be made to the way in which biodiversity is discovered and documented that will increase the pace [7Berger J.K. Mission possible: ALL Species Foundation and the call for discovery.Proc. Calif. Acad. Sci. 2005; 65: 114-118Google Scholar]. Here, we focus on what we view as a vital change: altering how and when discoveries are shared. We believe that species discovery will proceed most rapidly if data and species hypotheses are published as they are generated in Web-based data repositories. We articulate the advantages of this approach below and in Box 1.Box 1Benefits of publishing systematic research through quantum contributions on the Internet•Making knowledge of new species available more quickly before they are named or fully documented in revisionary works.•Placing new data more rapidly in the context of existing data, and allowing them to be explored using synthetic tools linked to Internet repositories.•Enabling collaborative work in which the complementary resources and talents of individuals can more quickly establish the existence of new species.•Reducing the loss of systematic knowledge that now occurs owing to the demise of systematists and their personal digital or paper records. •Making knowledge of new species available more quickly before they are named or fully documented in revisionary works.•Placing new data more rapidly in the context of existing data, and allowing them to be explored using synthetic tools linked to Internet repositories.•Enabling collaborative work in which the complementary resources and talents of individuals can more quickly establish the existence of new species.•Reducing the loss of systematic knowledge that now occurs owing to the demise of systematists and their personal digital or paper records. Any method of discovering species will be time-consuming when a taxonomic group is species-rich, occurs in logistically challenging areas, or requires special techniques to collect or acquire the necessary data. The traditional approach of producing large revisionary works slows the pace further and puts that work at risk of loss for the following reasons: (i) The traditional process of species discovery and documentation is lengthy and few tangible products are published along the way. Projects often take multiple years and, thus, are vulnerable to events that impede their completion. (ii) Until recently, much of this research has occurred in systematists’ minds and on paper. Lamentably, both minds and paper are subject to damage and demise. Knowledge is therefore vulnerable as careers inevitably end. Expertise gained through a lifetime may be lost before it is put down in a publicly available format. (iii) Taxonomic research has often been undertaken by individuals working in isolation. Solo research has advantages, accommodating the style of some systematists, but collaborative efforts can achieve synergies that speed and enhance research products. Similarly, working in isolation leads to missed opportunities to build the contributions and knowledge of others into research products. Knowledge of biodiversity increases through dynamic interplay between data and hypotheses. For macro-organisms, it begins with collecting and acquiring data from specimens; for microbes, the first data may come from DNA sequences. In either case, data are obtained that enable initial assessments of similarity compared to known relatives. At some point, a researcher obtains the first hints that data may represent an unrecognized species. In the next phase, the hypothesis is tested with more data, further refined, and so on. At a point along this path, the investigator becomes confident that the data represent a distinct species, after which the species receives a name. More data might be gathered after that point, and hypotheses about species boundaries further refined. Traditionally, publication is at the end of this process. We argue that knowledge about biodiversity will increase most rapidly if data and hypotheses are made available throughout the process, as quantum contributions. A quantum contribution in systematics might be a photograph of a specimen; a new specimen record, including geographic information; a DNA sequence; a description of a diagnostic feature; or a hypothesis that a distinct species exists. Venues for publication of quantum contributions are Internet-based data management systems. Data providing evidence of new species will need to be as high quality, and as rigorously gathered and presented, as ever. Species hypotheses will need to be carefully advanced and tested. It is our contention that the ‘publish-as-you-go’ model we espouse will often lead to species hypotheses that are supported by more extensive data than is typical, and will have been vetted more efficiently at all steps in the process by the community. One key advantage of quantum contributions is that they are easier for systematists to produce than are full revisionary publications and, thus, knowledge about biodiversity will enter the public sphere without long delays. Many systematists know of dozens of undescribed species that they plan to describe someday. If easy, comprehensive systems for quantum contributions are created, then spending a few minutes posting pictures or DNA sequences will make a positive and significant contribution to biodiversity knowledge, sometimes years before the systematist completes a formal species description. We are also convinced that the quantum contribution model is essential for safe-guarding products of systematic research. Doing so protects data from loss owing to any number of circumstances, from demise of one's personal computer to demise of oneself. Projects in progress are far safer from total loss before completion if they are web-based than if they are only in one's desktop computer, notebooks, filing cabinets and mind. We note that scientists who work on microbial diversity provide a proof of concept of this approach. This community focuses on sequencing and annotating organismal genes and genomes, and has championed and embraced the release of data before formal publication [8Bentley D.R. Genomic sequence information should be released immediately and freely in the public domain.Science. 