Multimodal Neuroimaging in Schizophrenia: Description and Dissemination
In this paper we describe an open-access collection of multimodal neuroimaging data in schizophrenia for release to the community. Data were acquired from approximately 100 patients with schizophrenia and 100 age-matched controls during rest as well as several task activation paradigms targeting a hierarchy of cognitive constructs. Neuroimaging data include structural MRI, functional MRI, diffusion MRI, MR spectroscopic imaging, and magnetoencephalography. For three of the hypothesis-driven projects, task activation paradigms were acquired on subsets of ~200 volunteers which examined a range of sensory and cognitive processes (e.g., auditory sensory gating, auditory/visual multisensory integration, visual transverse patterning). Neuropsychological data were also acquired and genetic material via saliva samples were collected from most of the participants and have been typed for both genome-wide polymorphism data as well as genome-wide methylation data. Some results are also presented from the individual studies as well as from our data-driven multimodal analyses (e.g., multimodal examinations of network structure and network dynamics and multitask fMRI data analysis across projects). All data will be released through the Mind Research Network’s collaborative informatics and neuroimaging suite (COINS).
KeywordsNeuroimaging Schizophrenia fMRI MEG Magnetoencephalography Spectroscopy Genetics Sensory gating Multimodal integration Memory Transverse patterning DTI ICA COINS COBRE MATRICS
The purpose of this paper is to announce the release of a large multimodal neuroimaging dataset on chronic schizophrenia patients and healthy controls [e.g., functional MRI, diffusion tensor imaging (DTI), proton MR spectroscopic imaging (1H–MRS), and magnetoencephalography (MEG) data] along with clinical, genetic, and neurocognitive assessments. Data were acquired from an NIH-supported COBRE (Centers of Biomedical Research Excellence) Phase I grant (2P20GM103472). Here we describe the projects that generated this large multimodal dataset for dissemination. An overview of the hypotheses motivating these projects and types of multimodal data to be released are presented in the Introduction, followed by details of each experimental protocol and a synopsis of results in the Methods & Results Section, followed by a plan for releasing the data in the Information Sharing section.
The multimodal data to be released were acquired across four hypothesis-driven projects with a goal of imaging the same sample of approximately 200 volunteers within each project. This included 100 patients with schizophrenia (SPs) and 100 healthy controls (HCs). In addition, analysis using data-driven approaches developed as part of the COBRE (e.g., multitask fMRI data analysis across projects) are presented. Schizophrenia is a serious illness that is primarily characterized by delusions, hallucinations, and disorganized speech. However, most patients also have cognitive dysfunction, resulting in problems with social and interpersonal interactions (Green et al. 2004, Green and Nuechterlein 2004). This dysfunction led to its original conception as an early form of dementia (dementia praecox) by Kraepelin (1896). Although current drug regimens are mostly successful at controlling the psychotic symptoms of the disorder (~20–30% do not respond adequately), cognitive and social functions generally remain impaired. Our Mind Research Network (MRN) COBRE team views schizophrenia as a disorder characterized by abnormalities in structural, functional, and effective connectivity between cortical and subcortical brain regions, producing abnormalities in information processing across distributed brain circuits. The novel combination of non-invasive neuroimaging techniques used here provided an unprecedented view of the neuronal pathologies that underlie the core cognitive dysfunctions of schizophrenia.
Overview of the 4 Hypothesis-Driven Projects
Numbers of participants for each project and task. Top portion—structural MR scans acquired on all SP and HC available to all projects. Bottom portion—Projects 1–3: functional scans (fMRI and MEG) acquired by some of the projects and for some of the SP and HC. Project 4: Structural and spectroscopy scans
I. Structural Scans
Structural MRI (sMRI)
HC = 200 SP = 224 (2 scans each)
II. Project Specific Scans
Project 1: Auditory Gating/Orienting and Multisensory Cognitive Control
Task 1—Gating In/Out
HC = 65 SP = 58
HC = 66 SP = 57
HC = 25 SP = 30
Task 3—Resting State
HC = 94 SP = 115
HC = 22 SP = 24
Task 4—Multisensory Cognitive Control (Visual/auditory)
HC = 27 SP = 55
Task 5—Multisensory Detection (Visual/auditory)
HC = 27 SP = 60
Project 2: Multisensory Integration (auditory and visual)
Auditory alone (near)
HC = 67 SP = 68
HC = 67 SP = 68
Visual alone (near)
HC = 67 SP = 68
HC = 67 SP = 68
HC = 67 SP = 68
HC = 67 SP = 68
HC = 67 SP = 68
HC = 67 SP = 68
Auditory alone (far)
HC = 67 SP = 68
HC = 67 SP = 68
Visual alone (far)
HC = 67 SP = 68
HC = 67 SP = 68
HC = 67 SP = 68
HC = 67 SP = 68
HC = 67 SP = 68
HC = 67 SP = 68
HC = 36 SP = 37
Project 3: Transverse Patterning (visual)
Task 1—Elemental Control (verbal)
HC = 61 SP = 47
HC = 60 SP = 49
Task 2—TP (verbal)
HC = 61 SP = 47
HC = 60 SP = 49
Task 3—Elemental Control (nonverbal)
HC = 61 SP = 47
HC = 60 SP = 49
Task 4—TP (nonverbal)
HC = 61 SP = 47
HC = 60 SP = 49
Project 4: No Experimental Tasks
Diffusion Tensor Imaging (DTI)
HC = 96 SP = 97
Proton MR Spectroscopy (1H–MRS)
HC = 69 SP = 62
There is general consensus in the schizophrenia literature that: 1) widespread, generalized deficits exist and appear to be related to disconnection between several regions; 2) deficits exist at both basic sensory and higher levels of functioning; 3) research can examine these disconnections but the problem is most likely more complex than one study can fully probe; and 4) multimodal imaging may provide complementary information (location, timing, content). However, there is no consensus on the primary sites of disconnection. Our four COBRE projects focused on anatomical and functional examples of cortico-cortical and cortico-subcortical disconnections in hemodynamic functioning (fMRI: Projects 1, 2 and 3), in the temporal dynamics of activity between these regions (MEG: Projects 1, 2 and 3), at the morphometric level (Project 4), and in white matter integrity (DTI: Projects 2, 3, and 4). In addition, Project 4 utilized MR Spectroscopic Imaging (1H–MRS) to examine metabolites of interest [e.g. N-acetylaspartate (NAA)] to potentially identify regions with primary neuronal damage.
Project 1: Multimodal Imaging of Auditory Pre-Attentive/Attentive and Cognitve Control Processes in SP (Andrew Mayer, PI)
This project acquired fMRI data from 4 task activation paradigms and MEG data from 1 task activation paradigm ranging from pre-attentive processes to multisensory cognitive control. Multiple lines of research have consistently documented auditory sensory processing deficits in SP. One such deficit which has been previously examined in SP is poor sensory gating; i.e., there is a failure to inhibit the electrophysiological response for the second of two rapidly presented stimuli in SP compared to HC (Adler et al. 1982, Freedman et al. 1987). This effect is typically reported in terms of a gating ratio comparing the amplitude of the response for the first (S1) and second (S2) stimulus (S2/S1*100). Poor sensory gating has been characterized as not only a deficit in selective attention and/or in the formation of memory traces, but also as a useful bio-marker of the cognitive and social dysfunction that is typically observed in SP (Cullum et al. 1993, Lijffijt et al. 2009). Electroencephalographic (EEG) and MEG studies of sensory gating have implicated the temporal lobes, including the superior temporal gyrus, as the most likely neuronal generator of the sensory gating deficit (Hanlon et al. 2005a, Huang et al. 2003, Thoma et al. 2003, 2005). Although other invasive neuroimaging techniques suggest a role for the hippocampus and prefrontal cortex in sensory gating (Boutros et al. 2005, Freedman et al. 1996, Grunwald et al. 2003, Korzyukov et al. 2007), neither EEG nor MEG studies have revealed hippocampal activation. While this negative result may be reflective of a true lack of hippocampal activation in gating, it may also be secondary to technical limitations with EEG and MEG, such as limits in spatial resolution, failure to examine appropriate time epochs, and/or difficulties localizing sources in deep structures under certain conditions. In contrast, fMRI is not restricted by these limitations, suggesting that it may be an alternative modality for studying the entire neuronal network underlying the gating deficit, including mesial temporal and prefrontal sources. In this protocol, SP and HC were presented with identical clicks, identical tones, and non-identical tones while undergoing functional imaging. It was predicted that a double dissociation of functioning would be observed with SP exhibiting higher gating ratios for the identical tones and clicks conditions and lower gating ratios for the non-identical tones condition, whereas HC would exhibit the opposite pattern. The functional data from these conditions were directly compared and correlated with behavioral and neuropsychological measures to further quantify the relationship between brain function and the observed clinical pathology that characterizes this basic inhibitory failure. Finally, since impaired sensory gating is thought to be an endophenotypical marker for schizophrenia, the genetic contribution to our measures of electrophysiological and hemodynamic functioning in both HC and SP were investigated as well. Data recorded from other tasks within this project included an auditory cueing paradigm that measured both exogenous orienting and inhibition of return and multisensory cognitive control and detection. A multisensory cognitive control paradigm was used to determine if deficits in SP result from dysfunction within the cognitive control network (CCN; top-down) and/or unisensory cortex (bottom-up), using congruent or incongruent multisensory (auditory and visual) numeric stimuli.
