Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Jan 1.
Published in final edited form as: Neuroimage. 2015 May 1;124(0 0):1149–1154. doi: 10.1016/j.neuroimage.2015.04.057

The Pediatric Imaging, Neurocognition, and Genetics (PING) Data Repository

Terry L Jernigan 1,2,3, Timothy T Brown 4,5, Donald J Hagler Jr 4,6, Natacha Akshoomoff 1,3, Hauke Bartsch 4, Erik Newman 1,3, Wesley K Thompson 3,7, Cinnamon S Bloss 8, Sarah S Murray 9, Nicholas Schork 10, David N Kennedy 11, Joshua M Kuperman 4,6, Connor McCabe 12, Yoonho Chung 13, Ondrej Libiger 8, Melanie Maddox 1, B J Casey 14, Linda Chang 15, Thomas M Ernst 15, Jean A Frazier 11, Jeffrey R Gruen 16, Elizabeth R Sowell 17, Tal Kenet 18, Walter E Kaufmann 19, Stewart Mostofsky 20, David G Amaral 21, Anders M Dale 2,4,5,6, for the Pediatric Imaging, Neurocognition, and Genetics Study
PMCID: PMC4628902  NIHMSID: NIHMS691152  PMID: 25937488

Abstract

The main objective of the multi-site Pediatric Imaging, Neurocognition, and Genetics (PING) study was to create a large repository of standardized measurements of behavioral and imaging phenotypes accompanied by whole genome genotyping acquired from typically-developing children varying widely in age (3 to 20 years). This cross-sectional study produced sharable data from 1493 children, and these data have been described in several publications focusing on brain and cognitive development. Researchers may gain access to these data by applying for an account on the PING Portal and filing a Data Use Agreement. Here we describe the recruiting and screening of the children and give a brief overview of the assessments performed, the imaging methods applied, the genetic data produced, and the numbers of cases for whom different data types are available. We also cite sources of more detailed information about the methods and data. Finally we describe the procedures for accessing the data and for using the PING data exploration portal.

Introduction

Here we describe the data generated in the Pediatric Imaging, Neurocognition, and Genetics (PING) study, and the data repository that is now accessible to investigators through the PING Portal (http://pingstudy.ucsd.edu). The repository contains aggregated data collected at 10 sites in the United States. Multimodal neuroimaging data, developmental histories, behavioral and cognitive assessments, and/or genome-wide genotypes are now available for 1493 children and adolescents, aged 3 to 21 years. The goal of PING was to address the imbalance in existing imaging genomics data resources between those containing data collected from adult and elderly participants and the very limited data available from pediatric and adolescent participants. A wide pediatric age range for participants was included to allow investigators to search for interactions between age and genotype (i.e., to discover gene associations with developmental phenotypes). The PING infrastructure was designed specifically to address the challenges of large, multi-site studies involving multimodal imaging and assessment of behavioral phenotypes in a developmental context, and to facilitate the exploration as well as the dissemination of the sharable data. While PING was a cross-sectional study, all features of the PING infrastructure were designed to be extensible to longitudinal designs and the infrastructure currently supports ongoing longitudinal studies that have followed PING. Below we describe briefly the PING study and cohort; procedures for facilitating, standardizing, and optimizing data acquisition; procedures for processing the imaging and genomics data; and the infrastructure for sharing and exploration of accumulated data.

The PING Cohort

Participants were recruited through local postings and outreach activities conducted in the greater metropolitan areas of Baltimore, Boston, Honolulu, Los Angeles, New Haven, New York, Sacramento, and San Diego. Children, adolescents, and young adults were screened as eligible for PING if they were between the ages of 3 and 20 years and fluent in English (some older 20-year-olds turned 21 by the completion of data collection). Exclusion criteria included: a) neurological disorders; b) history of head trauma; c) preterm birth (less than 36 weeks); d) diagnosis of an autism spectrum disorder, bipolar disorder, schizophrenia, or mental retardation; e) pregnancy; and f) daily illicit drug use by the mother for more than one trimester. Individuals with contraindications for MRI (such as dental braces, metallic or electronic implants, or claustrophobia) were also excluded from participating. Individuals with identified or suspected learning disability or ADHD were not excluded since these syndromes are fairly common in pediatric populations. Over 1700 participants were enrolled in PING at one of the 10 sites, however, only data from participants in whom acceptable data were obtained for at least two data types (i.e., demographic/developmental, behavioral, genomic, imaging) are included in the PING repository. This sample consists of 1,493 participants (780 males) for whom acceptable imaging, genomics, and/or cognition assessments were obtained; acceptable imaging data were acquired for 1,239 of these participants (645 males); acceptable cognition data for 1,453 participants (752 males), and acceptable genotyping is available for 1391 (719 males). Similar proportions of males and females participated across the entire age range. Written parental informed consent was obtained for all PING subjects below the age of 18, and child assent was also obtained for all participants between the ages of 7 and 17. Written informed consent was obtained directly from all participants aged 18 years or older. For more information about the PING cohort, see Brown et al., 20121 and Akshoomoff et al., 20142.

