Assessment of Progress - Research Progress on Environmental Health and Safety Aspects of Engineered Nanomaterials - page 25

an important opportunity for international collaboration to develop an infrastructure that can serve both
communities.
Although those new programs are promising, progress in developing elements of an informatics
infrastructure has been less encouraging. The foregoing examples show the need for libraries of
nanomaterials; for improved reporting on nanomaterial production processes and sample-preparation
techniques; for new methods for characterizing, tracking, and monitoring nanomaterials and their
transformations; for methods for quantifying the effects of nanomaterials; and for systems for sharing
research results and the development of predictive models for nanomaterial behaviors. Core systems,
services, and applications are not yet available or have been insufficiently adopted, and this gap impedes
research and the translation of research findings into products. For example, a harmonized nomenclature
system that facilitates and informs nanomaterial classification and development does not exist; data and
metadata standards are not established; reproducibility of methods (ruggedness testing) has not been
established; and the sensitivity data are not shared and therefore cannot be used to improve the
reproducibility of methods or to inform error propagation in risk analyses. The same general limitations
are present for model development: furnishing accurate nanomaterial and nanoproduct structural models
on the appropriate scales; developing and validating the models and their sensitivity to input parameters,
computer programs, the choice of run-time parameters, computer architectures, and compilers at the
relevant dimensions and time scales; and accessing and validating models for the physical, chemical, and
biologic systems of interest, also at the appropriate dimensions and time scales. In that regard, NanoHUB
constitutes a substantial and important start, providing a stable code for different users and assuming the
burden of hosting the code; providing computers, storage, and user services; archiving and sharing data,
metadata, and information about results; and comparison with related model results.
Finally, there is an overarching need for informatics to augment collaboration and accelerate
research and translation by facilitating access to data. Examples of the need for informatics include the
accelerated adoption of models through NanoHUB and the increased amount of interlaboratory testing of
methods by various organizations (NIEHS 2012; ILSI 2013a). There are abundant examples of data that
are not available through the publication process and that in many cases are not accessible on any
database—such as sensitivity data on methods and validation data on models—but there are several areas
of particular interest and activity. For example, high-throughput methods are increasingly used in
nanotechnology-EHS research, and applications from EPA-NSF funded centers (Thomas et al. 2011;
Mandrell et al. 2012) promise to generate large, correlated datasets obtained with standardized screening
methods. ISA-TAB-Nano 2 , a new standard for data exchange, is emerging; its harmonized data formats
incorporate high-throughput screening assays and methods for nanomaterial characterization. Metadata
capture will be possible through the NanoParticle Ontology (NPO) that builds on NIH's Enterprise
Vocabulary System. However, most important are the increasing informatics efforts (mentioned above)
that promise new support and substantially increased collaboration—the NKI, collaboration with the
MGI, and the other NNI signature initiatives, particularly the EU-US CoRs. Those developments
collectively signal heightened interest in increasing data quality throughout nanotechnology and
nanoscience and heightened activity in establishing a coherent infrastructure for increased collaborative
research among all the disciplines.
Additional data inputs are possible if databases are compiled from other studies. One potential
mechanism, as mentioned in the committee's first report, is NSF's requirement that all grant proposals
include a two-page plan for how data will be managed and shared publicly. However, modifications of
that requirement through creation of a data commons could allow the collection of all nanotoxicity data
from NSF-funded studies rather than siloed storage and retrieval sites established by each researcher.
2 This format is an extension of the Investigation-Study-Assay (ISA) Tabular formats used for genomics and
high-throughput screening (for example, MAGE-TAB) and adds a material file to permit transmission, linkage, and
provenance of data on the nanomaterial samples being studied. This publication represents an initial step to
providing one aspect of the needed infrastructure for sharing research data, and it is not yet clear how it will be
received by the research community.