Project 2: Multisensory Integration in SP using MEG and fMRI (Julia Stephen, PI)
The goal of this project was to study the neural mechanisms underlying multisensory integration of auditory and visual stimuli in SP and HC using MEG and fMRI combined with structural MRI (sMRI). Integration of multisensory information is critical to understanding our environment and provides an important bridge between simple sensory perception and more complex cognitive tasks. That is, the integrity of a larger, distributed neural circuit can be evaluated without engaging complex cognitive functions, where issues of strategy, for example, can confound the results. It has been established that multisensory stimulation generally improves reaction times and performance (based on percent correct) relative to unisensory stimulation at levels that are difficult to perceive (Stein et al. 1993), and more recent results indicate that multisensory facilitation can be observed across the perception spectrum (Stanford and Stein 2007). Some reports have shown that SP have slower reaction times to multisensory stimuli than HC [e.g., (Williams et al. 2010)]. Since the association areas of the temporal lobe are necessary for integration of auditory and visual information (Schroeder and Foxe 2002, Schroeder et al. 2001), altered multisensory processing is consistent with the findings that SP show functional and anatomical differences in the temporal lobes [e.g., (Patterson et al. 2008)]. In this project, auditory and visual multisensory integration was studied using an ecologically valid paradigm that simulates the sight and sound of a soccer ball bouncing either near or far from the subject. The stimuli were presented as auditory alone, visual alone, and synchronous auditory/visual conditions, to which subjects responded with a button press. Based on known deficits in the temporal lobe, it was hypothesized that the responses to combined auditory/visual stimuli would be impaired in SP relative to HC. In addition, due to a reported deficit in the visual dorsal stream in SP (Butler et al. 2007, Butler et al. 2005, Doniger et al. 2002, Foxe et al. 2001, 2005, Kim et al. 2006, 2005, O'Donnell et al. 1996, Schechter et al. 2006, 2005), a neural circuit sensitive to the spatial location of objects, it was hypothesized that there would be a difference in visual responses to the near versus far stimuli in SP, relative to HC. These deficits may be related to unisensory deficits and/or to difficulties in synchronizing activity across widespread regions connected within the circuit which will likely be seen differently in MEG versus fMRI measures, given the complementary nature of these measurement techniques. The long term goal was to identify the local cortical deficits, as well as the deficits in cortical networks that lead to the abnormalities observed in SP. Knowledge about these deficits should permit the evaluation of relationships with neurochemical abnormalities, ultimately leading to more targeted pharmacological interventions.
Project 3: Fronto-temporal Coherence: A Test of the Disconnection Hypothesis in SP (Faith Hanlon, PI)
Prefrontal cortex (PFC) and hippocampal structures play a central role in working memory and relational memory impairments exhibited in SP (Goldman-Rakic 1994, Hanlon et al. 2005a, 2006, Honey and Fletcher 2006, Wolf et al. 2009). These PFC and hippocampal functional deficits have traditionally been attributed to properties of the cortical structures themselves. However, an alternative view is that the deficits are due to the disconnection between these structures in SP (Friston 1999, 2002, Friston and Frith 1995, Johnson 2006). Project 3 tested this disconnection hypothesis; that is, there is a functional disconnection between frontal and temporal cortices due to an abnormal anatomical connection of the underlying white matter tract, uncinate fasciculus, that links the two. The most striking evidence supporting this abnormal fronto-temporal connectivity in SP is found in neuroanatomical studies that describe abnormalities in the uncinate fasciculus (Burns et al. 2003, Kubicki et al. 2002, Park et al. 2004). Despite this finding of abnormal anatomical connectivity, the relationship between functional and anatomical connectivity of this PFC-hippocampal network has not been sufficiently assessed. The focus of this project was on examining the functional and anatomical connectivity in SP and HC using MEG and fMRI during a working-relational memory task, transverse patterning or TP (Burns et al. 2003, Hanlon et al. 2012, 2011, 2003, 2005b, Kubicki et al. 2002, Park et al. 2004). TP is very similar to the childhood-game “Rock, Paper, Scissors.” In TP, subjects chose between stimuli presented in pairs, with the correct choice being a function of the specific pairing. To complete the task the subject must discover, encode, and maintain the distinct relationships among the stimuli, thus requiring working memory and relational memory integration. Fronto-temporal functional connectivity was examined via assessing the temporal correlation, or coherence between the PFC (BA 9 and BA 10) and hippocampus during TP performance. In addition, anatomical connectivity using fractional anisotropy (FA) measures derived from DTI of the uncinate fasciculus was also evaluated. Finally, the functional/structural relationships between PFC and hippocampus, memory function, overall functioning, and clinical symptomatology were examined. Understanding these relationships could potentially lead to treatments or therapies aimed at improving memory, thereby improving overall functioning and clinical symptoms.
Project 4: Fronto-Subcortical Disconnection underlying Neurocognitive Dysfunction in SP (Rex Jung, PI)
This project used 3 T MRI to investigate whether general cognitive functioning in SP is related to circuit level white matter (WM), metabolic, and volumetric changes in subcortical gray matter (GM) and WM regions suggestive of fronto-subcortical disconnection. The available neuroscientific literature in HC, including functional (i.e., fMRI, positron emission tomography) and biochemical/structural (i.e., 1H–MRS, DTI, voxel-based morphometry or VBM) neuroimaging paradigms, show a striking consensus suggesting that variations in a distributed network predict individual differences found on intelligence and reasoning tasks in HC (Jung and Haier 2007). This parieto-frontal integration theory, or P-FIT model includes the dorsolateral PFC (BAs 6, 9, 10, 45, 46, 47), the inferior (BAs 39, 40) and superior (BA 7) parietal lobule, the anterior cingulate (BA 32), and regions within the temporal (BAs 21, 37) and occipital (BAs 18, 19) lobes. The main structural brain abnormality established in SP research is modest lateral ventricle enlargement indicative of atrophy (Chua and McKenna 1995), a finding established in the very first report utilizing computerized tomography (Johnstone et al. 1976), and suggestive of WM atrophic changes. Since then various brain abnormalities have been identified in SP, both cortically and subcortically, including gross morphological changes, reduced metabolism, and cellular pathology. This diffuse pathology has led some to hypothesize that SP is “a disease of neuronal connectivity,” arising from abnormalities at the level of the neuron and myelin (Andreasen 2000). Evidence for WM abnormality in SP includes structural oligodendrocyte abnormalities (Hof et al. 2002, 2003, Orlovskaya et al. 1997, 1999, Uranova et al. 2001), myelin protein alterations (Chambers and Perrone-Bizzozero 2004), myelin gene polymorphisms (Liu et al. 2005, Novak et al. 2002, Peirce et al. 2006, Qin et al. 2005, Tan et al. 2005, Wan et al. 2005), and altered expression of genes involved in the formation and maintenance of myelin sheaths as well as oligodendrocyte survival and proliferation revealed by DNA microarray methods (Bunney and Bunney 2000, Goldman-Rakic and Selemon 1997, Hakak et al. 2001). The simultaneous utilization of DTI, 1H–MRS, and sMRI allowed us to relate WM abnormalities to both cortical and subcortical metabolic and morphological changes that, in turn, may underlie ongoing neurocognitive decline in SP. Since SP was hypothesized to be a disorder that involves the integration of information among distributed brain circuits, Project 4 investigated: 1) whether fronto-subcortical network abnormalities are present in SP; 2) whether cognitive functioning in SP is related to circuit level dysfunction; and 3) whether regions beyond the identified brain circuits contribute significantly to broad cognitive functioning deficits characteristic of SP. A hierarchical approach was explored, in which cognitive functioning was predicted by cortical thinning, WM microstructure changes, and metabolic changes within discrete fiber tracts in SP. Also tested was the hypothesis that ongoing neurocognitive impairment in SP is related to chronic disconnection of fronto-subcortical circuits, reflected in morphological contraction of discrete frontal lobe regions critical to higher cognitive functioning. Finally, we explored whether specific genes associated with the formation, development, and functioning of myelin are associated with circuit level dysfunction in SP. No research to date has simultaneously explored diffusivity, metabolic, and morphometric abnormalities within these critical brain circuits as they relate to neurocognitive dysfunction in SP. We see this as a critical line of investigation that will help us understand and potentially intervene in the ongoing cognitive dysfunction that currently limits the ability of treated patients to return to more normal occupational and social functioning.