Participant Assessments

Neuromedical History

The PING Study Demographics and Child Health History Questionnaire was completed by parents or guardians of the minor participants, and participants aged 18 and over were given a self-report version of this questionnaire. The questionnaire assessed basic medical, developmental, and behavioral history, as well as family history of medical and neuropsychiatric disorders, including substance abuse. The measures from this questionnaire are identified in the PING Portal ontology and defined in the data dictionary with the prefix “FDH_”.

NIH Toolbox Cognition Battery Measures

Cognitive assessments for the PING project were conducted using the NIH Toolbox Cognition Battery (NTCB). The NTCB was designed to tap key functions (executive function, attention, episodic memory, working memory, language, and processing speed) across the lifespan (ages 3 to 85 years). This computerized approach provides an economical method for assessing a wide range of cognitive abilities, which is appealing for large-scale studies. For pediatric studies, this also has the advantage of providing the same set of measures for use with young children, older children, and adolescents in an appealing format that provides automated data collection, storage, and scoring3,4. A detailed description of the NTCB results in the PING study is provided in Akshoomoff et al., 20142. More detailed information on the NIH Toolbox for Cognition is available at http://www.nihtoolbox.org/. The eight NTCB subtests for which data are available are: Dimensional Change Card Sort Test, Flanker Inhibitory Control and Attention Test, Picture Sequence Memory Test, Pattern Comparison Processing Speed Test, Oral Reading Recognition Test, List Sorting Working Memory Test, and Picture Vocabulary Test. The measures obtained with the NTCB are identified in the PING Portal ontology and data dictionary with the prefix “TBX_”.

Social-Emotional and Substance Use Assessments

In a subset of participants, a limited number of additional assessments of social and emotional functions and substance exposure were acquired through the PhenX Rising project. More information about the PhenX Toolkit and PhenX Rising is available at https://www.phenxtoolkit.org and about the data collected in PING in association with PhenX Rising in McCarty et al., 20143. PhenX assessments were obtained on only a subset of participants because PING was already underway when the PhenX Rising study began. Also, many of the assessments are age-specific. Variables associated with PhenX assessments are identified in the PING ontology and data dictionary with the prefix “PHX_”.

Multimodal Image Acquisition and Quality Control

In PING, the imaging, quality control, and analysis protocols were developed specifically to meet the challenges associated with multisite, multimodal imaging of children. These procedures were also designed to ensure that acquisition and preprocessing methods were compatible with, and facilitated, the use of post-processing methods developed by leading imaging groups throughout the neuroimaging community as well as those applied by consortium investigators.

Image Acquisition and Preprocessing

The PING imaging protocol takes advantage of key technologies developed for the consortium and builds on earlier methods development performed as part of the Biomedical Informatics Research Network (BIRN4) and the Alzheimer's Disease Neuroimaging Initiative (ADNI5). Specifically, a standard PING scan session included: 1) a 3D T1-weighted inversion prepared RF-spoiled gradient echo scan using prospective motion correction (PROMO), for cortical and subcortical segmentation; 2) a 3D T2-weighted variable flip angle fast spin echo scan, also using PROMO, for detection and quantification of white matter lesions and segmentation of CSF; 3) a high angular resolution diffusion imaging (HARDI) scan, with integrated B0 distortion correction (DISCO), for segmentation of white matter tracts and measurement of diffusion parameters; and 4) a resting state blood oxygenation level-dependent (BOLD) fMRI scan, with integrated distortion correction. Pulse sequence parameters used across (3T) scanner manufacturers (GE, Siemens, and Phillips) and models were optimized for equivalence in contrast properties and consistency in image-derived quantitative measures.

