By Doug Macron
The National Institutes of Health last month announced the first funding opportunity under a three-part program designed to support the collection and analysis of "informative molecular activity and cellular feature signatures" in response to siRNAs, small molecules, and other perturbing agents in a variety of cell types.
The program, called the Library of Integrated Network-Based Cellular Signatures, or LINCS, was established by the NIH in early February based in part on the outcome of a 2008 workshop held to develop strategies for generating useful data from cell-based systems that had been disrupted systemically by perturbagens, and using that information to better understand biological networks in health and disease.
According to LINCS coordinator Alan Michelson, associate director for basic research at the National Heart, Lung, and Blood Institute, that meeting was prompted by a 2006 paper in Science in which Todd Golub and colleagues from the Broad Institute and Harvard University reported their so-called connectivity map, which "showed you could use gene-expression signatures in various drug-treated cells to reveal unanticipated relationships among different small molecules."
"With that [paper] in mind, we had a lot of interest in assessing what the state of the art is … in exploiting that particular strategy for looking at small-molecule effects and perhaps combining it with RNAi," which could, for example, provide insights into the targets of certain drugs, he said.
"With a small molecule, you usually don't know what your target is, but with RNAi, you know at least what you've intended [to be] the target with the particular nucleotide sequences you've designed," Michelson explained. "The potential to actually combine the small-molecule screening with RNAi screening … [might] help get at what [the] small-molecule targets might be, at least through the proxy of having common signatures."
But the overall goal of the LINCS program "is not to address any one biological question," the NIH said. Rather, it is "to develop a resource [that] can be used broadly by the biomedical research community to address a wide variety of biological questions. Potential applications of the LINCS knowledge base include … modeling of biological networks, elucidating how human genetic variants cause disease, classifying diseases by functional molecular criteria, developing novel molecular diagnostics … or identifying new therapeutic targets."
In fact, the NIH is purposely taking a sort of wait-and-see approach when it comes to applications for funding under the program. "After consulting with the community, we decided that it was probably not a good idea to be over-proscriptive with respect to specifying any aspect of this," Michelson said.
Broadly, the NIH views the program as a "three-way matrix" of an assay used to evaluate cellular response, perturbagens, and the cells themselves, he said.
When it comes to the assay, "it doesn't have to be just gene expression; it could be virtually any high-throughput assay that is going to be informative for a particular set of perturbagens in cells," he said. "For example, it could also be high-content, microscopic imaging of cellular responses with appropriate fluorescent-based reporters."
In terms of perturbagens, funded projects would "preferably" evaluate a mix of small molecules and RNAi agents, but "it's not clear what the optimum combination" would be, he noted. "That would need to be justified by the original applicant."
[ pagebreak ]
At the same time, "there are other potential perturbagens that could be employed," Michelson said. These could include "environmental conditions that might have particular physiological relevance, pH manipulations, or oxygen-tension manipulations in cell culture." And those could all be further explored in combination with siRNAs and small molecules.
And while the NIH is interested in gathering data from all cell types, it has specified that these be human cells, "simply because we want to see the direct relationship to human cellular responses, the relationship to disease signatures, and the potential to relate that more directly to … drug therapy," he noted.
Last month, the NIH released the first of three requests for applications that will be funded under the initial phase of the LINCS program. Entitled "Large-Scale Production of Perturbagen-Induced Cellular Signatures," it will fund "data-production efforts that will produce a usable initial resource, while addressing issues of optimization of methods, integration of data, and scalability," according to the NIH.
The agency has set aside $2.7 million for fiscal 2010 to fund two awards under the RFA. These funds will increase to $5 million a year in 2011 and 2012.
The funding is specifically designed to support two "pilot data-production centers," which will be charged with producing "a necessarily limited but coherent initial LINCS data resource that can be used by the general research community," the NIH said in the RFA.
"Because this … seeks the production of an initial LINCS resource having immediate utility, it is expected that applications will propose approaches that are already mature in terms of throughput, scalability, and biological informativeness," the RFA adds.
The agency plans to hold a conference call on March 31 from 1 to 2 pm Eastern to provide applicants with background information and answer questions about the preparation of applications for the RFA. Details about the call will be released at a later date.
The NIH expects to issue two additional LINCS RFAs later this year, one of which will support "a computational data integration/data-analysis tool-development approach," Michelson said. The other will cover "wet-lab technology development."
In the end, the NIH hopes that the LINCS program will "set a standard for how to go about doing this type of work" that can be used by other researchers who aren't funded by the RFAs, he noted.
"We'd also like to see a sort of living database accumulate," expanded with data from investigators within the LINCS program and those doing similar work in independently funded projects, both in the US and overseas, he said.
"And, of course, [we'd like] the data that are generated here be made available to lots of other users who are not part of the program," Michelson added. "This is meant to be a truly general community resource."