Methods & Results
An overview of participant characteristics as well as data acquisition parameters common across the four projects using structural/functional imaging measures is presented here. Details of each project’s experimental protocols are discussed separately under each project description. Table 1 contains a breakdown of the number of SP and HC involved in each task and the types of data acquired for each task. All structural imaging, shared across projects, were acquired under Project 4. All studies were approved by the institutional review board (IRB) of the University of New Mexico (UNM) and all subjects provided written informed consent.
Subject Recruitment and Evaluation
Patients were recruited primarily from the UNM Psychiatric Center and secondarily from the Raymond G. Murphy Veterans Affairs Medical Center, given the investigators’ clinical appointments. Some were recruited from other psychiatric clinics in the Albuquerque metropolitan area as well. Inclusion criteria for patient selection included diagnosis of schizophrenia or schizoaffective disorder between 18 and 65 years of age. Each SP completed the Structured Clinical Interview for DSM-IV Axis I Disorders [SCID--(First et al. 1995a, b)] for diagnostic confirmation (consensus was reached by two research psychiatrists using the SCID-DSM-IV-TR, patient version) and evaluation for co-morbidities. SP had to demonstrate retrospective and prospective clinical stability to be included in this investigation.
Because multiple neuroimaging sessions were required from each subject on different days, a concerted effort to prevent meaningful variation in the SP’s critical state variables (e.g., symptoms, medication dose, neurological side-effects) was built into the design of the study. Hence SP were seen weekly by the study clinicians during the period of data collection. The study clinicians and staff were responsible for subject recruitment, determining that subjects met clinical stability criteria, genetic testing, and administering all clinical assessments and standardized cognitive batteries to subjects. Standard symptom ratings [e.g., Positive and Negative Syndrome Scale (PANSS) (Kay et al. 1987)] as well as neurological side-effects [e.g., akathisia—(Barnes 1989); parkinsonism—(Simpson and Angus 1970); and tardive dyskinesia—(Schooler and Kane 1982)], were collected within one week of every neuroimaging assessment. The COBRE Stability Clinic determined retrospective stability from relevant psychiatric records documenting that no change in symptomatology or type/dose of psychotropic medications occurred during the three months prior to the referral. This Clinic assessed prospective stability during three consecutive weekly visits and during each imaging assessment. Prospective stability was defined as no change in clinical symptoms > two points from the positive symptom items on the PANSS, no score of “worse” or “much worse” on the Clinical Global Impression Scale (CGI) (Guy 1976), no suicidal or violent ideation, and no psychiatric or medical hospitalizations. The doses of antipsychotic medications were converted to olanzapine equivalents to estimate medication load (Gardner et al. 2010). Subjects with a history of neurological disorders including head trauma (loss of consciousness greater than 5 min), mental retardation or history of active substance dependence or abuse (except for nicotine) within the past year were excluded (i.e., history of dependence on phencyclidine (PCP), amphetamines or cocaine, or history of PCP, amphetamine, or cocaine use within the last 12 months). All subjects had a negative toxicology screen for drugs of abuse at the start of the study.
HCs were recruited from the same geographic location via IRB-approved advertisement and completed the SCID-Non-Patient Edition to rule out Axis I conditions (First et al. 1995a, b). Additional exclusion criteria for HCs included a current or past psychiatric disorder (with the exception of one lifetime major depressive episode), head trauma with a loss of consciousness greater than 5 min, recent history of substance abuse or dependence, depression or antidepressant use within the past 6 months, lifetime antidepressant use of more than one year, and history of a psychotic disorder in a first-degree relative.
SP and HC groups were similar in age (37.9 ± 14 vs. 37.5 ± 11.8) and male/female proportion (81/19 vs. 72/26), respectively. The SPs had been chronically ill (age of first psychotic symptoms 21.2 ± 8) and had lower education compared to HCs (3.8 ± 1.4 vs. 4.6 ± 1.3, respectively; 3 = high school graduate or equivalent, 4 = some college). All participants refrained from smoking for at least one hour before scanning.
MRI/fMRI data collection
EPI slices were collected in sequential ascending order on a Siemens 3 T TIM Trio scanner, located at MRN, using a 12-channel head coil. A sagittal gradient echo scout image through the midline was obtained to prescribe oblique axial image slices parallel to the anterior-posterior commissure (AC-PC) line. Oblique slices were used to minimize the orbitofrontal susceptibility artifact (Deichmann et al. 2003). High resolution T1-weighted images were acquired with a 5-echo multi-echo MPRAGE sequence [TE (echo times) = 1.64, 3.5, 5.36, 7.22, 9.08 ms, TR (repetition time) = 2.53 s, TI (inversion time) = 1.2 s, 7○ flip angle, number of excitations (NEX) = 1, slice thickness = 1 mm, FOV (field of view) = 256 mm, resolution = 256 × 256] for region of interest analyses and spatial normalization. All fMRI data were collected using a conventional single-shot, gradient-echo echoplanar pulse sequence with lipid suppression [TR = 2000 ms; TE = 29 ms; flip angle = 75○; FOV = 240 mm; matrix size = 64 × 64; 33 slices; voxel size: 3.75 × 3.75 × 4.55 mm]. The first image of each run was eliminated to account for T1 equilibrium effects.
Diffusion MRI (dMRI) Data Collection
dMRI data were collected along the AC-PC line, throughout the whole brain, with FOV = 256 × 256 mm, 128 × 128 matrix, 72 slices with a slice thickness of 2 mm (isotropic 2 mm resolution), NEX = 1, TE = 84 ms and TR = 9000 ms. A multiple-channel radiofrequency (RF) coil was used, with GRAPPA (X2), 30 gradient directions with b = 800 s/mm2. The b = 0 experiment was repeated five times (Jones et al. 1999), and equally interspersed between the 30 gradient directions. The total imaging time was approximately 6 min. This protocol was repeated twice to increase the signal to noise ratio (SNR).
Proton MR Spectroscopic Imaging (1H–MRS)
A standard PRESS chemical shift imaging 1H–MRS was performed with a phase-encoded version of a point-resolved spectroscopy sequence (PRESS) both with and without water pre-saturation as described (Gasparovic et al. 2011). Briefly, the following parameters were used: TE = 40 ms, TR = 1500 ms, slice thickness = 15 mm, FOV = 220 × 220 mm, circular k-space sampling (diameter = 24), total scan time = 2 × 582 s. A TE of 40 ms was chosen to improve detection of the Glx (combined glutamate and glutamine) signal (Mullins et al. 2008). The nominal voxel size was 6.9 × 6.9 × 15 mm3 (0.71 cm3) but the effective voxel volume is estimated to be 2.4 cm3. The 1H–MRS volume of interest (VOI) was prescribed from an axial T2-weighted image to lie immediately above the lateral ventricles and parallel to the AC-PC axis, and included portions of the cingulate gyrus and the medial frontal and parietal lobes. To minimize the chemical shift artifact, the transmitter was set to the frequency of the NAA methyl peak during the acquisition of the metabolite spectra and to the frequency of the water peak during the acquisition of the unsuppressed water spectra. Additionally, the outermost rows and columns of the VOI were excluded from analysis. Spectra were automatically fitted with LCModel (version 6.1) in the spectral range between 1.8–4.2 ppm in reference to the non-water-suppressed data using “water-scaling” (Gasparovic et al. 2006), blind to diagnostic group. Only metabolite values with goodness of fit, as measured by the standard-deviation (SD), of ≤20 were further analyzed. Finally, results from LCModel for the metabolites of interest were corrected for partial volume (using SPM-5 segmented T1 images) and relaxation effects (from literature values), as outlined previously (Gasparovic et al. 2011). Metabolites sampled include creatine (Cr), phosphocreatine (PCr), glutamate (Glu), glutamine (Gln), choline (Ch), myo-inositol (Ins), NAA and N-acetylasparthglutamate (NAAG).