Gradient nonlinearity correction (3D GradWarp)

Nonlinearity of the gradient fields used for spatial encoding in MRI is one of the most prominent sources of spatial distortion in MRI scans6,7. Through the involvement of PING neuroimaging investigators in the Biomedical Informatics Research Network (BIRN) and the Alzheimer's Disease Neuroimaging Initiative (ADNI), the group has led the development of a fully automated procedure to correct for gradient field nonlinearities using displacement maps computed based on scanner-specific specifications provided by the MRI scanner manufacturers. The correction software developed in the UCSD MultiModal Imaging Laboratory (MMIL) was adopted as part of the routine preprocessing routine for all scans acquired as part of ADNI, and has been shown to significantly improve the accuracy of longitudinal change estimates based on serial MRI scans8.

Motion correction (PROMO)

An important recent advance in MRI acquisition technology is the development of real-time, or prospective, motion correction. The PROMO approach9, first applied widely in PING, utilizes three orthogonal spiral navigators together with a recursive image-based estimation strategy based on the extended Kalman filter (EKF) for motion measurement. The spiral k-space trajectory allows image-domain reconstruction prior to motion estimation, which when combined with the flexible EKF framework, allows for efficient image-based tracking within an a priori region of interest. Significant reduction of motion-related image degradation in pediatric imaging is possible with this method10,11.

EPI B0 distortion correction (DISCO)

Single-shot echo planar imaging (EPI) is an efficient MRI acquisition scheme for producing fast, high-definition images for diffusion weighted imaging and fMRI. However, EPI suffers from severe spatial distortions and intensity variations due to susceptibility-induced B0 field inhomogeneity. Anatomically accurate, undistorted images are essential for integrating the HARDI and fMRI images with anatomical (T1-weighted) images in PING, i.e., for achieving accurate spatial registration of the information from different modalities. Since the B0 distortion pattern depends on the exact position of the subject in the scanner, which may vary across scan sessions, correcting for such distortions is also essential for obtaining accurate estimates of change based on longitudinal MRI scans. Our group has developed a fast, robust, and accurate procedure for removing such spatial and intensity distortions from the EPI images obtained for HARDI and fMRI12. The method involves acquisition of brief scans with opposite phase encoding polarities (resulting in opposite spatial distortion patterns) and subsequent alignment of the resulting images using a fast nonlinear registration procedure. The DISCO method, which requires minimal additional scan time, provides superior accuracy and better cross-modality registration relative to the more commonly used, and more time consuming, field mapping approach.

Multimodal Image Analysis

Morphometric analysis of structural MRI data was performed using a specialized processing stream developed for PING that is based on FreeSurfer, with additional corrections and analyses developed at UCSD MMIL.

Structural MRI Preprocessing

As described above, distortions caused by nonlinearity of the spatial encoding gradient fields were corrected with predefined, scanner specific, nonlinear transformations, provided by MRI scanner manufacturers13. Non-uniformity of signal intensity was reduced using the nonparametric non-uniform intensity normalization (N3) method14. Images were rigidly registered and resampled into alignment with an atlas brain with 1 mm isotropic voxels, facilitating standardized viewing and analysis of brain structure. If multiple, good quality scans were obtained for a participant, they were registered to each other and averaged.

Morphometric Analysis

FreeSurfer encompasses tools for cortical surface reconstruction, subcortical segmentation, cortical parcellation, and estimation of various measures of brain morphometry using routinely acquired T1-weighted MRI volumes15–27. Important extensions made at MMIL include maps of relative cortical surface area changes28,29 and genetically informed cortical parcellations28,30,31. Cortical surface reconstruction involves skull-stripping26, non-uniformity correction32, white matter segmentation, initial mesh creation15, correction of topological defects20,27, and generation of optimal white and pial surfaces15,16,19. Subcortical structures were labeled using an automated, atlas-based, volumetric segmentation procedure21; volumes in mm3 and average T1-weighted intensity (T1w) were calculated for each structure. Labels for cortical gray matter and underlying white matter voxels were assigned based on surface-based nonlinear registration to atlas based on gyral and sulcal patterns23 and Bayesian classification rules17,18,24,25. White matter voxels adjacent to the cortical parcels were also labeled25. Fuzzy-cluster parcellations based on genetic correlation of surface area were used to calculate weighted averages of cortical surface measures28. Cortical thickness was calculated as the shortest distance between the white and pial surfaces19. Maps of relative cortical areal expansion were created by resampling individual subject surfaces to a standard tessellation, such that the area assigned to each mesh vertex reflects the degree of expansion or contraction relative to the atlas28,29. T1w was sampled to the cortical surface at a distance of ±0.2 mm along the normal vector at each surface location, and T1w cortical contrast was calculated from gray and white matter values33. Average thickness, area, and T1w were calculated for each cortical parcel. Surface-based maps were sampled to the FreeSurfer atlas22 and smoothed along the cortical surface34.

