Title: | Regional Vulnerability Index |
Version: | 0.3.3 |
Description: | The Regional Vulnerability Index (RVI), a statistical measure of brain structural abnormality, quantifies an individual's similarity to the expected pattern (effect size) of deficits in schizophrenia (Kochunov P, Fan F, Ryan MC, et al. (2020) <doi:10.1002/hbm.25045>). |
License: | GPL-2 | GPL-3 [expanded from: GPL (≥ 2)] |
Encoding: | UTF-8 |
LazyData: | true |
RoxygenNote: | 7.3.2 |
Depends: | R (≥ 2.10) |
Imports: | stats |
NeedsCompilation: | no |
Packaged: | 2025-03-11 02:29:53 UTC; AbestSG |
Maintainer: | Si Gao <Si.Gao@uth.tmc.edu> |
Author: | Si Gao [aut, cre, dtc] (RVIpkg method development, code development, maintenance, data collection), Peter Kochunov [aut, dtc, cph, fnd] (RVIpkg concept, method development, data collection and funding), Kathryn Hatch [ctb, dtc] (RVIpkg testing and data collection), Yizhou Ma [ctb, dtc] (RVIpkg testing, method development and data collection), Fatima Talib [ctb] (RVIpkg testing and method) |
Repository: | CRAN |
Date/Publication: | 2025-03-11 16:00:02 UTC |
Optimizing data from UK Biobank
Description
The Ave_func() can optimize data from UK Biobank(UKB). It will rename field IDs of regional neuroimaging traits to abbreviation names, and then average data of left and right hemispheres of the same field.
Usage
Ave_func(resp.range, type = "all", data)
Arguments
resp.range |
a numeric vector specifying column range of regional neuroimaging traits. |
type |
a character string specifying data types of regional neuroimaging traits(i.e. All traits(type='all'), White matter(type='WM'),Gray matter(type='GM') or Subcortical(type='Subcortical')) |
data |
a data frame contains regional neuroimaging traits with field IDs from UKBB. Default(type='all') |
Value
a dataframe of regional neuroimaging traits with abbreviated field names.
Note
The Ave_func() function is developed at the Maryland Psychiatric Research Center, Department of Psychiatry, University of Maryland School of Medicine. This project is supported by NIH R01 EB015611 grant. Please cite our funding if you use this software.
References
Kochunov P, Fan F, Ryan MC, et al. Translating ENIGMA schizophrenia findings using the regional vulnerability index: Association with cognition, symptoms, and disease trajectory (2020). Hum Brain Mapp. 2020;10.1002/hbm.25045. doi:10.1002/hbm.25045
R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
Expected patterns of gray matter
Description
the expected patterns of gray matter in schizophrenia derived from large-scale meta-analyses by the ENIGMA.
Usage
EP.GM
Format
A data frame with 33 rows and 13 variables:
- GM
grey matter thickness
- SSD
the expected pattern of Schizophrenia Spectrum Disorder (doi:10.1016/j.biopsych.2018.04.023 table S4a)
- MDD
the expected pattern of Major Depressive Disorder (doi:10.1038/mp.2016.60 table 1)
- AD_ADNI
the expected pattern of Alzheimer's Disease (doi:10.1016/j.nicl.2021.102574 table S1)
- AD_ADNIOSYRIX
the expected pattern of Alzheimer's Disease Neuroimaging Initiative + OSIRIX
- BD
the expected pattern of Bipolar Disorder (doi:10.1038/mp.2017.73 table 1)
- PD
the expected pattern of Parkinson's disease
- Diabetes
the expected pattern of Diabetes
- HighBP
the expected pattern of Hypertension
- HighLipids
the expected pattern of Hyperlipidemia
- MET
the expected pattern of Metabolic diseases (Diabetes, Hypertension and Hyperlipidemia)
- DS_22q
the expected pattern of 22q11.2 deletion syndrome (doi:10.1038/s41380-018-0078-5 table S10a)
- Suicide
the expected pattern of suicidal ideation (doi:10.1101/2021.09.27.21264068 table S8)
- OCD_pediatric
the expected pattern of Obsessive-compulsive disorder in pediatric subjects (doi:10.1176/appi.ajp.2017.17050485 table S6)
- OCD_adult
the expected pattern of Obsessive-compulsive disorder in adults (doi:10.1176/appi.ajp.2017.17050485 table S4)
- AN
the expected pattern of Anorexia Nervosa
Expected patterns of subcortical
Description
the expected patterns of subcortical in schizophrenia derived from large-scale meta-analyses by the ENIGMA.