MEG data collection
MEG data were collected in a magnetically shielded room (AK3b, Vacuumschmelze GmbH & Co. KG) using the Elekta Neuromag 306 channel biomagnetometer (VectorView, Elekta AB) located at MRN. Before acquisition of MEG scans, four head-position indicator (HPI) coils were placed around the participant’s head and secured with tape. The HPI coil location and head shape digitization was obtained using the Polhemus 3-D tracking device. Three fiducials (left and right preauricular and nasion) were identified, along with ~200 additional points around the scalp, to define the head-centered coordinate system and for co-registration of the MEG data to the MRI structural image. Two channels of electrooculogram (EOG), vertical (one electrode placed above the left eyebrow and one placed below the left eye) and horizontal (one electrode each was placed lateral to the outer canthus of the left and right eye), were recorded to provide signals for artifact rejection of eye movements and blinks. One cardiac channel (leads placed just below the left and right clavicle) was also recorded simultaneously with the MEG data for identification of heart beat artifacts. Eye-tracking (Eyelink 1000, SR Research) was used during MEG data collection for Project 2 when available (corrective lenses interfere with eye-tracking). MEG data were collected at a sampling rate of 1000 Hz, DC coupled with a low-pass filter at 330 Hz. Continuous MEG raw data were collected in conjunction with continuous head position data (cHPI) for offline correction of head motion using Neuromag’s MaxFilter 2.1 MaxMove. Trials were rejected if magnetic activity was greater than 3000 femtotesla (fT) peak-to-peak in any channel. A subject-specific averaged eyeblink was identified and modeled using the signal space projection method (Uusitalo and Ilmoniemi 1997) within the Minimum Norm Estimates (MNE) Python software (Gramfort et al. 2014). Presentation software (Neurobehavioral Systems) was used for the presentation of stimuli.
An extensive neuropsychological battery was performed for each of the participants with neuroimaging data (same as the other projects). The battery of tests administered to SPs included: CPT-IP (Continuous Performance Test-Identical pairs), Quality of Life Scale, MATRICS (Measurement and Treatment Research to Improve Cognition in Schizophrenia), WASI (Wechsler Abbreviated Scale of Intelligence), WAIS-IV (Processing Speed Index of the Wechsler Adult Intelligence Scale-IV which includes symbol search and coding), WTAR (Wechsler Test of Adult Reading), and UCSD Performance-based Skills Assessment (UPSA/Psychosocial Functioning). HCs were administered the following tests: CPT-IP, MATRICS, WASI, WAIS-IV (symbol search, coding PSI) and WTAR.
In addition, our COBRE I studies included the collection of genetic material from subjects to evaluate the genetic contributions to neuroimaging findings in all the four projects and a pilot project. A number of genes have been associated with increased vulnerability for schizophrenia (Ripke et al. 2013). This genetic information adds a new and important dimension for the interpretation of endophenotypes revealed by neuroimaging studies. Saliva was collected and DNA was isolated and analyzed using genome wide genotyping chips. Saliva was collected from 100 SP and 100 HC subjects. Of these, 84 samples were genotyped using Illumina’s HumanOmni Quad 1 M Beadchips (Illumina, San Diego) spanning 1,140,419 SNPs and the rest were genotyped using Illumina’s HumanOmni Quad 5 M beadchips containing about 5 million SNPs and two samples were genotyped in both arrays. GenomeStudio software was used to make the final genotype calls. A series of quality control procedures following the recommendation by Anderson et al. (2010) was performed in PLINK (Purcell et al. 2007), including less than 5% per-sample missingness, samples’ gender match, heterozygosity within 3 SD, relatedness <0.18 (no 2nd degree or closer relatives), Hardy-Weinberg equilibrium <1 × 10−6, and minor allele frequency (MAF) > 0.05 (Chen et al. 2012). Although the vast majority of subjects self-reported racial status, we validated this information by computing the population structure from the genotyping data. Specifically, we applied PLINK multidimensional scaling analysis to our samples and used the HapMap3 as the reference. In addition, we used TaqMan® genotyping probes to assay another SNP not present in the 1 M chip, the rs1625579 in the miR-137 host gene.
Description of Tasks and Data Acquired
The first phase of Project 1 focused on basic pre-attentional (Task 1) and attentional processes (Task 2), whereas the second phase of the project focused on multisensory cognitive control (Task 4) and detection (Task 5). Task 3 was common to both phases of the project and involved a resting state fMRI scan. All fMRI/MRI data were collected on a 3.0 Tesla Siemens Trio scanner using a 12-channel head coil as described previously under the “MRI/fMRI data collection” section. MEG data were acquired for two of the tasks using an Elekta Neuromag 306 channel system as described under the “MEG data collection” section. See Table 1 for the number of SPs and HCs participating in each task. MEG task parameters are described below.
Project 1 Tasks 1 and 2
For Task 1 hemodynamic response functions were explicitly modeled for single (S1) and pairs (S1 + S2) of identical (IT; “gating-out” redundant information) or non-identical (NT; “gating-in” novel information) tones (Mayer et al. 2013). For Task 2, cues presented to the left or right ear (100 ms, 2000 Hz tone pip) correctly predicted the location of the targets (100 ms, 1000 Hz tone pip) in 50% of the trials (Abbott et al. 2012). Participants pressed a button with their right middle finger for tones on the right headphone and a button with their right index finger for tones on their left. The stimulus onset asynchrony (SOA) between cue and target was 200, 400, 600 or 800 ms, which was designed to effectively capture both facilitation (200 ms SOA) and inhibition of return or IOR (800 ms), as well as approximate the crossing of the two functions (400 and 600 ms SOA). For Task 2 the onset of a cue and the onset of the next cue ranged from 3 to 5 s at multiples of 1 s, with 240 trials acquired for each ISI (interstimulus interval). Cue and Target durations were 100 ms. The majority of participants who completed Tasks 1 and 2 also completed a resting state fMRI scan (Task 3), in which participants passively stared at a fixation cross for approximately 5 min. An MEG resting state scan (Task 3) utilized the same parameters as for the fMRI rest condition. A variant of Tasks 2 and 3 was also collected using MEG.
Project 1 Tasks 4 and 5
Forced Choice Multisensory Auditory/Visual Task [see (Stone et al. 2011) for a complete description]
The task for Project 2 was designed to investigate cortical disconnection in SP using fMRI and MEG measures from an auditory and visual multisensory task (EEG data were acquired simultaneously for a subset of SP and HC—these data are available upon request). The following 8 conditions were presented in random order to each participant: auditory alone (near/far = cond1/cond2), visual alone (near/far = cond3/cond4), synchronous auditory and visual (AV: near/far = cond5/cond6) and asynchronous AV (near/far = cond7/cond8). The near/far manipulation represented two different volume levels for the auditory stimuli (loud = near, quiet = far) and two different locations within the visual field with a soccer ball presented in a soccer field background (near = peripheral, far = foveal). The image of the black and white soccer ball appeared on the static soccer field background for 200 ms. For the V Near condition, the ball was centered at 8 degrees below fixation and subtended 2.7 degrees visual angle. For the V Far condition the soccer ball appeared at 1.8 degrees below fixation and subtended 1 degree of visual angle. For the auditory Near and Far conditions a 550 Hz 200 ms duration tone was presented at 63 and 45 dB SPL, respectively. The same stimuli were used for the multisensory conditions. In all cases the conditions were congruent (e.g., Near ball/Near tone). During 20% of the trials, feedback indicating a correct or incorrect response was provided to participants. The feedback was consistent with the trial condition (auditory, visual or combined) and consisted of either a crowd cheer and the soccer ball rolling into the goal for correct responses or a crowd groan and the soccer ball missing the goal for incorrect responses. For the MEG study, the ISI was randomized between 1500 and 2000 ms and all 8 conditions were randomized and presented in 6 separate blocks. Each participant was presented with 150 trials of each condition resulting in ~140 trials per condition after artifact rejections. An auditory threshold test was performed in the scanner prior to data collection to normalize the sound volume across participants. Please see Fig. 1 in (Stephen et al. 2013) or (Stone et al. 2014) for examples of the visual display used for data acquisition. Synchronous AV stimuli had a 5 ms lag between the presentation of the auditory and visual stimuli (with visual presented prior to auditory), whereas for the asynchronous AV condition the auditory stimulus lagged the visual stimulus by 50 ms (representing the natural delay experienced when a sound source is located approximately 50 ft. from the listener). The task was a simple forced choice task that required the participant to respond to all 8 randomized conditions by deciding whether the stimulus was “near” or “far” governed by the conditions stated above. As noted in Table 1, approximately 67 HC and 68 SP participated in each of the tasks during fMRI and MEG scanning.
