Simulated GTEx-style cis-eQTL example

Simulated cis-eQTL data inspired by published GTEx case studies (AHCYL1, SCD, HSP90AA1, STMN2, ABHD17A): a functional response on 256 gene-body positions, three causal SNPs with localized peak effects at distinct 5', mid-body, and 3' positions. Used by vignette("fsusie_gtex_case_study").

Usage

gtex_example

Format

A list with components

X

n x p genotype matrix (p = 200) sliced from susieR::N3finemapping$X.

Y

n x T log1p-coverage matrix (T = 256).

pos

length-T integer vector of gene-body positions.

causal_snps

integer vector of column indices in X that carry true effects.

causal_betas

length(causal_snps) x T matrix of true per-position effects.

description

free-text description.

Source

Simulated. See data-raw/make_data.R.

Examples

# \donttest{
data(gtex_example)
fit <- fsusie(gtex_example$Y, gtex_example$X,
              pos = gtex_example$pos, L = 15, L_greedy = 5,
              verbose = TRUE)
#> HINT: ncol(Y) is not 2^J or positions are unevenly spaced; interpolated to a regular dyadic grid.
#> iter          ELBO       delta   sigma2      mem      V  extras
#>    1   -205841.8728           -   [0.998, 0.998, 1.000]   0.20 GB  [1.75e-02, 1.16e-02, 9.48e-03, 1.01e-03, 0 x 1]  pi_null=[0.92, 1.00]
#> iter   2: max|d(alpha,PIP)|=1.60e-01, V=[1.63e-02, 1.50e-02, 9.98e-03, 1.38e-03, 0 x 1] [mem: 0.20 GB]
#> iter   3: max|d(alpha,PIP)|=1.70e-01, V=[1.61e-02, 1.51e-02, 9.99e-03, 1.33e-03, 0 x 1] [mem: 0.20 GB]
#> iter   4: max|d(alpha,PIP)|=1.65e-01, V=[1.60e-02, 1.51e-02, 1.00e-02, 1.22e-03, 0 x 1] [mem: 0.20 GB]
#> iter   5: max|d(alpha,PIP)|=1.56e-01, V=[1.58e-02, 1.51e-02, 1.00e-02, 1.13e-03, 0 x 1] [mem: 0.20 GB]
#> iter   6: max|d(alpha,PIP)|=1.46e-01, V=[1.57e-02, 1.51e-02, 1.00e-02, 1.05e-03, 0 x 1] [mem: 0.20 GB]
#> iter   7: max|d(alpha,PIP)|=3.00e-02, V=[1.56e-02, 1.51e-02, 1.00e-02, 9.84e-04, 0 x 1] [mem: 0.20 GB]
#> iter   8: max|d(alpha,PIP)|=8.12e-02, V=[1.53e-02, 1.51e-02, 1.00e-02, 9.47e-04, 0 x 1] [mem: 0.20 GB]
#> iter   9: max|d(alpha,PIP)|=8.00e-02, V=[1.45e-02, 1.51e-02, 1.00e-02, 7.18e-04, 0 x 1] [mem: 0.20 GB]
#> iter  10: max|d(alpha,PIP)|=7.13e-02, V=[1.40e-02, 1.51e-02, 1.00e-02, 7.15e-04, 0 x 1] [mem: 0.20 GB]
#> iter  11: max|d(alpha,PIP)|=1.86e-05, V=[1.40e-02, 1.51e-02, 1.00e-02, 7.15e-04, 0 x 1] -- converged (alpha_pip_fixed_point) [mem: 0.20 GB]
#> [L_greedy] 1 round, greedy_lbf_cutoff=0.100, final L=5
#> round  L     min(lbf)   action
#> 1      5     0.000      saturated
fit_s <- mf_post_smooth(fit, method = "TI")
mfsusie_plot(fit_s)

# }