1996; 274: 533-534Crossref PubMed Scopus (48) Google Scholar, 9The Wellcome Trust Sharing Data from Large-scale Biological Research Projects: A System of Tripartite Responsibility. The Wellcome Trust, 2003Google Scholar]. Rapid, open access to these data has allowed the attendant repositories (e.g. GenBank) to increase rapidly in size and complexity. Such repositories have spawned tools for managing, visualizing and analyzing sequence data, and this reciprocal development has greatly expanded knowledge of microbial diversity (Box 2). Similar tools geared for systematists working on other groups of organism are becoming available (e.g. the Barcode of Life Database; Box 2). Although current tools have limitations (Box 2), we argue that the rapid release of data will be similarly beneficial to documenting the ‘genome’ of Earth; that is, the organismal diversity of the world.Box 2Current informatics endeavors for biodiversity compilation: existing pieces of the quantum contributions puzzleSocial networking platforms for sharing biodiversity knowledge and resources for coordinating new data into existing phylogenetic contexts are rapidly being developed. Missing is the ability to integrate these different kinds of platform in ways that maximally support discovery within and across platforms, and that seamlessly link taxonomy, phylogenetic trees and the data objects that adorn the branches of these trees, and that have data-entry, visualization and presentation tools elegant enough to attract most systematists. Here, we highlight some current efforts that individually provide great utility.Tools for collaboratively generating biodiversity contentScratchpadsScratchpads (http://scratchpads.eu/) provide a social networking platform to create, share and manage taxonomic data online. Especially valuable is automatic association of data uploaded to Scratchpads with Encyclopedia of Life (EOL) taxonomies. This is powerful for providing summary content of known taxa and beginning to associate existing, but not new, data into taxonomic frameworks.Life DesksLife Desks (http://www.lifedesks.org/) are similar to Scratchpads and seamlessly integrate with EOL taxon pages that include aggregated content from across the web. LifeDesk and EOL taxon pages include undescribed species and thus provide a mechanism to locate provisional taxa.CATECATE (Creating a Taxonomic e-Science, http://www.cate-project.org/) is focused on developing unitary taxonomies overseen by the community, along with development of modern, web-based treatments of taxa.WikispeciesWikispecies (http://species.wikimedia.org/wiki/Main_Page) is a free species directory, open to creation and editing by the public. It does not integrate with other databases, nor does it include automatic data-quality annotations.Tools for phylogenetic ordination and taxonomic identificationRibosomal Database ProjectThe Ribosomal Database Project (http://rdp.cme.msu.edu/) includes classifier and tree-building tools that allow new sequences to be ordinated in context with existing sequences, but lacks community workbenches to further utilize these results.BOLDBOLD (http://www.boldsystems.org), the Barcode of Life Database, provides identification of unknown sequences and placement within trees but lacks functionality to utilize this output fully.Metagenomics databases and toolsMetagenomics databases and tools, such as Camera, MG-RAST and IMG/M, are proliferating and provide both phylogenetic analysis tools and binning methods to make first-cut taxonomic matches. However, these are not specifically geared toward doing biodiversity discovery and documentation.Workflows for speeding documentation and publication processesNovel approaches for hastening movement from provisional assessment of new units of biodiversity to fully documented units are also being developed. In particular, automated workflows, such as those documented by Blagoderov et al. [25Blagoderov V. et al.Streamlining taxonomic publication: a working example with Scratchpads and ZooKeys.ZooKeys. 2010; 50: 17-28PubMed Google Scholar], extend Scratchpads projects with name registration in ZooBank and publication in ZooKeys. Social networking platforms for sharing biodiversity knowledge and resources for coordinating new data into existing phylogenetic contexts are rapidly being developed. Missing is the ability to integrate these different kinds of platform in ways that maximally support discovery within and across platforms, and that seamlessly link taxonomy, phylogenetic trees and the data objects that adorn the branches of these trees, and that have data-entry, visualization and presentation tools elegant enough to attract most systematists. Here, we highlight some current efforts that individually provide great utility. Scratchpads (http://scratchpads.eu/) provide a social networking platform to create, share and manage taxonomic data online. Especially valuable is automatic association of data uploaded to Scratchpads with Encyclopedia of Life (EOL) taxonomies. This is powerful for providing summary content of known taxa and beginning to associate existing, but not new, data into taxonomic frameworks. Life Desks (http://www.lifedesks.org/) are similar to Scratchpads and seamlessly integrate with EOL taxon pages that include aggregated content from across the web. LifeDesk and EOL taxon pages include undescribed species and thus provide a mechanism to locate provisional taxa. CATE (Creating a Taxonomic e-Science, http://www.cate-project.org/) is focused on developing unitary taxonomies overseen by the community, along with development of modern, web-based treatments of taxa. Wikispecies (http://species.wikimedia.org/wiki/Main_Page) is a free species directory, open to creation and editing by the public. It does not integrate with other databases, nor does it include automatic data-quality annotations. The Ribosomal Database Project (http://rdp.cme.msu.edu/) includes classifier and tree-building tools that allow new sequences to be ordinated in context with existing sequences, but lacks community workbenches to further utilize these results. BOLD (http://www.boldsystems.org), the Barcode of Life Database, provides identification of unknown sequences and placement within trees but lacks functionality to utilize this output fully. Metagenomics databases and tools, such as Camera, MG-RAST and IMG/M, are proliferating and provide both phylogenetic analysis tools and binning methods to make first-cut taxonomic matches. However, these are not specifically geared toward doing biodiversity discovery and documentation. Novel approaches for hastening movement from provisional assessment of new units of biodiversity to fully documented units are also being developed. In particular, automated workflows, such as those documented by Blagoderov et al. [25Blagoderov V. et al.Streamlining taxonomic publication: a working example with Scratchpads and ZooKeys.ZooKeys. 