Behavioral data were collected for both the fMRI and MEG tasks. Individual trial reaction times (RTs) and correct/incorrect responses were recorded. For fMRI data acquisition, 80 averages were acquired for each condition. An ORASI rest paradigm was used for resting MEG acquisition which consisted of 3 min eyes closed, 1 min eyes open, with instructions to blink at the beginning to allow for good characterization of eye blinks during the eyes open portion of the rest task. Raw and Maxfiltered MEG data (.fif format) are available. DICOM and preprocessed MRI data (through MRN’s SPM pipeline) are also available.
TP and Elemental Tasks [for a complete description of methods see (Hanlon et al. 2012)]
Across our studies using MEG and the verbal and nonverbal versions of the TP task, we have found lateralized hippocampal and PFC activation deficits in SP, as well as lower fronto-temporal anatomical connectivity which is related to working-relational memory performance deficits and worse every day functioning in SP (Hanlon et al. 2012, 2011, 2003, 2005b). Specifically, Hanlon et al. (2011) found that SP exhibited lower mean behavioral performance than HC on both the nonverbal and verbal version of TP in the MEG, with no decrement in performance on the non-hippocampal-dependent elemental (EL) control task versions. Fig. 4B displays an example of the hippocampal and PFC activation (shown in color) found in a HC using the standardized Low Resolution Brain Electromagnetic Tomography (sLORETA) analysis (Pascual-Marqui 2002, Wagner et al. 2004). HC showed more right hippocampal activation during nonverbal TP and more left hippocampal activation during verbal TP (Hanlon et al. 2011). This lateralized hippocampal activation was not seen in patients, who instead showed more bilateral or left hippocampal activation for both TP versions. We also found that SP exhibited more left PFC activation (BA 9 and 10) while HC exhibited more right PFC activation for both versions of TP (Fig. 4B). In addition, using COBRE data we examined the relationship between fronto-temporal anatomical connectivity with working/relational memory performance (nonverbal and verbal versions of TP) and everyday functioning (Hanlon et al. 2012). Fronto-temporal anatomical connectivity was assessed using dMRI measures of FA in the uncinate fasciculus. The tract based spatial statistics (TBSS/FSL) method was used to calculate participant specific skeletons (Smith et al. 2006). The skeleton was further restricted to uncinate fasciculus as defined by the JHU atlas (Wakana et al. 2004). Group differences were evaluated by comparing group mean values for FA over the uncinate fasciculus skeleton (Fig. 4D; green voxels; MNI z-coordinate indicated). The UPSA-2 (Patterson et al. 2001) was included to evaluate everyday functioning in patients. Results again showed that patients performed worse than HC on both the verbal and nonverbal versions of the TP task, but did not show a performance decrement on the verbal or nonverbal versions of the elemental (EL) control task (Fig. 4C). Also, FA in bilateral uncinate fasciculus was lower in SP compared to HC (Fig. 4D; red voxels). Finally, lower fronto-temporal anatomical connectivity (lower FA) was related to lower working-relational memory performance, and both predicted worse every day functioning (Hanlon et al. 2012).
The analysis of the P-FIT model of creativity has been studied in healthy individuals (Vakhtin et al. 2014) but has not yet been evaluated in the schizophrenia data. Multiple studies, however, using structural connectivity, 1H–MRS, and sMRI data, have been published as discussed below. Each COBRE investigator utilized structural and biochemical data acquired under Project 4 to enhance their existing projects (i.e., converging multimodal data); therefore, the focus here is on descriptions of the structural and 1H–MRS data that is available to others in terms of raw and preprocessed data.
Diffusion Tensor Imaging (DTI)
A DTI processing pipeline has been developed that was used by several studies at MRN that needed DTI analysis (Aine et al. 2011, Caprihan et al. 2011, 2015, Hanlon et al. 2012, Monnig et al. 2013, Wu et al. 2015) and by other research groups outside of MRN (Bessette et al. 2014, Haney-Caron et al. 2014). The preprocessing is primarily based on FSL software with some custom MATLAB programs for quality control. The preprocessing steps consisted of data quality check, motion and eddy current distortion correction, and correcting diffusion gradient directions (Caprihan et al. 2011). The output of the preprocessing step is then used for calculating scalar diffusion parameters, TBSS based analysis, source based morphometry (SBM), and structural connectivity analysis. The group studies were based on differences observed in tracts defined by an atlas (Aine et al. 2011), differences based on the skeleton as defined by the TBSS (Caprihan et al. 2015), a data driven method based on SBM of diffusion data to look for group differences in the loading coefficients (Caprihan et al. 2011), and whole brain connectivity matrix differences (Wu et al. 2015). The preprocessed data of subjects scanned under the COBRE project is available for other users through COINS.
Proton MR Spectroscopic Imaging (1H–MRS)
The structural T1-weighted images are freely available, and have been included with other datasets in the analysis of GM covariation patterns in schizophrenia (Gupta et al. 2015) and a world-wide meta-analysis of subcortical volumes in schizophrenia (van Erp et al. 2016). We have also combined the healthy controls from the COBRE study with multiple other data sets collected at MRN in order to evaluate the relationship between structural patterns of covariation and resting fMRI networks (Segall et al. 2012). The structural MRI from project 4 and functional MRI data from project 1 were also utilized in a schizophrenia classification challenge hosted on the kaggle.com competition site and garnered over 2000 entries [(Silva et al. 2014) more details are provided in the section below].
Sample Data-Driven Analyses of COBRE Data
In this section we describe three more recent endeavors using these data sets: 1) analysis of multi-task fMRI data across projects; 2) use of these data sets for an international signal processing competititon; and 3) examination of functional network connectivity (i.e., network structure and network dynamics) using resting state fMRI and MEG in SP and HC.
Multi-Task Analysis of fMRI Data across Project Tasks
Tenth Annual MLSP Competition—Schizophrenia Classification Challenge
Functional Network Connectivity in both fMRI and MEG
These datasets also provide a unique opportunity for fusing data of various types, besides combining data across modalities. The first study mentioned in this section (Cetin et al. 2014) can be considered as multi-task data fusion and the Silva et al. (2014) competition challenge paper requires data fusion across sMRI/fMRI. Stephen et al. (2013) and Sui et al. (2015) provide more examples of the richness of these datasets where MEG data is fused with DTI data, and where MATRICS responses are fused with fMRI/dMRI/sMRI.
Neuroimaging studies of schizophrenia have typically identified small differences between SP and HC in electrophysiological, structural, functional and metabolic brain measures. However, many of the functional/structural studies have focused on single task measures (or resting state), unisensory measures, or single modality measures (including volumetric) rather than utilizing complementary measures of brain function and structure that are needed to fully understand the complex brain dynamics that underlie this disorder. Project 1, which utilized several tasks and resting state, suggests that schizophrenia is characterized by inhibitory deficits that extend throughout the hemodynamic response function and across multiple tasks. Project 2 shows that although unisensory visual RTs were significantly slower for SP compared to HC, multisensory RTs (i.e., visual and auditory stimuli were presented together) did not differ by group, suggesting that the addition of the auditory stimulus helped to normalize the behavioral response in the SP group; the neurophysiological measures were consistent with this finding. Project 3 found lateralized hippocampal and PFC activation deficits in SP for nonverbal/verbal tasks, whereas a non-hippocampal elemental control task did not reveal differences between SP and HC. These results highlight the necessity of examining a hierarchy of tasks/functions across the same general set of patients/healthy controls as we did in our COBRE project. While use of rest fMRI data has become popular due to its ease of collection and use, the integration of (extended) rest fMRI data with task data is a largely understudied area. We have shown, using a hierarchy of tasks including rest, that schizophrenia patients exhibit consistent changes in connectivity from rest to tasks and in some cases show decreased connectivity at rest and increased connectivity during the most difficult tasks.