Usage
EP.Subcortical
Format
A data frame with 8 rows and 13 variables:
- Subcortical
subcortical grey matter volume
- SSD
the expected pattern of Schizophrenia Spectrum Disorder (doi:10.1038/mp.2015.63 table 1)
- MDD
the expected pattern of Major Depressive Disorder (doi:10.1038/mp.2015.69 table 1)
- AD_ADNI
the expected pattern of Alzheimer's Disease (doi:10.1016/j.nicl.2021.102574 table S1)
- AD_ADNIOSYRIX
the expected pattern of Alzheimer's Disease Neuroimaging Initiative + OSIRIX
- BD
the expected pattern of Bipolar Disorder (doi:10.1038/mp.2015.227 table 1)
- PD
the expected pattern of Parkinson's disease
- Diabetes
the expected pattern of Diabetes
- HighBP
the expected pattern of Hypertension)
- HighLipids
the expected pattern of Hyperlipidemia
- MET
the expected pattern of Metabolic diseases (Diabetes, Hypertension and Hyperlipidemia)
- DS_22q
the expected pattern of 22q11.2 deletion syndrome (doi:10.1176/appi.ajp.2019.19060583 table S14a)
- Suicide
the expected pattern of suicidal ideation (doi:10.1101/2021.09.27.21264068 table S8)
- OCD_pediatric
the expected pattern of Obsessive-compulsive disorder in pediatric subjects (doi:10.1176/appi.ajp.2016.16020201 table S2)
- OCD_adult
the expected pattern of Obsessive-compulsive disorder in adults (doi:10.1176/appi.ajp.2016.16020201 table 3)
- AN
the expected pattern of Anorexia Nervosa
Expected patterns of white matter
Description
the expected patterns of white matter in schizophrenia derived from large-scale meta-analyses by the ENIGMA.
Usage
EP.WM
Format
A data frame with 24 rows and 14 variables:
- WM
white matter fractional anisotropy
- SSD
the expected pattern of Schizophrenia Spectrum Disorder (doi:10.1038/mp.2017.170 table 1; doi:10.1002/hbm.24998 table 2)
- MDD
the expected pattern of Major Depressive Disorder (doi:10.1038/s41380-019-0477-2 Table S4; doi:10.1002/hbm.24998 table 2)
- AD_ADNI
the expected pattern of Alzheimer's Disease (doi:10.1016/j.nicl.2021.102574 table S1)
- AD_ADNIOSYRIX
the expected pattern of Alzheimer's Disease Neuroimaging Initiative + OSIRIX
- BD
the expected pattern of Bipolar Disorder (doi:10.1002/hbm.24998 table 2)
- Diabetes
the expected pattern of Diabetes
- HighBP
the expected pattern of Hypertension
- HighLipids
the expected pattern of Hyperlipidemia
- MET
the expected pattern of Metabolic diseases (diabetes, hypertension and hyperlipidemia)
- DS_22q
the expected pattern of 22q11.2 deletion syndrome (doi:10.1002/hbm.24998 table 2; doi:10.1038/s41380-019-0450-0 table S6)
- PTSD
the expected pattern of Post-traumatic stress disorder (doi:10.1002/hbm.24998 table 2)
- TBI
the expected pattern of Traumatic brain injury (doi:10.1002/hbm.24998 table 2)
- OCD_pediatric
the expected pattern of Obsessive-compulsive disorder in pediatric subjects (doi:10.1038/s41398-021-01276-z table 4)
- OCD_adult
the expected pattern of Obsessive-compulsive disorder in adults (doi:10.1038/s41398-021-01276-z table 3)
Regional Vulnerability Index
Description
The Regional Vulnerability Index (RVI), a statistical measure of brain structural abnormality, quantifies an individual’s similarity to the expected pattern (effect size) of deficits seen in schizophrenia derived from large-scale meta-analyses by the ENIGMA consortium. This package outputs the inverse-normal transformed (INT) residuals, z-normalized INT residuals, RVI and Alignment Vulnerability Index (AVI).
