March 2010

TAIRing at Research

Highlights from the Policy Blotter

The American Society for Biochemistry and Molecular Biology Policy Blotter blog posts regular news and commentary about current science policy issues. Below are some recent highlights. You can read them and other posts at

• A Strong Science Budget, but NIH Heading for a Cliff
 In his annual budget request to Congress, President Obama recommended strong funding increases for many scientific agencies that support the life sciences.

• NIH Needs $37 Billion in 2011
 On Jan. 28, the Federation of American Societies for Experimental Biology and its member societies, including ASBMB, recommended the National Institutes of Health budget be increased to $37 billion during 2011.

• Collins: Reinvigorate the Research Community
 Francis Collins, director of the National Institutes of Health, has given several interviews and authored a high-profile “policy forum” piece in the journal Science. In each case, he continued to encourage robust support for the NIH that he says will bring about great new societal benefits.

• What Makes DARPA So Special?  
On Jan. 7, at a meeting of the President’s Council of Advisors on Science and Technology, officials from the Defense Advanced Research Project Agency presented what they believed were the characteristics of this defense-oriented research agency that have allowed it to innovate and succeed over its decades-long history.


Molecular biologists around the world have come to rely on actively curated genome databases of model organisms. But the National Science Foundation has decided to end its support for The Arabidopsis Information Resource, known as TAIR, and Arabidopsis and plant biochemists risk losing a vital resource. If the NSF’s decision sets a precedent, Wormbase, Flybase and other databases may be similarly at risk.


“TAIR is where you go for plant genome science,” said Rebekah Rogers, a Harvard doctoral candidate and former plant molecular biologist.

Like many similar databases, TAIR provides a host of information related to the genetics of the model plant Arabidopsis thaliana. At, TAIR users can find Arabidopsis genome information ranging from the very basic to the very applied, said Eva Huala, director and principal investigator of TAIR.

Like a Web-based library, TAIR’s unique and integrated set of resources requires an active curatorial effort. The database and its 20 staff members have relied upon funding provided by the NSF for more than 10 years.

The genomic resources of TAIR have helped unlock the research potential of Arabidopsis for an entire community of researchers. TAIR has about 40,000 unique users each month from the Americas, Europe and Asia, and the average number of users each month has grown steadily since TAIR’s founding.

But those who utilize the database often are interested in more than just Arabidopsis. “It’s what everyone in plant biology uses, even in crop science,” Rogers said.

A Community Shrinks

Having awarded TAIR two, five-year grants, the NSF in May 2009 declined to fully renew TAIR’s funding. Instead, the NSF has granted TAIR an additional four years of steeply decreasing funding and encouraged the database to seek funding from other sources. By 2011, TAIR’s NSF funding will cease.

The potential collapse of TAIR’s funding threatens the field of plant genomics.

“The first people to go will be the computational biologists,” Huala said. As these researchers rely upon publically accessible data, they are unlikely to pursue plant research if the information is not readily available. If computational biologists leave plant genomics, plant biology may fall behind animal research, Huala said.

Other biologists also may be driven away from plant research. Because it provides graphical, easy-to-use interfaces, TAIR gives researchers access to genome-based data without requiring them to write computer programs, Rogers said.

Innovation vs. Infrastructure

Continuing to fund research infrastructure often runs counter to the NSF’s focus on funding innovative research. When a resource or program like TAIR ceases to be innovative, the NSF would like to use its limited budget in other places, Huala said.

Indeed, TAIR may have fallen victim to an emphasis on new innovations in sequencing technology.

“With the flood of genomic data, it may not be the best expenditure to put so many resources into a few species,” said Scott Roy, a postdoctoral fellow at Stanford University.  A computational biologist, Roy said model organisms may begin to occupy a smaller percentage of genome data that technological advances have made inexpensive to produce.   However, the direction of the field is still uncertain, Roy said.

But, although financial resources may limit their numbers, genomic databases have “a tremendous utility to inform closely related genomes,” Huala said.

Additionally, though new sequencing technology can produce staggering amounts of raw data, genome databases integrate sequence information with gene descriptions and relevant publications. Some databases also are repositories for unpublished data and minor comments that would not otherwise be available.

Without genome databases, “that kind of information would be lost,” Rogers said.

The Future of Databases

Like other National Institutes of Health-funded projects, many genomics databases are supported by grants that must be renewed every several years. While the NIH continues to support several databases, the grants for two major databases, Flybase and Saccharomyces Genome Database, are up for renewal in 2011.

As for TAIR, Huala has discussed the situation with officials at the NIH in hopes that they might fund the database. Although conversations are ongoing, the NIH seems “reluctant to take on another model,” Huala said.

For now, TAIR is exploring other funding sources, including corporate sponsorships. Huala said she believes requiring users or institutions to purchase subscriptions may drive away many academic researchers.

Kyle M. Brown ( is an ASBMB science policy fellow.

Do you use TAIR? Tell us how you feel about the funding situation in the comment section below.

found= true556