When examining linkages across modalities, in order to provide a better understanding of the structure/function networks that underlie the cognitive and social impairments experienced by SP, Project 2 results indicate that variations in MEG amplitude/timing were directly associated with alterations in WM integrity in SP versus HC. Similarly, Project 3 found lower fronto-temporal anatomical connectivity in SP, compared to HC, which was related to working-relational memory performance deficits and worse every day functioning in SP. Furthermore, our studies using multimodal functional data (e.g., MEG and fMRI), identified consistent functional components during rest, yet distinct patterns of functional network connectivity, results which emphasize the inadequacy of unimodal data collection and support the need to evaluate and integrate complex mental illness via multiple types of information. While the effect sizes for linking genome-wide individual genetic mutations with brain imaging results continue to be very small, the connection of these two modalities can give us important clues about the underlying mechanisms associated with schizophrenia. For example, multivariate association analyses can integrate multiple genetic risk loci’s effects on brain endophenotypes (Pearlson et al. 2015). Through these analyses, we were able to identify specific genetic contributions to structural abnormalities in patients with schizophrenia using a set of reference genes [Guided ICA (Chen et al. 2013) and PCA with reference (Gupta et al. 2015)]. Furthermore, we were able to demonstrate the contributions of a top schizophrenia risk gene expressing microRNA miR-137 (MIR137) and its targets to schizophrenia risk using pathways analyses and meta-analysis gene-set enrichment of variant associations (MAGENTA) (Wright et al. 2015, 2016) as well as gray matter concentration (Wright et al. 2016) and corpus callosum volume (Patel et al. 2015).
Ongoing advancements in computational power and algorithm development now allow us to capture some of this complexity across multiscale measures of brain function. For example, the use of dynamic (time-varying) connectivity (Calhoun et al. 2014) shows us that SPs do not always experience reduced connectivity; SP tend to spend less overall time in the more strongly connected states (Damaraju et al. 2014), which also correlates with the cortico-thalamic connectivity. Finally, we would like to emphasize the importance of data sharing and community competitions. The COBRE data competition, with over 2200 participants, demonstrated that structural and functional brain imaging is reliably predictive of schizophrenia. While this work is not yet complete, this represented one of the largest competitions demonstrating the potential for using brain imaging to inform us about diagnosis at the individual level.
Information Sharing Statement
Data sharing vehicle—COINS
This work was supported by a NIH COBRE Phase I grant (1P20RR021938, Lauriello, PI and 2P20GM103472, Calhoun, PI) awarded to the Mind Research Network. We wish to express our gratitude to numerous investigators who were either external consultants to the Cores and projects (Nancy Andreasen MD, Matti Hämäläinen PhD, Rik Henson PhD, Kent Hutchinson PhD, John Mosher PhD, Scott Sponheim PhD, Larry Wald PhD, Keith Worsley PhD); mentors on the projects (Robert Freedman MD, Kelvin Lim MD, Claudia Tesche, PhD), members of the external advisory committee (e.g., William Carpenter MD, John Lauriello MD, Godfrey Pearlson MD, Martin Reite MD, Peter Williamson MD) and members of the internal advisory committee (e.g., Samuel Keith MD, Yoshio Okada PhD, John Rasure PhD, Daniel Savage PhD). A special thank you goes to Margaret King for neuroinformatics management and COINS support.
- Abbott, C. C., Merideth, F., Ruhl, D., Yang, Z., Clark, V. P., Calhoun, V. D., Hanlon, F. M., & Mayer, A. R. (2012). Auditory orienting and inhibition of return in schizophrenia: a functional magnetic resonance imaging study. Prog Neuro-Psychopharmacol Biol Psychiatry, 37, 161–168.CrossRefGoogle Scholar
- Bockholt, H. J., Scully, M., Courtney, W., Rachakonda, S., Scott, A., Caprihan, A., Fries, J., Kalyanam, R., Segall, J. M., de la Garza, R., Lane, S., & Calhoun, V. D. (2010). Mining the mind research network: a novel framework for exploring large scale, heterogeneous translational neuroscience research data sources. Front Neuroinform, 3, 36.PubMedPubMedCentralGoogle Scholar
- Calhoun, V. D., & Sui, J. (2016). Multimodal fusion of brain imaging data: A key to finding the missing link(s) in complex illness. Biol Psychiatry: Cognitive Neuroscience and Neuroimaging, 1, 230–244.Google Scholar
- Caprihan, A., Jones, T., Chen, H., Lemke, N., Abbott, C., Qualls, C., Canive, J., Gasparovic, C., & Bustillo, J. R. (2015). The Paradoxical Relationship between White Matter, Psychopathology and Cognition in Schizophrenia: A Diffusion Tensor and Proton Spectroscopic Imaging Study. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology, 40, 2248–2257.CrossRefGoogle Scholar
- Cetin, M. S., Christensen, F., Abbott, C. C., Stephen, J. M., Mayer, A. R., Canive, J. M., Bustillo, J. R., Pearlson, G. D., & Calhoun, V. D. (2014). Thalamus and posterior temporal lobe show greater inter-network connectivity at rest and across sensory paradigms in schizophrenia. NeuroImage, 97, 117–126.PubMedPubMedCentralCrossRefGoogle Scholar
- Cetin, M. S., Houck, J. M., Rashid, B., Agcaoglu, O., Stephen, J. M., Sui, J., Canive, J., Mayer, A. R., Aine, C., Bustillo, J. R., & Calhoun, V. D. (2016). Multimodal classification of schizophrenia patients with MEG and fMRI data using static and dynamic connectivity measures. Front Neurosci, 10, 466.PubMedPubMedCentralCrossRefGoogle Scholar
- Chen, J., Calhoun, V. D., Pearlson, G. D., Perrone-Bizzozero, N., Sui, J., Turner, J. A., Bustillo, J. R., Ehrlich, S., Sponheim, S. R., Canive, J. M., Ho, B. C., & Liu, J. (2013). Guided exploration of genomic risk for gray matter abnormalities in schizophrenia using parallel independent component analysis with reference. NeuroImage, 83, 384–396.PubMedCrossRefGoogle Scholar
- Damaraju, E., Allen, E. A., Belger, A., Ford, J. M., McEwen, S., Mathalon, D. H., Mueller, B. A., Pearlson, G. D., Potkin, S. G., Preda, A., Turner, J. A., Vaidya, J. G., van Erp, T. G., & Calhoun, V. D. (2014). Dynamic functional connectivity analysis reveals transient states of dysconnectivity in schizophrenia. Neuroimage Clin, 5, 298–308.PubMedPubMedCentralCrossRefGoogle Scholar