Diffusion MRI Preprocessing

As described above, spatial and intensity distortions caused by B0 field inhomogeneity were reduced using a robust and accurate procedure for reducing spatial and intensity distortions in EPI images12 that relies on the reversing gradient method6,34. Eddy current distortions were corrected with a nonlinear estimation procedure that used the diffusion gradient orientations and amplitudes to predict the pattern of distortions across the entire set of diffusion weighted volumes35.

Microstructural Analysis

Diffusion parameters were computed for a set of major brain fiber tracts36, as well as for other brain structures of interest. Conventional DTI methods were used to calculate measures related to microstructural tissue properties37-39, including the principal diffusion orientation, fractional anisotropy (FA), and mean, longitudinal, and transverse diffusivity (MD, LD, and TD). T2-weighted intensity (T2w) was calculated from the b=0 image (averaged if multiple b=0 images). To remove arbitrary intensity variation across subjects due to scanner settings (i.e., gain), T2w images were normalized for each subject. A linear fit with zero intercept was calculated between MD and b=0 intensity values using each voxel within a brain mask as a separate data point. The slope of the linear relationship was used to scale the T2w images.

AtlasTrack was used to automatically label long-range white matter tracts based on a probabilistic atlas of fiber tract locations and orientations35. The fiber atlas contains prior probabilities and orientation information for specific long-range projection fibers, including some additional fiber tracts not included in the original description, such as cortico-striate connections and inferior to superior frontal cortico-cortical connections.

Imaging variables associated with morphometry are identified in the PING ontology and data dictionary with the prefix “MRI_” followed by additional labels appropriate for specific measures (e.g., “cort_area”, “cort_thick”, “subcort_vol”, etc); and those associated with diffusion data with the prefix “DTI_” followed by additional labels appropriate for specific measures (e.g., “fiber”, “aseg”, etc). The “aseg” designation refers to regions of interest delineated in the volumetric analysis.

Imaging Data Quality Control

Raw Image Quality Control

Through the secure web-based application, individual sites uploaded DICOM images for each scan session. The data were automatically checked for completeness and protocol compliance, and images were reviewed for image quality by technicians trained by faculty. Specifically, images were inspected for motion artifacts, excessive distortion, operator error, or scanner malfunction. Quality ratings—good, average (usable), and bad (unacceptable)—were entered into the quality control utility within 24 hours from time of upload to allow re-scanning of subjects when possible.

T1-weighted images were examined slice-by-slice for evidence of excessive motion, such as stark ribbon or criss-cross artifacts within parenchyma and ghosting artifacts outside the head. Each volume was rated as either acceptable or recommended for rescan. Similarly, diffusion images were examined across all slices for signs of artifacts and poor image quality. Volumes with five or more slices showing significant slice-to-slice motion, motion artifacts, or whole-slice dropout were rejected (i.e., recommended for rescan). BOLD data were inspected for excessive subject movement and artifacts, and the mean frame-to-frame head motion was calculated.

Processed Image Quality Control

Processed images from all modalities were also examined for all participants, including subcortical volumetric segmentations, cortical areal parcellations, and white and pial surface reconstructions. A series of QC movies were also produced using Matlab scripts for each subject that assisted in data examination. A movie showing coronal views in sequence was used to judge white matter texture consistency and possible temporal underestimation. A related horizontal sequence was used to check for temporal underestimation in other regions (i.e., superior). A movie showing sagittal views was used for examination and rating of pial and dural overestimation along parietal regions and for signs of excessive head motion. White matter tracts produced using AtlasTrack were inspected for contiguity and overall quality and rated as acceptable or not.

Processing and Analysis of Genetic Information

Acquisition of Samples and DNA Extraction

Saliva collection in PING was performed using two different products from DNAgenoTek, Oragene•DISCOVER (OGR-500) and Oragene•DISCOVER (OGR-250). DNA extraction was carried out using respective protocols provided by Oragene, and DNA quantity was assessed using a Nanodrop fluorometer. DNA samples with at least 3ug total DNA were carried forward for further processing. Additional saliva samples were requested on specimens with less than 3ug total DNA. Stock DNA was stored at -80C for long-term storage. Ultimately samples were processed for 1411 PING participants.

