Abstract
Imaging genetic analyses use heritability calculations to measure the fraction of phenotypic variance attributable to additive genetic factors. We tested the agreement between heritability estimates provided by four methods that are used for heritability estimates in neuroimaging traits. SOLAR-Eclipse and OpenMx use iterative maximum likelihood estimation (MLE) methods. Accelerated Permutation inference for ACE (APACE) and fast permutation heritability inference (FPHI), employ fast, non-iterative approximation-based methods. We performed this evaluation in a simulated twin-sibling pedigree and phenotypes and in diffusion tensor imaging (DTI) data from three twin-sibling cohorts, the human connectome project (HCP), netherlands twin register (NTR) and BrainSCALE projects provided as a part of the enhancing neuro imaging genetics analysis (ENIGMA) consortium. We observed that heritability estimate may differ depending on the underlying method and dataset. The heritability estimates from the two MLE approaches provided excellent agreement in both simulated and imaging data. The heritability estimates for two approximation approaches showed reduced heritability estimates in datasets with deviations from data normality. We propose a data homogenization approach (implemented in solar-eclipse; www.solar-eclipse-genetics.org) to improve the convergence of heritability estimates across different methods. The homogenization steps include consistent regression of any nuisance covariates and enforcing normality on the trait data using inverse Gaussian transformation. Under these conditions, the heritability estimates for simulated and DTI phenotypes produced converging heritability estimates regardless of the method. Thus, using these simple suggestions may help new heritability studies to provide outcomes that are comparable regardless of software package.
| Original language | English |
|---|---|
| Article number | 16 |
| Pages (from-to) | 1-11 |
| Number of pages | 11 |
| Journal | Frontiers in Neuroinformatics |
| Volume | 13 |
| Issue number | March |
| Early online date | 12 Mar 2019 |
| DOIs | |
| Publication status | Published - Mar 2019 |
Fingerprint
Keywords
- Computational methods
- DTI
- Genetics
- Heritability
- Imaging genetics
- Population
- Reproducability
Cite this
}
Homogenizing estimates of heritability among SOLAR-eclipse, OpenMX, APACE, and FPHI software packages in neuroimaging data. / Kochunov, Peter; Patel, Binish; Ganjgahi, Habib; Donohue, Brian; Ryan, Meghann; Hong, Elliot L.; Chen, Xu; Adhikari, Bhim; Jahanshad, Neda; Thompson, Paul M.; Van’t Ent, Dennis; den Braber, Anouk; de Geus, Eco J.C.; Brouwer, Rachel M.; Boomsma, Dorret I.; Hulshoff Pol, Hilleke E.; de Zubicaray, Greig I.; McMahon, Katie L.; Martin, Nicholas G.; Wright, Margaret J.; Nichols, Thomas E.
In: Frontiers in Neuroinformatics, Vol. 13, No. March, 16, 03.2019, p. 1-11.Research output: Contribution to Journal › Article › Academic › peer-review
TY - JOUR
T1 - Homogenizing estimates of heritability among SOLAR-eclipse, OpenMX, APACE, and FPHI software packages in neuroimaging data
AU - Kochunov, Peter
AU - Patel, Binish
AU - Ganjgahi, Habib
AU - Donohue, Brian
AU - Ryan, Meghann
AU - Hong, Elliot L.
AU - Chen, Xu
AU - Adhikari, Bhim
AU - Jahanshad, Neda
AU - Thompson, Paul M.
AU - Van’t Ent, Dennis
AU - den Braber, Anouk
AU - de Geus, Eco J.C.
AU - Brouwer, Rachel M.
AU - Boomsma, Dorret I.
AU - Hulshoff Pol, Hilleke E.
AU - de Zubicaray, Greig I.
AU - McMahon, Katie L.
AU - Martin, Nicholas G.
AU - Wright, Margaret J.
AU - Nichols, Thomas E.
PY - 2019/3
Y1 - 2019/3
N2 - Imaging genetic analyses use heritability calculations to measure the fraction of phenotypic variance attributable to additive genetic factors. We tested the agreement between heritability estimates provided by four methods that are used for heritability estimates in neuroimaging traits. SOLAR-Eclipse and OpenMx use iterative maximum likelihood estimation (MLE) methods. Accelerated Permutation inference for ACE (APACE) and fast permutation heritability inference (FPHI), employ fast, non-iterative approximation-based methods. We performed this evaluation in a simulated twin-sibling pedigree and phenotypes and in diffusion tensor imaging (DTI) data from three twin-sibling cohorts, the human connectome project (HCP), netherlands twin register (NTR) and BrainSCALE projects provided as a part of the enhancing neuro imaging genetics analysis (ENIGMA) consortium. We observed that heritability estimate may differ depending on the underlying method and dataset. The heritability estimates from the two MLE approaches provided excellent agreement in both simulated and imaging data. The heritability estimates for two approximation approaches showed reduced heritability estimates in datasets with deviations from data normality. We propose a data homogenization approach (implemented in solar-eclipse; www.solar-eclipse-genetics.org) to improve the convergence of heritability estimates across different methods. The homogenization steps include consistent regression of any nuisance covariates and enforcing normality on the trait data using inverse Gaussian transformation. Under these conditions, the heritability estimates for simulated and DTI phenotypes produced converging heritability estimates regardless of the method. Thus, using these simple suggestions may help new heritability studies to provide outcomes that are comparable regardless of software package.
AB - Imaging genetic analyses use heritability calculations to measure the fraction of phenotypic variance attributable to additive genetic factors. We tested the agreement between heritability estimates provided by four methods that are used for heritability estimates in neuroimaging traits. SOLAR-Eclipse and OpenMx use iterative maximum likelihood estimation (MLE) methods. Accelerated Permutation inference for ACE (APACE) and fast permutation heritability inference (FPHI), employ fast, non-iterative approximation-based methods. We performed this evaluation in a simulated twin-sibling pedigree and phenotypes and in diffusion tensor imaging (DTI) data from three twin-sibling cohorts, the human connectome project (HCP), netherlands twin register (NTR) and BrainSCALE projects provided as a part of the enhancing neuro imaging genetics analysis (ENIGMA) consortium. We observed that heritability estimate may differ depending on the underlying method and dataset. The heritability estimates from the two MLE approaches provided excellent agreement in both simulated and imaging data. The heritability estimates for two approximation approaches showed reduced heritability estimates in datasets with deviations from data normality. We propose a data homogenization approach (implemented in solar-eclipse; www.solar-eclipse-genetics.org) to improve the convergence of heritability estimates across different methods. The homogenization steps include consistent regression of any nuisance covariates and enforcing normality on the trait data using inverse Gaussian transformation. Under these conditions, the heritability estimates for simulated and DTI phenotypes produced converging heritability estimates regardless of the method. Thus, using these simple suggestions may help new heritability studies to provide outcomes that are comparable regardless of software package.
KW - Computational methods
KW - DTI
KW - Genetics
KW - Heritability
KW - Imaging genetics
KW - Population
KW - Reproducability
UR - http://www.scopus.com/inward/record.url?scp=85064277755&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85064277755&partnerID=8YFLogxK
U2 - 10.3389/fninf.2019.00016
DO - 10.3389/fninf.2019.00016
M3 - Article
VL - 13
SP - 1
EP - 11
JO - Frontiers in Neuroinformatics
JF - Frontiers in Neuroinformatics
SN - 1662-5196
IS - March
M1 - 16
ER -