- van Erp, T. G., Hibar, D. P., Rasmussen, J. M., Glahn, D. C., Pearlson, G. D., Andreassen, O. A., Agartz, I., Westlye, L. T., Haukvik, U. K., Dale, A. M., Melle, I., Hartberg, C. B., Gruber, O., Kraemer, B., Zilles, D., Donohoe, G., Kelly, S., McDonald, C., Morris, D. W., Cannon, D. M., Corvin, A., Machielsen, M. W., Koenders, L., de Haan, L., Veltman, D. J., Satterthwaite, T. D., Wolf, D. H., Gur, R. C., Gur, R. E., Potkin, S. G., Mathalon, D. H., Mueller, B. A., Preda, A., Macciardi, F., Ehrlich, S., Walton, E., Hass, J., Calhoun, V. D., Bockholt, H. J., Sponheim, S. R., Shoemaker, J. M., van Haren, N. E., Pol, H. E., Ophoff, R. A., Kahn, R. S., Roiz-Santianez, R., Crespo-Facorro, B., Wang, L., Alpert, K. I., Jonsson, E. G., Dimitrova, R., Bois, C., Whalley, H. C., McIntosh, A. M., Lawrie, S. M., Hashimoto, R., Thompson, P. M., & Turner, J. A. (2016). Subcortical brain volume abnormalities in 2028 individuals with schizophrenia and 2540 healthy controls via the ENIGMA consortium. Mol Psychiatry, 21, 585.PubMedCrossRefGoogle Scholar
- First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1995a). Structured clinical interview for DSM-IV axis I disorders, non-patient edition (SCID-NP), Version 2. New York: New York State Psychiatric Institute, Biometrics Research.Google Scholar
- First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1995b). Structured clinical interview for DSM-IV axis I disorders, patient edition (SCID-P), Version 2. New York: New York State Pscychiatric Institute, Biometrics Research.Google Scholar
- Freedman, R., Adler, L. E., Myles-Worsley, M., Nagamoto, H. T., Miller, C., Kisley, M., McRae, K., Cawthra, E., & Waldo, M. (1996). Inhibitory gating of an evoked response to repeated auditory stimuli in schizophrenic and normal subjects. Human recordings, computer simulation, and an animal model. Arch Gen Psychiatry, 53, 1114–1121.PubMedCrossRefGoogle Scholar
- Gupta, C. N., Calhoun, V. D., Rachakonda, S., Chen, J., Patel, V., Liu, J., Segall, J., Franke, B., Zwiers, M. P., Arias-Vasquez, A., Buitelaar, J., Fisher, S. E., Fernandez, G., van Erp, T. G., Potkin, S., Ford, J., Mathalon, D., McEwen, S., Lee, H. J., Mueller, B. A., Greve, D. N., Andreassen, O., Agartz, I., Gollub, R. L., Sponheim, S. R., Ehrlich, S., Wang, L., Pearlson, G., Glahn, D. C., Sprooten, E., Mayer, A. R., Stephen, J., Jung, R. E., Canive, J., Bustillo, J., & Turner, J. A. (2015). Patterns of Gray Matter Abnormalities in Schizophrenia Based on an International Mega-analysis. Schizophr Bull, 41, 1133–1142.PubMedCrossRefGoogle Scholar
- Guy, W. (1976). ECDEU Assessment manual for psychopharmacology: Publication ADM 76–338 (pp. 534–537). Washington, DC: US Department of Health, Education and Welfare.Google Scholar
- Hakak, Y., Walker, J. R., Li, C., Wong, W. H., Davis, K. L., Buxbaum, J. D., Haroutunian, V., & Fienberg, A. A. (2001). Genome-wide expression analysis reveals dysregulation of myelination-related genes in chronic schizophrenia. Proc Natl Acad Sci U S A, 98, 4746–4751.PubMedPubMedCentralCrossRefGoogle Scholar
- Hanlon, F. M., Houck, J. M., Klimaj, S. D., Caprihan, A., Mayer, A. R., Weisend, M. P., Bustillo, J. R., Hamilton, D. A., & Tesche, C. D. (2012). Frontotemporal anatomical connectivity and working-relational memory performance predict everyday functioning in schizophrenia. Psychophysiology, 49, 1340–1352.PubMedPubMedCentralCrossRefGoogle Scholar
- Hanlon, F. M., Shaff, N. A., Dodd, A. B., Ling, J. M., Bustillo, J. R., Abbott, C. C., Stromberg, S. F., Abrams, S., Lin, D. S., & Mayer, A. R. (2016). Hemodynamic response function abnormalities in schizophrenia during a multisensory detection task. Hum Brain Mapp, 37, 745–755.PubMedCrossRefGoogle Scholar
- Houck, J. M., Cetin, M. S., Mayer, A. R., Bustillo, J. R., Stephen, J., Aine, C. J., Canive, J., Perrone-Bizzozero, N., Thoma, R. J., Brookes, M. J., & Calhoun, V. D. (2017). Magnetoencephalographic and functional MRI connectomics in schizophrenia via intra- and inter-network connectivity. NeuroImage, 145(Pt A), 96–106.Google Scholar
- Huang, M. X., Edgar, J. C., Thoma, R. J., Hanlon, F. M., Moses, S. N., Lee, R. R., Paulson, K. M., Weisend, M. P., Irwin, J. G., Bustillo, J. R., Adler, L. E., Miller, G. A., & Canive, J. M. (2003). Predicting EEG responses using MEG sources in superior temporal gyrus reveals source asynchrony in patients with schizophrenia. Clin Neurophysiol, 114, 835–850.PubMedCrossRefGoogle Scholar
- Kalyanam, D., Boutte, C., Hutchinson, K. E., & Calhoun, V. D. (2015). Application of ICA to realistically simulated 1H-MRS data. Brain and Behavior, 5(7), e00345.Google Scholar
- Kraepelin, E. (1896). Lehrbuch der Psychiatrie. Leipzig: Barth.Google Scholar
- Kubicki, M., Westin, C. F., Maier, S. E., Frumin, M., Nestor, P. G., Salisbury, D. F., Kikinis, R., Jolesz, F. A., McCarley, R. W., & Shenton, M. E. (2002). Uncinate fasciculus findings in schizophrenia: a magnetic resonance diffusion tensor imaging study. Am J Psychiatry, 159, 813–820.PubMedPubMedCentralCrossRefGoogle Scholar
- Monnig, M. A., Caprihan, A., Yeo, R. A., Gasparovic, C., Ruhl, D. A., Lysne, P., Bogenschutz, M. P., Hutchison, K. E., & Thoma, R. J. (2013). Diffusion tensor imaging of white matter networks in individuals with current and remitted alcohol use disorders and comorbid conditions. Psychology of addictive behaviors : journal of the Society of Psychologists in Addictive Behaviors, 27, 455–465.CrossRefGoogle Scholar
- Orlovskaya, D., Vikhreva, O., Zimina, I. S., Denisov, D. V., & Uranova, N. A. (1999). Ultrastructural dystrophic changes of oligodendroglial cells in autopsied prefrontal cortex and striatum in schizophrenia: a morphometric study. Schizophr Res, 36, 82–83.Google Scholar
- Park, H. J., Westin, C. F., Kubicki, M., Maier, S. E., Niznikiewicz, M., Baer, A., Frumin, M., Kikinis, R., Jolesz, F. A., McCarley, R. W., & Shenton, M. E. (2004). White matter hemisphere asymmetries in healthy subjects and in schizophrenia: a diffusion tensor MRI study. NeuroImage, 23, 213–223.PubMedPubMedCentralCrossRefGoogle Scholar
- Patel, V. S., Kelly, S., Wright, C., Gupta, C. N., Arias-Vasquez, A., Perrone-Bizzozero, N., Ehrlich, S., Wang, L., Bustillo, J. R., Morris, D., Corvin, A., Cannon, D. M., McDonald, C., Donohoe, G., Calhoun, V. D., & Turner, J. A. (2015). MIR137HG risk variant rs1625579 genotype is related to corpus callosum volume in schizophrenia. Neurosci Lett, 602, 44–49.PubMedPubMedCentralCrossRefGoogle Scholar
- Pearlson, G. D., Liu, J., & Calhoun, V. D. (2015). An introductory review of parallel independent component analysis (p-ICA) and a guide to applying p-ICA to genetic data and imaging phenotypes to identify disease-associated biological pathways and systems in common complex disorders. Front Genet, 6, 276.PubMedPubMedCentralCrossRefGoogle Scholar