Genome-Wide Genotyping

Genome-wide genotyping was performed on the extracted DNA using the Illumina Human660W-Quad BeadChip. The Illumina Human660W-Quad BeadChip (see www.illumina.com) contains more than 550,000 genetic markers (single nucleotide polymorphisms or SNPs and other variants) and is designed to measure most of the genetic variation present in the human genome (based on Hapmap release 21 reference data, see http://hapmap.ncbi.nlm.nih.gov/). The BeadChip measures variants on all autosomes (i.e., non-sex-chromosomes), the X and Y chromosomes, as well as mitochondrial DNA. The SNP call rate was >99% (i.e., >99% of the 539,865 SNPs were called). Acceptable genotyping data could be obtained by the Genomics Core for 1391 of the participant samples processed (out of the total 1396 received, or > 99.5%), including 727 males and 679 females.

Up to 1000 single nucleotide polymorphism values can be downloaded through the PING Portal Genetics Browser and the entire set of genotyping data is available in bulk.

Genetic Ancestry Assessment

In order to assess each participant's ancestry based on their genotype information, we constructed an ancestry-informative reference panel by bringing together genotype data from 2,513 individuals of known ancestry from 63 populations around the world using several publicly available sources: 1) the Human Genome Diversity Project (HGDP)36; 2) the Population Reference (POPRES)37; 3) the International HapMap 3 Consortium (HapMap3)38; and 4) the University of Utah dataset39. The reference panel was created in a stepwise fashion in order to ensure that the included individuals were not admixed among six major continental populations (African, Central Asian, East Asian, European, Native American, and Oceanic) and that each continental population was represented by a reasonably large number of diverse individuals originating in the relevant continent. The assembled reference panel contained genotype information at 16,433 strand-unambiguous SNPs. These markers exhibited low LD (r-squared less than 0.1 was observed between 99% of marker pairs), and allele frequency was higher than 1%.

To assess ancestry and admixture proportions in the PING participants, we used a supervised clustering approach implemented in the ADMIXTURE software40 and probabilistically assigned each participant to six clusters corresponding to the six major continental populations. The genotype profiles of the six populations were defined by the individuals who made up the reference panel. Although some individuals could be easily associated with a particular continental population, other individuals were clearly admixed. Such admixture could easily be associated with important phenotypic variation and hence needs to be quantified41–44. We therefore determined the degree of ancestry of each participant, effectively quantifying the amount of their genome that is likely to be derived from each of the six populations. Variables associated with genetic ancestry factors are identified in the PING ontology and data dictionary with the prefix “GAF_”.

Data Sharing

Access to the PING Data Resource is available through an online web interface at http://pingstudy.ucsd.edu. Here, information about the study, the consortium, and the methods are available for browsing, and instructions are given for applying for approval to explore, download, or request bulk shipping of data (for a fee covering media and shipping charges). Full sharing of all data is not permitted by the IRB for some PING sites. This includes restrictions on the sharing of some raw image data on the NITRC site and restrictions on sharing of some genetics data. All sharable data are available through the PING Data Portal to any researcher who holds a position in a research institution and is at least at the postdoctoral level (upon assent to the PING Data Use Agreement and approval of a brief data use description). Students can gain access to data if sponsored by eligible supervising researchers who agree to supervise the students’ compliance with the data use agreement. Raw image data for a subset of the participants is available through NITRC after an account is approved through the PING Portal.

Large data request downloads for PING are handled in two ways. First, raw image data in DICOM format is distributed using the dedicated image distribution platform on NITRC.org. The system used by NITRC (XNAT) is specifically built to host and download data across wide-area networks. Processed imaging and genetics data are also shared by PING using hard drives that can be ordered as PING-IN-A-BOX systems. The hard drive is shipped to customers (drive and shipping costs are billed to the recipient) and contains data in DICOM, Nifti, MGZ, and plink binary formats for easy integration into other existing post-processing pipelines.

Although PING data are publicly available, new data cannot be contributed into the PING data repository from outside sources. Given its strict standardization procedures for behavioral, imaging, and genomic data acquisition and processing, it is considered to be a completed resource.

