Data Science & Informatics

SUMMARY Brain-mapping initiatives are acquiring increasingly large and comprehensive neuroimaging and multiomic— e.g. genomic and transcriptomic—datasets. Existing analyses of such data in human neuroscience tend to search for links between cognition, behavior or disease on the one hand, and properties of genomes, transcrip- tomes or brain morphology and connectivity on the other. Such valuable analyses have steadily advanced our knowledge of human brain function.

PROJECT SUMMARY/ABSTRACT The growing availability of large functional magnetic resonance imaging (fMRI) datasets has enabled new investigations into functional systems of the human brain. A challenge – but also opportunity – of fMRI arises from the fact that BOLD signal stems from multiple intertwined neural and physiological sources. One major contributor to fMRI signals arises from slow (

This two-year project will advance, integrate, document, and promote the use of the Hierarchical Event Descriptor (HED) system to describe events in human neuroimaging and behavioral data from research experiments and other sources in sufficient detail to support comparative analysis of human brain dynamics across studies. Relating the recorded data dynamics to temporally- specifiable changes in subject experience, action, and cognition is a major goal (and challenge) for both neuroimaging and biomechanical imaging.

Project Summary/Abstract Reuse of existing neuroscience data relies, in part, on our ability to understand the experimental design and study data. Historically, a description of the experiment is provided in textual documents, which are often difficult to search, lack the details necessary for data reuse, and are hampered by differences in terminologies across related fields of neuroscience.

Abstract The goal of this project is to develop an archival data format and a formal task specification language to serve as standards for describing behavioral experiments. Because different laboratories use different behavioral systems, hardware, and software, it has been difficult to communicate behavioral task design, share data, or reproduce experiments.

Neuroscientific data contain information from an incredible diversity of species, are generated by a plethora of devices, and encapsulate the results of scientific thinking and decision making. Most of this generated data remains confined within laboratories and is not accessible to the broader scientific community. The research projects awarded under the Brain Initiative are generating a diverse collection of data that can transform and accelerate the pace of discovery. These datasets are large--ranging in size from GBs to PBs-- and represent diverse data types and assorted metadata.

To take advantage of recent and ongoing advances in intensive and large-scale computational methods, and to preserve the scientific data created by publicly funded research projects, data archives must be created as well as standards for specifying, identifying, and annotating deposited data. The value of and interest in such archives among researchers can be greatly increased by adding to them an active computational capability and framework of analysis and search tools that support further analysis as well as larger scale meta-analysis and large scale data mining.

Project Summary (maximum 30 lines) The BRAIN initiative seeks to develop and apply technologies in order to understand of how brain cells interact in both time and space to give rise to brain function.

Project Summary Population neuroscience requires investigation of brain-behavioral relationships within epidemiological samples studied in longitudinal designs, including a large number of assessment modalities and with subjects enrolled at many data collection sites to achieve a large sample and broad representation of the population.

PROJECT SUMMARY Overview: We will extend and develop implementations of foundational methods for analyzing populations of attributed connectomes. Our toolbox will enable brain scientists to (1) infer latent structure from individual connectomes, (2) identify meaningful clusters among populations of connectomes, and (3) detect relationships between connectomes and multivariate phenotypes. The methods we develop and extend will naturally overcome the challenges inherent in connectomics: high-dimensional non-Euclidean data with multi-level nonlinear interactions.