Usage
RVI_func(
ID,
DXcontrol,
covariates = NULL,
resp.range,
EP,
sign = FALSE,
fisherZ = FALSE,
data
)
Arguments
ID |
a column name of subject IDs in data. |
DXcontrol |
a character string specifying control subset(i.e. DXcontrol='DX==0'or DXcontrol='DX=="CN"'). Mean and standard deviation of z-normalization should be calculated in healthy controls. |
covariates |
an optional character vector specifying column names of covariates (i.e. Age, Sex). If covariates=NULL (the default), residuals will not be adjusted for any covariate. If covariates are specified (i.e. covariates=c('Age','Sex')), residuals will be adjusted for covariates. |
resp.range |
a numeric vector specifying column indices of regional neuroimaging traits. |
EP |
a numeric vector specifying an expected pattern of regional neuroimaging traits. The expected patterns(EP.WM, EP.GM and EP.Subcortical) for white matter fractional anisotropy (FA), cortical matter thickness and subcortical volume are included in the package (Note: If you use an expected pattern, you need to make sure the order of regional neuroimaging traits in your data match up the corresponding order of the expected pattern). The patterns can be extract in the package (i.e. RVIpkg::EP.WM$SSD, RVIpkg::EP.WM$MDD, RVIpkg::EP.WM$AD, RVIpkg::EP.WM$BD ,RVIpkg::EP.WM$PD .etc.). They were developed using neuroimaging data of UK Biobank (UKBB). |
sign |
a logical value indicating whether the AVI use signs from RVI. |
fisherZ |
a logical value indicating whether the result should generate fisher-z transformed RVI. |
data |
a data frame contains a column of subject IDs, a column of controls, columns of covariates, columns of responses. |
Details
The RVI is developed as a simple measure of agreement between an individual's pattern of regional neuroimaging traits and the expected pattern of schizophrenia. First, all observations of each regional neuroimaging trait are regressed out optional covariates using linear regression, and then residuals are extracted from the model after removing effects of the optional covariates. The optional covariates could be age, sex, intracranial brain volume and/or .etc in the data. After that the residuals are inverse-normal transformed based on residuals' ranks, and then the INT residuals are z-normalized/standardized using mean and standard deviation of healthy controls to get z-normalized INT residuals. For each subject, the RVI is then calculated as a Pearson correlation coefficient between the z-normalized INT residuals of the traits and corresponding expected pattern of the traits and the AVI is the dot product of the z-normalized INT residuals of the traits and corresponding expected pattern of the traits. These expected patterns include cortical thickness, subcortical volume, and white matter FA for mental illnesses and metabolic diseases.
Value
A list with the following elements:
i.norm.resid |
INT residuals |
z.norm.resid |
z-normalized/standardized INT residuals |
RVI |
RVI: the Pearson correlation coefficient between the z-normalized INT residuals and corresponding expected pattern; AVI: the dot product of the z-normalized INT residuals and corresponding expected pattern; RVI.fisherz: Fisher z-transformed RVI |
Note
The RVI_func() function is developed at the Maryland Psychiatric Research Center, Department of Psychiatry, University of Maryland School of Medicine. This project is supported by NIH R01 EB015611 grant. Please cite our funding if you use this software.
References
Kochunov P, Fan F, Ryan MC, et al. Translating ENIGMA schizophrenia findings using the regional vulnerability index: Association with cognition, symptoms, and disease trajectory (2020). Hum Brain Mapp. 2020;10.1002/hbm.25045. doi:10.1002/hbm.25045
R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
Examples
EP1 <- c(-0.37,0.31,-0.02,-0.08,-0.21,0.46,0.31,0.25)
RVI1 <- RVI_func(ID='ID', DXcontrol='DX==0', covariates=c('Age','Sex'), resp.range=c(5:12),
EP=EP1, data=RVIpkg::data)
RVI2 <- RVI_func(ID='ID', DXcontrol='DX==0', covariates=NULL, resp.range=c(5:12),
EP=EP1, data=RVIpkg::data)
EP2 <- RVIpkg::EP.Subcortical$SSD
RVI3 <- RVI_func(ID='ID', DXcontrol='DX==0', covariates=c('Age','Sex'), resp.range=c(5:12),
EP=EP2, data=RVIpkg::data)
Simulated volumes of subcortical structures
Description
Simulated volumes of subcortical structures of Schizophrenia spectrum disorder are used as an example for this function. You can calculate RVI for this dataset
Usage
data
Format
A data frame with 196 rows and 12 variables:
- ID
subjects' ID
- DX
indicators of control group
- Age
subjects' age
- Sex
subjects' gehder
- Lateral.Ventricle
simulated volumes of lateral Ventricle
- Thalamus
simulated volumes of Thalamus
- Caudate
simulated volumes of Caudate
- Putamen
simulated volumes of Putamen
- Pallidum
simulated volumes of Pallidum
- Hippocampus
simulated volumes of Hippocampus
- Amygdala
simulated volumes of Amygdala
- Left.Accumbens.area
simulated volumes of left Accumbens
Source
The 'data' dataset is from Maryland Psychiatric Research Center, Department of Psychiatry, University of Maryland School of Medicine.