The primary sponsors of the PING repository, NIDA and NICHD, as well as the NIH Office of the Director, have made a major, long term commitment to preservation of informative data repositories, particularly those with valuable genomic data, for future use. The contents of the repository may be transferred to one or more of the major NIH sponsored data repositories, such as NDAR or dbGaP, but it is likely that continuing support of the PING dataset will be provided. The Center for Human Development at UCSD is also committed to long-term maintenance of this resource.

PING Data Exploration Portal

An intelligent data exploration tool is available to facilitate the application of advanced statistical models to PING data, and to enable region of interest- and vertex-wise mapping of effects and 3D visualization of the results onto the cortical surface45. In contrast to many data sharing tools, the Portal also integrates appropriate statistical modeling capabilities, with structured descriptors for all of the PING measurements and an intuitive user interface to control diverse quality control and analysis workflows.

The Portal supports online collaborative exploration of interrelationships among measurements obtained from structural, behavioral, and genetic analyses. Using this web application, a user can select a specific variable of interest (e.g., cortical thickness) from the data dictionary and, with a single click of a button fit a statistical model (e.g., generalized additive model; GAM) with one or more independent variables and covariates (e.g., age, sex, scanner, genetic ancestry factor), and plot the resulting model fits along with the individual data points. For example, using the Portal one can interactively produce a scatter plot of each participant's total cortical surface area as a function of age, color-coded by sex, and controlling for other factors such as socioeconomic status (SES) and genetically derived ethnic ancestry. This statistical model can then be applied to every cortical vertex to produce a map of this relationship between age and surface area with covariates of interest. Main effects and interactions can be modeled for summary variables, averages, or by vertex, and displayed interactively, in real-time, using a WebGL-based application with control over color mapping, orientation of brain hemispheres, and corrections for multiple comparisons by controlling the false discovery rate (FDR)46. The same modeling functions can be applied to region of interest analyses either specified by the user or using built-in cortical parcellations from Freesurfer.

The model, or query, can then be stored by the user and shared with other users. The ease of interaction permits deeper understanding of the complex relationships within the dataset and facilitates the discovery of hidden structure in the data. In PING, for example, the Portal has been an effective way to visualize scanner effects and the results of different methods for modeling them. The Portal supports an online chat feature that is used to send model descriptions to other people visiting the page, as well as a user feedback forum monitored by the developer. Users can construct and download datasets, statistical reports, results tables, model specifications, and figures for off-line analysis, archiving, or publication.

Highlights.

  • We provide a brief description of the Pediatric Imaging Neurocognition and Genetics (PING) Study.

  • We describe the data in the PING Data Repository.

  • We outline the methods used to generate the data.

  • We describe the procedure for accessing and exploring the data through the PING Portal.