- Peirce, T. R., Bray, N. J., Williams, N. M., Norton, N., Moskvina, V., Preece, A., Haroutunian, V., Buxbaum, J. D., Owen, M. J., & O'Donovan, M. C. (2006). Convergent evidence for 2′,3′-cyclic nucleotide 3′-phosphodiesterase as a possible susceptibility gene for schizophrenia. Arch Gen Psychiatry, 63, 18–24.PubMedCrossRefGoogle Scholar
- Ripke, S., O'Dushlaine, C., Chambert, K., Moran, J. L., Kahler, A. K., Akterin, S., Bergen, S. E., Collins, A. L., Crowley, J. J., Fromer, M., Kim, Y., Lee, S. H., Magnusson, P. K., Sanchez, N., Stahl, E. A., Williams, S., Wray, N. R., Xia, K., Bettella, F., Borglum, A. D., Bulik-Sullivan, B. K., Cormican, P., Craddock, N., de Leeuw, C., Durmishi, N., Gill, M., Golimbet, V., Hamshere, M. L., Holmans, P., Hougaard, D. M., Kendler, K. S., Lin, K., Morris, D. W., Mors, O., Mortensen, P. B., Neale, B. M., O'Neill, F. A., Owen, M. J., Milovancevic, M. P., Posthuma, D., Powell, J., Richards, A. L., Riley, B. P., Ruderfer, D., Rujescu, D., Sigurdsson, E., Silagadze, T., Smit, A. B., Stefansson, H., Steinberg, S., Suvisaari, J., Tosato, S., Verhage, M., Walters, J. T., Multicenter Genetic Studies of Schizophrenia, C, Levinson, D. F., Gejman, P. V., Kendler, K. S., Laurent, C., Mowry, B. J., O'Donovan, M. C., Owen, M. J., Pulver, A. E., Riley, B. P., Schwab, S. G., Wildenauer, D. B., Dudbridge, F., Holmans, P., Shi, J., Albus, M., Alexander, M., Campion, D., Cohen, D., Dikeos, D., Duan, J., Eichhammer, P., Godard, S., Hansen, M., Lerer, F. B., Liang, K. Y., Maier, W., Mallet, J., Nertney, D. A., Nestadt, G., Norton, N., O'Neill, F. A., Papadimitriou, G. N., Ribble, R., Sanders, A. R., Silverman, J. M., Walsh, D., Williams, N. M., Wormley, B., Psychosis Endophenotypes International, C, Arranz, M. J., Bakker, S., Bender, S., Bramon, E., Collier, D., Crespo-Facorro, B., Hall, J., Iyegbe, C., Jablensky, A., Kahn, R. S., Kalaydjieva, L., Lawrie, S., Lewis, C. M., Lin, K., Linszen, D. H., Mata, I., McIntosh, A., Murray, R. M., Ophoff, R. A., Powell, J., Rujescu, D., Van Os, J., Walshe, M., Weisbrod, M., Wiersma, D., Wellcome Trust Case Control, C, Donnelly, P., Barroso, I., Blackwell, J. M., Bramon, E., Brown, M. A., Casas, J. P., Corvin, A. P., Deloukas, P., Duncanson, A., Jankowski, J., Markus, H. S., Mathew, C. G., Palmer, C. N., Plomin, R., Rautanen, A., Sawcer, S. J., Trembath, R. C., Viswanathan, A. C., Wood, N. W., Spencer, C. C., Band, G., Bellenguez, C., Freeman, C., Hellenthal, G., Giannoulatou, E., Pirinen, M., Pearson, R. D., Strange, A., Su, Z., Vukcevic, D., Donnelly, P., Langford, C., Hunt, S. E., Edkins, S., Gwilliam, R., Blackburn, H., Bumpstead, S. J., Dronov, S., Gillman, M., Gray, E., Hammond, N., Jayakumar, A., McCann, O. T., Liddle, J., Potter, S. C., Ravindrarajah, R., Ricketts, M., Tashakkori-Ghanbaria, A., Waller, M. J., Weston, P., Widaa, S., Whittaker, P., Barroso, I., Deloukas, P., Mathew, C. G., Blackwell, J. M., Brown, M. A., Corvin, A. P., McCarthy, M. I., Spencer, C. C., Bramon, E., Corvin, A. P., O'Donovan, M. C., Stefansson, K., Scolnick, E., Purcell, S., McCarroll, S. A., Sklar, P., Hultman, C. M., & Sullivan, P. F. (2013). Genome-wide association analysis identifies 13 new risk loci for schizophrenia. Nat Genet, 45, 1150–1159.PubMedPubMedCentralCrossRefGoogle Scholar
- Schechter, I., Butler, P. D., Zemon, V. M., Revheim, N., Saperstein, A. M., Jalbrzikowski, M., Pasternak, R., Silipo, G., & Javitt, D. C. (2005). Impairments in generation of early-stage transient visual evoked potentials to magno- and parvocellular-selective stimuli in schizophrenia. Clin Neurophysiol, 116, 2204–2215.PubMedPubMedCentralCrossRefGoogle Scholar
- Scott, A., Courtney, W., Wood, D., de la Garza, R., Lane, S., King, M., Wang, R., Roberts, J., Turner, J. A., & Calhoun, V. D. (2011). COINS: An Innovative Informatics and Neuroimaging Tool Suite Built for Large Heterogeneous Datasets. Front Neuroinform, 5, 33.PubMedPubMedCentralCrossRefGoogle Scholar
- Silva R, Castro E, Gupta N, Cetin M, Arbabshirani M, Potluru V, Plis SM, Calhoun VD, "The Tenth Annual MLSP Competition: Schizophrenia Classification Challenge," in IEEE International Workshop on Machine Learning for Signal Processing. IEEE Reims, France, Sept. 21-24, 2014.Google Scholar
- Smith, S. M., Jenkinson, M., Johansen-Berg, H., Rueckert, D., Nichols, T. E., MacKay, C. E., Watkins, K. E., Ciccarelli, O., Cader, M. Z., Mathews, P. M., & Behrens, T. E. J. (2006). Tract-based spatial statistics: Voxelwise analysis of multi-subject diffusion data. NeuroImage, 31, 1487–1505.PubMedCrossRefGoogle Scholar
- Talairach, J., & Tournoux, P. (1988). Co-lanar stereotaxic atlas of the human brain. New York: Thieme.Google Scholar
- Thoma, R. J., Hanlon, F. M., Moses, S. N., Edgar, J. C., Huang, M., Weisend, M. P., Irwin, J., Sherwood, A., Paulson, K., Bustillo, J., Adler, L. E., Miller, G. A., & Canive, J. M. (2003). Lateralization of auditory sensory gating and neuropsychological dysfunction in schizophrenia. Am J Psychiatry, 160, 1595–1605.PubMedCrossRefGoogle Scholar
- Wolf, R. C., Vasic, N., Sambataro, F., Hose, A., Frasch, K., Schmid, M., & Walter, H. (2009). Temporally anticorrelated brain networks during working memory performance reveal aberrant prefrontal and hippocampal connectivity in patients with schizophrenia. Prog Neuro-Psychopharmacol Biol Psychiatry, 33, 1464–1473.CrossRefGoogle Scholar
- Wright, C., Gupta, C. N., Chen, J., Patel, V., Calhoun, V. D., Ehrlich, S., Wang, L., Bustillo, J. R., Perrone-Bizzozero, N. I., & Turner, J. A. (2016). Polymorphisms in MIR137HG and microRNA-137-regulated genes influence gray matter structure in schizophrenia. Transl Psychiatry, 6, e724.PubMedPubMedCentralCrossRefGoogle Scholar
- Zuo, X. N., Anderson, J. S., Bellec, P., Birn, R. M., Biswal, B. B., Blautzik, J., Breitner, J. C., Buckner, R. L., Calhoun, V. D., Castellanos, F. X., Chen, A., Chen, B., Chen, J., Chen, X., Colcombe, S. J., Courtney, W., Craddock, R. C., Di Martino, A., Dong, H. M., Fu, X., Gong, Q., Gorgolewski, K. J., Han, Y., He, Y., He, Y., Ho, E., Holmes, A., Hou, X. H., Huckins, J., Jiang, T., Jiang, Y., Kelley, W., Kelly, C., King, M., LaConte, S. M., Lainhart, J. E., Lei, X., Li, H. J., Li, K., Li, K., Lin, Q., Liu, D., Liu, J., Liu, X., Liu, Y., Lu, G., Lu, J., Luna, B., Luo, J., Lurie, D., Mao, Y., Margulies, D. S., Mayer, A. R., Meindl, T., Meyerand, M. E., Nan, W., Nielsen, J. A., O'Connor, D., Paulsen, D., Prabhakaran, V., Qi, Z., Qiu, J., Shao, C., Shehzad, Z., Tang, W., Villringer, A., Wang, H., Wang, K., Wei, D., Wei, G. X., Weng, X. C., Wu, X., Xu, T., Yang, N., Yang, Z., Zang, Y. F., Zhang, L., Zhang, Q., Zhang, Z., Zhang, Z., Zhao, K., Zhen, Z., Zhou, Y., Zhu, X. T., & Milham, M. P. (2014). An open science resource for establishing reliability and reproducibility in functional connectomics. Sci Data, 1, 140049.PubMedPubMedCentralCrossRefGoogle Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.