Sometimes the most foundational resources only get their loudest recognition when they vanish. Consider the outrage if, suddenly, there was no more BLAST server, no more Sequence Read Archive, no more Pubmed. If you’re a scientist, it isn’t hard to imagine how damaging it would be if some of these services disappeared overnight. Within one’s own field there are probably some more niche online services which, too, everyone quietly relies upon. This is the fate that is facing a database used by multitudes of parasite and vector researchers, who now are fighting to save a crucial pillar of their research.
VEuPathDB (roughly the Vector and Eukaryotic Pathogen Database) was a searchable resource of collated information concerning numerous disease vector species (like mosquitoes), infectious fungal species, and parasites (like malaria). It was formed in 2019 from a merger of the separate incarnations of some of these databases; however, the legacy websites that precedes the existing service are themselves much older, such as the malaria-focussed PlasmoDB (founded in 2001). Hence, the platform is over 20 years old, constantly updating and upgrading with time, having collected 1775 ‘datasets’ by 2022. The unified product included detail on Microsporidia, Cryptosporidium, Giardia, Plasmodium, Toxoplasma, Trichinella, fungi, vectors, amoeba, host interactions, and even genetic orthogony (the preservation of gene locations between different species). The consistently high usage of the tool attested to its impact on research: “In 2023, it averaged 46,000 unique users and more than 11 million page hits per month from researchers in more than 150 countries.”
Like the online BLAST server, Sequence Read Archive, and Pubmed, VEuPathDB is one of many resources funded by the National Institute of Health (NIH), the US-based scientific powerhouse. Unlike BLAST, VEuPathDB will no longer receive NIH support. As a consequence, it has shut down. The decision to no longer fund has been an unwelcome surprise to the field of infectious disease, with outpourings of support from across the user base hoping to convince the funding board to change its mind. Many big names have spoken up about the damage that could be caused by losing VEuPathDB, including Leann Tilley, Boris Striepen, Elizabeth Winzeler, Abdoulaye Djimde, David Fidock, Dyann Wirth, and Dionicia Gamboa. The outcry encompasses voices from all over the world, with researchers from high- and low-resource nations united in hope that a solution will be found to save the database.
I did my PhD on malaria, including work on different species that infect humans, apes, and even mice, so I can testify first-hand that VEuPathDB (and specifically PlasmoDB) provided an excellent tool for research across organisms. If you had a question to ask of the genome (such as what type of protein a gene might encode, what is a gene’s DNA sequence, when is this protein being produced, in which other species a gene has orthologous cousins, etc) it was easy to answer. Queries of groups of genes could be made and then combined to obtain subsets of genes, and the queries themselves could be saved so that you could ask them again when the data updated. These updates occurred frequently, and were a good way to see what the malaria world was publishing. None of these actions were unique to PlasmoDB – that is, you could accomplish them through other means. But PlasmoDB made them simple to use, and put them all into one place. The team behind the site were active in the community, regularly hosting training sessions and attending conferences to advertise features. To disappear so suddenly represents an immediate threat to students writing theses, and academics writing papers, who may have been relying on the software staying available up until their deadline. They, like me, may have taken the continued existence of VEuPathDB for granted.
The intention of the NIH appears to be pivoting towards a new resource, the Bioinformatics Resource Analytics Centre (BRC-Analytics), which will include the pre-existing VEuPathDB data integrated with tools from UCSC Genome Browser, HyPhy (for analysing evolutionary relationships), and Galaxy (a popular interface for using bioinformatic software without requiring any coding). Many in the parasitology community have yet to be convinced by this change of direction. In fact, BRC-Analytics was likely intended as a complementary resource to work alongside VEuPathDB instead of replacing it, suggesting that the funding committee unilaterally decided that the former should take its place. Granted, many of the methods would likely be overlapping between both the softwares (since VEuPathDB also offered levels of genome browsing, evolution tracking, and Galaxy integration), so it may make sense to merge instead of maintaining two separate projects. However, the outcome of the decision has been jarring instead of prescient. There has been no proposed merge or carry-over process, so VEuPathDB has gone dark and the fate of their 40 staff is in limbo. BRC-Analytics is a nascent project that is around six months from being up-and-running, so in the interim there is nothing.
Prof David Roos, formerly faculty at UPenn, was one of the founders of VEuPathDB, and he recently stopped leading a research lab in order to focus on managing the service full-time. He has not held back his exasperation for the situation, likening it to the US outright banning gasoline to force a transition to electric cars. Moving to sleeker, and perhaps more sustainable, databases may be a good long-term plan, but this can’t be at the expense of having any useful resource in the present. Prof George Christophides, one of the founders of the database VectorBase which became part of VEuPathDB, stated that “the decision to fund alternatives without consulting the very community that built and relied on VectorBase speaks to a disconnect between funders and the researchers they aim to support”. He also echoes a concern that the suggested replacement will be aimed at specialised bioinformaticians instead of a more general academic user base. A comment in The Lancet Microbe supports these sentiments and highlights the major/neglected tropical diseases that VEuPathDB covered. Many researchers focussing on these pathogens work in low-income nations, where the parasites cause the most suffering, and their work could be the most severely affected by the loss of an open resource that required minimal training to use. The British Society of Parasitology, too, has drawn attention to the situation.
As the title suggests, there is a light at the end of the tunnel for VEuPathDB. The concern from academics globally did not go unnoticed by other funding bodies, some of whom have stepped in to discuss supporting the platform in the immediate-term: The Foundation for the Advancement of Science & Medicine, Open Philanthropy, the Chan Zuckerberg Initiative, and the University of Pennsylvania and the University of Georgia. The closing message on the shut-up shop of VEuPathDB’s current website (and the website of all constituent databases) ends on this glimmer of hope, tempered by the acknowledgement that a long-term plan is needed to maintain the resource into the next twenty years or more.
As ever close to their community, the database team are also asking concerned users to fill in a survey about the role of the software in their work, and their opinions on its future. If you’ve ever used a VEuPathDB service, it might be worth demonstrating your appreciation now, before the effort is gone for good.
Comments