Acknowledgments

The PING Project was supported by the National Institute on Drug Abuse and the Eunice Kennedy Shriver National Institute of Child Health and Human Development with the following awards: RC2DA029475 and R01 HD061414.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Brown TT, et al. Neuroanatomical Assessment of Biological Maturity. Curr Biol. 2012 doi: 10.1016/j.cub.2012.07.002. doi:S0960-9822(12)00793-2 [pii] [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Akshoomoff N, et al. The NIH Toolbox Cognition Battery: results from a large normative developmental sample (PING) Neuropsychology. 2014;28:1–10. doi: 10.1037/neu0000001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.McCarty CA, et al. PhenX RISING: real world implementation and sharing of PhenX measures. BMC Med Genomics. 2014;7:16. doi: 10.1186/1755-8794-7-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Keator DB, et al. A national human neuroimaging collaboratory enabled by the Biomedical Informatics Research Network (BIRN) IEEE Trans Inf Technol Biomed. 2008;12:162–172. doi: 10.1109/TITB.2008.917893. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Jack CR, Jr, et al. The Alzheimer's Disease Neuroimaging Initiative (ADNI): MRI methods. J Magn Reson Imaging. 2008;27:685–691. doi: 10.1002/jmri.21049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Jovicich J, et al. Reliability in multi-site structural MRI studies: effects of gradient non-linearity correction on phantom and human data. Neuroimage. 2006;30:436–443. doi: 10.1016/j.neuroimage.2005.09.046. [DOI] [PubMed] [Google Scholar]
  • 7.Chang H, Fitzpatrick JM. A technique for accurate magnetic resonance imaging in the presenceof field inhomogeneities. IEEE Trans Med Imaging. 1992;11:319–329. doi: 10.1109/42.158935. [DOI] [PubMed] [Google Scholar]
  • 8.Holland D, Dale AM. Nonlinear registration of longitudinal images and measurement of change in regions of interest. Med Image Anal. 2011;15:489–497. doi: 10.1016/j.media.2011.02.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.White N, et al. et al. PROMO: Real-time prospective motion correction in MRI using image-based tracking. Magn Reson Med. 2010;63:91–105. doi: 10.1002/mrm.22176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Brown TT, et al. Prospective motion correction of high-resolution magnetic resonance imaging data in children. Neuroimage. 2010;53:139–145. doi: 10.1016/j.neuroimage.2010.06.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Kuperman JM, et al. Prospective motion correction improves diagnostic utility of pediatric MRI scans. Pediatr Radiol. 2011;41:1578–1582. doi: 10.1007/s00247-011-2205-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Holland D, Kuperman JM, Dale AM. Efficient correction of inhomogeneous static magnetic field-induced distortion in Echo Planar Imaging. Neuroimage. 2010;50:175–183. doi: 10.1016/j.neuroimage.2009.11.044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Jovicich J, et al. Reliability in multi-site structural MRI studies: effects of gradient non-linearity correction on phantom and human data. Neuroimage. 2006;30:436–443. doi: 10.1016/j.neuroimage.2005.09.046. [DOI] [PubMed] [Google Scholar]
  • 14.Sled J, Z A, E A. A nonparametric method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans Med Imaging. 1998;7:87–97. doi: 10.1109/42.668698. [DOI] [PubMed] [Google Scholar]
  • 15.Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage. 1999;9:179–94. doi: 10.1006/nimg.1998.0395. [DOI] [PubMed] [Google Scholar]
  • 16.Dale AM, Sereno MI. Improved localization of cortical activity by combining EEG and MEG with MRI cortical surface reconstruction: A linear approach. J Cogn Neurosci. 1993;5:162–176. doi: 10.1162/jocn.1993.5.2.162. [DOI] [PubMed] [Google Scholar]
  • 17.Desikan RS, et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage. 2006;31:968–980. doi: 10.1016/j.neuroimage.2006.01.021. [DOI] [PubMed] [Google Scholar]
  • 18.Destrieux C, Fischl B, Dale A, Halgren E. Automatic parcellation of human cortical gyri and sulci using standard anatomical nomenclature. Neuroimage. 2010;53:1–15. doi: 10.1016/j.neuroimage.2010.06.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Fischl B, Dale AM. Measuring the thickness of the human cerebral cortex from magnetic resonance images. Proc Natl Acad Sci U S A. 2000;97:11050–11055. doi: 10.1073/pnas.200033797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Fischl B, Liu A, Dale AM. Automated manifold surgery: constructing geometrically accurate and topologically correct models of the human cerebral cortex. IEEE Trans Med Imaging. 2001;20:70–80. doi: 10.1109/42.906426. [DOI] [PubMed] [Google Scholar]
  • 21.Fischl B, et al. Whole Brain Segmentation: Automated Labeling of Neuroanatomical Structures in the Human Brain. Neurotechnique. 2002;33:341–355. doi: 10.1016/s0896-6273(02)00569-x. [DOI] [PubMed] [Google Scholar]
  • 22.Fischl B, Sereno MI, Tootell RB, Dale AM. High-resolution intersubject averaging and a coordinate system for the cortical surface. Hum Brain Mapp. 1999;8:272–284. doi: 10.1002/(SICI)1097-0193(1999)8:4<272::AID-HBM10>3.0.CO;2-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Fischl B, Sereno MI, Dale AM. Cortical surface-based analysis II: Inflation, flattening, a surface-based coordinate system. Neuroimage. 1999;9:195–207. doi: 10.1006/nimg.1998.0396. [DOI] [PubMed] [Google Scholar]