2010; 50: 17-28PubMed Google Scholar], extend Scratchpads projects with name registration in ZooBank and publication in ZooKeys. In the fields of astronomy and physics, the Sloan Digital Sky Survey (SDSS) [10York D.G. et al.The Sloan Digital Sky Survey: Technical summary.Astronomical J. 2000; 120: 1579-1587Crossref Scopus (7652) Google Scholar] and ArXiv projects represent two remarkably successful experiments in providing immediate, freely available resources. These transformational successes, which have allowed new kinds of science (e.g. citizen science utilizing SDSS data [11Raddick M.J. Szalay A.S. The universe online.Science. 2010; 329: 1028-1029Crossref PubMed Scopus (11) Google Scholar]), are often discussed as models, but remain unreplicated in biodiversity sciences. We agree that the future of publication of taxonomic results lies in the Internet, as cogently argued by Godfray et al. [12Godfray H.C.J. et al.The Web and the structure of taxonomy.Syst. Biol. 2007; 56: 943-955Crossref PubMed Scopus (59) Google Scholar]. We also agree with Mietchen et al. [13Mietchen D. et al.Wikis in scholarly publishing.Inf. Serv. Use. 2011; 31: 53-59Google Scholar] that Internet resources (e.g. wikis) can improve works after species have been formally named. We are promoting a more pervasive change: early Internet publication of quantum contributions about undiscovered or incompletely known organisms (and not just finished taxonomic works or modifications of these) has the potential to speed biodiversity discovery and documentation as it transforms our field. Some digital tools already available can be components of a workflow via quantum contributions (Box 2). These tools provide ways to compile biodiversity information collaboratively and efficiently, while also allowing experts to share knowledge via annotations (e.g. expert opinions). Such digital environments will support the ‘publish-as-you-go’ research approach that we argue is vital to hastening biodiversity discovery and documentation. With further development and better, more open linkages across platforms, web-based tools such as these will allow individuals and to new data in quickly the current of knowledge and resources to species documentation. The by quantum contributions will speed species discovery and documentation. systematist an of an might it as similar to has along a and then collect for DNA sequencing and share results via A systematist might note similar from and records. At that point, these systematists might and that there is enough evidence to an name for the species. The workflow that we will also contributions to the process by and With access to information about a new species, a might the that organisms of this species have an association with species. A might to an and that has collected a species. An who or similar species might a process that is already in for a systematist might take years to a and might not be that systematist has collected might not have the resources to that to the or might not have access to of by the the ‘publish-as-you-go’ approach will increase the data species hypotheses while the between the first hints of a new species and public of Data that are to biodiversity databases be in the context of current One mechanism be to data that may new species to an Internet-based of the of that existing knowledge of organismal diversity and and taxon through as a for this obtained data be to this along with existing be it DNA sequences, of or of evidence a new species, that species an name and on the at which all data about the species be The of the for a species in the along with be by creation of a formal taxonomic name and documentation. The be a of all This workflow is in A for systematics in a quantum knowledge model is of and taxonomic These are not only with hypotheses about and species boundaries that systematists but also of that are to Traditionally, species are not in they have formal and species are named only after they are in publishing (Box the steps in the current documentation process. In the of publishing quantum the first data often be published before any researcher has enough to a formal name. Data may be available before the researcher they represent a new species. information about species be publicly while the new species has at most an name. taxonomic will thus be the (i) a species hypothesis or a that a species that of a of and (ii) an name or taxon with the to taxonomic The Press, Google and (iii) a formal taxonomic name In web-based pieces of be to including a or to and the to taxonomic The Press, Google Scholar] have argued for as taxon to that they are and an be as components of both and formal Such and of data provide to and enable linkages between them and other data. This that creation of a species and formal are distinct in documenting a species, is in to taxonomic of formal only with publication of species hypotheses For some publishing data on species before they are formally named is already the In for new species are often discovered via and before they formal (e.g. and et al.The of Scholar] formally named years after it described as of in a J. Google More and new species are being discovered and DNA sequences published before they are described formally (e.g. in et 2000; Scopus Google Scholar], et species in using and 2005; PubMed Scopus Google Scholar], organisms D. an approach the diversity of organisms on 2005; PubMed Scopus Google Scholar], et of the PubMed Scopus Google Scholar] and et of an 2005; PubMed Scopus Google Scholar, et of a from PubMed Scopus Google the for taxa by DNA sequences in that not have formal species Some represent described species that have not been so others represent new species by but not formally and the further research to their These taxa in are on but early of the of a taxonomic their of formal will that a formal taxonomic name is should documentation it is as as in traditional species This will be by community Some might as of DNA sequences and other data more some traditional data has than on to species. the approach we also increase the pace of discovery by documentation. We have described a process to speed and systematic work that publishing data and hypotheses as they are all systematists will the fully open approach The approach may from to as a of the of the community that works on that the of the and the of available data. Some may more only sharing data We that progress will be most rapid in those for which the community a model of quantum contributions.