This article has been updated to state that the dollar amount of NIH's infrastructure investment is on a per year basis rather than the total investment for the entire program as previously reported.
As part of its efforts to strengthen biomedical research in the US, the National Institutes of Health is revisiting the informatics infrastructure it has in place for managing and sharing large biomedical datasets and has proposed a number of initiatives that are expected to be in place by 2015.
The recommendations were put together by the data and informatics working group of the NIH’s Advisory Committee to the Director, which was comprised of researchers from the NIH as well as Stanford, Princeton, Harvard, and Columbia Universities, the Broad Institute, and others.
The NIH decided to revisit its infrastructure because of the “explosion” of data generated by biomedical research efforts it supports and because of technology advances in genomics, imaging, and electronic health records, Lawrence Tabak, NIH's principal deputy director and co-chair of the working group, told BioInform this week.
“We’ve got so much more data that is available to us and it hasn’t gotten lost on us that perhaps that we are not fully exploiting these data,” he said.
In its report, the data and informatics group asked the NIH to promote data sharing through central and federal catalogs; support the development, implementation, evaluation, maintenance, and dissemination of informatics methods and applications; build capacity by providing training in relevant quantitative sciences; develop an NIH-wide data strategic plan; and provide the financial backing needed to support these activities.
The group then came up with implementation strategies for its recommendations that took into account the need for an infrastructure environment that supports optimal use of big data and recognized that “the ability to use the data that we are generating is a responsibility that [should be] shared across all institutes and centers at NIH not just a small subset,” Tabak told BioInform.
The working group came up with two strategies that will be led by NIH Advisory Data Councils. The first, which will benefit the scientific community at large, Tabak said, is the Big Data to Knowledge (BD2K) initiative, through which the NIH expects to extract maximum value from the biomedical data it generates.
Activities under this initiative will include developing and implementing improved data and software sharing policies; creating catalogs of existing research data; and engaging the community to develop data/metadata standards to facilitate broader use of biomedical data, he said.
The NIH will also create and disseminate analysis methods and software, provide training, and establish new centers of excellence, he said.
Other efforts include the InfrastructurePlus adaptive environment, which will explore high-performance computing, hosting, and storage approaches at NIH as well as support the modernization of the NIH network among other activities.
On this front, the data council will assess NIH’s current computing capabilities and its current and projected computational and storage needs, and then make recommendations about how best to implement a shared computational environment as well as establish cloud infrastructure among other activities.
The NIH plans to begin implementing its data and informatics recommendations next year, Tabak said.
In 2013, the working group’s activities will include a series of workshops that will address the different aspects of its initiative as well as announce some funding opportunities, he said.
He also said that the group estimates that it will invest on average around $125M yearly over the life of the program “with the majority of the resources planed for extramural activities.
We obviously must ramp up certain aspects and will not start at the average from the inception," he said
Besides data and informatics, other NIH initiatives include a new program called Building Infrastructure Leading to Diversity that is intended to provide mentored research experiences for undergraduate students at participating schools; financial support to pursue biomedical research careers; and faculty support for training mentors.
The institute also intends to establish a National Research Mentoring Network; set up an NIH steering committee working group to ensure that diversity remains a core consideration; offer a grants program that will enhance training of graduate students and postdoctoral researchers; and encourage the adoption of individual development plans for all trainees among other activities.