Psych_LDSC: Psychiatric and educational attainment Genomic SEM LDSC...

In umx: Structural Equation Modeling and Twin Modeling in R

Psychiatric and educational attainment Genomic SEM LDSC dataset.

Description

A dataset containing the genetic covariance structure from LD score regression (LDSC) for 5 psychiatric and educational attainment phenotypes: SCZ, BIP, MDD, EA, and INSOM. This is the object shape returned by GenomicSEM / umxGSEM_ldsc() and the example covstruc used throughout the GSEM helpers (see also Anthro_LDSC).

Usage

data(Psych_LDSC)

Format

A list with elements S, V, I, N, and m (see Details).

Details

LDSC list fields (same S/V/I/N/m layout as GenomicSEM ldsc; V names are OpenMx residual labels):

• S: 5 x 5 genetic covariance matrix among the phenotypes (diagonals are SNP heritabilities on the LDSC scale).

• V: 15 x 15 sampling covariance of the unique elements of S (OpenMx vech order; dimnames like var_SCZ, poly_BIP_SCZ). Used as asymptotic covariance for WLS/DWLS. Not GenomicSEM SCZ_SCZ / BIP_SCZ.

• I: 5 x 5 LDSC intercept matrix. Diagonal = per-trait LDSC intercepts (expected near 1 without confounding). Off-diagonal = cross-trait intercepts reflecting sample overlap. Preferred input for GC in umxGSEM_GWAS().

• N: 1 x 15 effective sample sizes for each unique S element.

• m: Number of SNPs used in LDSC.

See Also

Anthro_LDSC, umxGSEM(), umxGSEM_GWAS(), umxGSEM_ldsc()

Other GSEM: Anthro_LDSC, plot.umx_GWAS(), umxGSEM(), umxGSEM_GWAS(), umxGSEM_dl_tutorial_files(), umxGSEM_label_ldsc(), umxGSEM_ldsc(), umxGSEM_munge(), umxGSEM_std(), umxGSEM_sumstats(), umxSummary.MxDataStatic(), umxSummary.list()

Other datasets: Anthro_LDSC, Fischbein_wt, GFF, HSwls, docData, iqdat, umx, us_skinfold_data

Examples

data(Psych_LDSC)
str(Psych_LDSC, max.level = 1)
names(Psych_LDSC) # "V" "S" "I" "N" "m"
# Pretty-print S, I, V, ... (see also umxSummary on WLS mxData)
umxSummary(Psych_LDSC)
# I = LDSC intercept matrix (diag ~ 1; off-diag = sample overlap)
round(Psych_LDSC$I, 3)
# Common factor of all five traits (Grotzinger et al. 2019 style DWLS GSEM)
## Not run: 
m1 = umxGSEM(model = "g ~= SCZ + BIP + MDD + EA + INSOM",
             covstruc = Psych_LDSC, estimation = "DWLS")
umxSummary(m1)

## End(Not run)

umx documentation built on Oct. 11, 2026, 5:07 p.m.