  • 24.Fischl B, et al. Sequence-independent segmentation of magnetic resonance images. Neuroimage. 2004;23(Suppl 1):S69–84. doi: 10.1016/j.neuroimage.2004.07.016. [DOI] [PubMed] [Google Scholar]
  • 25.Salat DH, et al. Regional white matter volume differences in nondemented aging and Alzheimer's disease. Neuroimage. 2009;44:1247–1258. doi: 10.1016/j.neuroimage.2008.10.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Ségonne F, et al. A hybrid approach to the skull stripping problem in MRI. Neuroimage. 2004;22:1060–1075. doi: 10.1016/j.neuroimage.2004.03.032. [DOI] [PubMed] [Google Scholar]
  • 27.Ségonne F, Pacheco J, Fischl B. Geometrically Accurate Topology-Correction of Cortical Surfaces Using Nonseparating Loops. IEEE Trans Med Imaging. 2007;26:518–529. doi: 10.1109/TMI.2006.887364. [DOI] [PubMed] [Google Scholar]
  • 28.Chen CH, et al. Hierarchical genetic organization of human cortical surface area. Science (80-.) 2012;335:1634–1636. doi: 10.1126/science.1215330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Joyner AH, et al. A common MECP2 haplotype associates with reduced cortical surface area in humans in two independent populations. Proc Natl Acad Sci U S A. 2009;106:15483–15488. doi: 10.1073/pnas.0901866106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Chen CH, et al. Genetic topography of brain morphology. Proc Natl Acad Sci U S A. 2013;110:17089–17094. doi: 10.1073/pnas.1308091110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Chen CH, et al. Genetic influences on cortical regionalization in the human brain. Neuron. 2011;72:537–544. doi: 10.1016/j.neuron.2011.08.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Sled JG, Zijdenbos AP, Evans AC. A non-parametric method for automatic correction of intensity non-uniformity in MRI data. IEEE Trans Med Imaging. 1998;17:87–97. doi: 10.1109/42.668698. [DOI] [PubMed] [Google Scholar]
  • 33.Westlye LT, et al. Increased sensitivity to effects of normal aging and Alzheimer's disease on cortical thickness by adjustment for local variability in gray/white contrast: a multi-sample MRI study. Neuroimage. 2009;47:1545–1557. doi: 10.1016/j.neuroimage.2009.05.084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Hagler DJ, Jr, Saygin AP, Sereno MI. Smoothing and cluster thresholding for cortical surface-based group analysis of fMRI data. Neuroimage. 2006;33:1093–1103. doi: 10.1016/j.neuroimage.2006.07.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Hagler DJ, Jr, et al. Automated white-matter tractography using a probabilistic diffusion tensor atlas: Application to temporal lobe epilepsy. Hum Brain Mapp. 2009;30:1535–1547. doi: 10.1002/hbm.20619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Cann HM, et al. A human genome diversity cell line panel. Science. 2002;296:261–2. doi: 10.1126/science.296.5566.261b. [DOI] [PubMed] [Google Scholar]
  • 37.Nelson MR, et al. The Population Reference Sample, POPRES: a resource for population, disease, and pharmacological genetics research. Am J Hum Genet. 2008;83:347–358. doi: 10.1016/j.ajhg.2008.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Altshuler DM, et al. Integrating common and rare genetic variation in diverse human populations. Nature. 2010;467:52–8. doi: 10.1038/nature09298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Xing J, et al. Mobile elements create structural variation: analysis of a complete human genome. Genome Res. 2009;19:1516–26. doi: 10.1101/gr.091827.109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Alexander DH, Novembre J, Lange K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 2009;19:1655–64. doi: 10.1101/gr.094052.109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Goetz LH, Uribe-Bruce L, Quarless D, Libiger O, Schork NJ. Admixture and clinical phenotypic variation. Hum Hered. 2014;77:73–86. doi: 10.1159/000362233. [DOI] [PubMed] [Google Scholar]
  • 42.Norden-Krichmar TM, et al. Correlation analysis of genetic admixture and social identification with body mass index in a Native American community. Am J Hum Biol. 26:347–60. doi: 10.1002/ajhb.22521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Nievergelt CM, et al. Chip-based direct genotyping of coding variants in genome wide association studies: utility, issues and prospects. Gene. 2014;540:104–9. doi: 10.1016/j.gene.2014.01.069. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Libiger O, Schork NJ. A Method for Inferring an Individual's Genetic Ancestry and Degree of Admixture Associated with Six Major Continental Populations. Front Genet. 2012;3:322. doi: 10.3389/fgene.2012.00322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Bartsch H, Thompson WK, Jernigan TL, Dale AM. A web-portal for interactive data exploration, visualization, and hypothesis testing. Front Neuroinform. 2014;8:25. doi: 10.3389/fninf.2014.00025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Genovese CR, Lazar NA, Nichols T. Thresholding of statistical maps in functional neuroimaging using the false discovery rate. Neuroimage. 2002;15:870–8. doi: 10.1006/nimg.2001.1037. [DOI] [PubMed] [Google Scholar]

RESOURCES