Skip to content

S-LDSC Analysis

I applied Stratified Linkage Disequilibrium Score Regression1 (S-LDSC) to summary statistics from Betham et al.'s GWAS of lupus2.

Reference Data Sources

I used the standard reference datasets prepared by the authors of the S-LDSC method1.

Results

GTEx and Franke lab tissue expression data

The plot below shows the results of the application of S-LDSC to Bentham et al.'s study using the GTEx and Franke lab gene expression datasets. In the plot, the x-axis corresponds to cell type, while the y-axis corresponds to \(-\log_{10}(p)\). Points are colored according to broad tissue category. Large points correspond to cell/tissue types deemed significant by an application of the Benjamini-Hochberg procedure at an FDR of 0.013.

As would be expected from an immunological disease like lupus, the top tissue types are all immune-related.

Roadmap Chromatin data

I next applied S-LDSC using the reference dataset derived from the Roadmap epigenetic project. The results are in the plot below:

Again, as would be expected of an immune condition, immune-related cell types rank highly. Somewhat surprisingly, the signal for T-cells is stronger than the B-cell signal, despite lupus being considered a primarily immunoglobulin-driven disease.

ImmGen data

Next, I applied S-LDSC using reference data from the ImmGen project.

There were no significant cell types.

The cell types with the lowest p values are given in the table below:

Name Coefficient Coefficient_P_value Reject Null
MF.11cloSer.Salm3.SI 1.11878e-07 0.0001879 False
DC.8-4-11b-.SLN 1.19332e-07 0.000208158 False
NKT.44-NK1.1-.Th 9.40865e-08 0.000465922 False
DC.103+11b-.Lu 9.82035e-08 0.000488146 False
DC.8-.Th 1.11345e-07 0.000564316 False
DC.8+.Sp.ST 8.85064e-08 0.00164973 False
DC.IIhilang-103-11b+.SLN 9.31931e-08 0.00178242 False
NKT.4-.Lv 8.2538e-08 0.00211748 False

Its surprising that no cell types were found to be significant in this analysis, given lupus's status as an immune-mediated disease. To speculate, this may be due to differences between the mouse and human immune systems.

Corces et al. ATAC-seq data

The results of applying S-LDSC using the epigenetic reference data from Corces et al. ATAC-seq analysis of blood cells are shown below.

Name Coefficient Coefficient_P_value Reject Null
Bcell 1.53276e-06 6.87243e-06 True
NK 1.38321e-06 0.00385197 False
CD8 1.26043e-06 0.0155016 False
CD4 1.02203e-06 0.023981 False
CLP 6.16263e-07 0.120099 False
Mono 4.66559e-07 0.247512 False
CMP 1.60724e-07 0.298654 False
MPP 1.39026e-07 0.331436 False
GMP 1.20103e-07 0.349802 False
LMPP 1.32611e-07 0.360171 False
MEP 9.50534e-08 0.373571 False
HSC 7.32517e-08 0.411087 False
Erythro -9.56772e-08 0.569965 False

There is one significant cell type: B-cells. This is highly consistent with known status of lupus as a disease mediated by immunoglobulin.

Cahoy and GTEx-Brain data

The next two reference datasets pertain to the nervous system. The results of running S-LDSC with these two datasets are shown below:

Name Coefficient Coefficient_P_value Reject Null
Oligodendrocyte -9.80353e-10 0.517435 False
Astrocyte -8.08454e-09 0.637238 False
Neuron -1.87689e-08 0.804622 False
Name Coefficient Coefficient_P_value Reject Null
Brain_Spinal_cord_(cervical_c-1) 5.77991e-08 0.00452981 False
Brain_Frontal_Cortex_(BA9) 1.90353e-08 0.139255 False
Brain_Substantia_nigra 1.94392e-08 0.20111 False
Brain_Cerebellum 9.35862e-09 0.344303 False
Brain_Cerebellar_Hemisphere 9.03005e-09 0.351129 False
Brain_Putamen_(basal_ganglia) 5.30034e-09 0.384532 False
Brain_Cortex 4.89259e-09 0.405465 False
Brain_Amygdala 3.51586e-09 0.42135 False
Brain_Hypothalamus 1.05231e-09 0.478304 False
Brain_Anterior_cingulate_cortex_(BA24) -2.08092e-09 0.548644 False
Brain_Nucleus_accumbens_(basal_ganglia) -1.57096e-08 0.791281 False
Brain_Hippocampus -1.64556e-08 0.804464 False
Brain_Caudate_(basal_ganglia) -3.65828e-08 0.978487 False

There are no significant cell types. This is consistent with lupus being primarily non-neurological.

  1. Hilary K Finucane, Yakir A Reshef, Verneri Anttila, Kamil Slowikowski, Alexander Gusev, Andrea Byrnes, Steven Gazal, Po-Ru Loh, Caleb Lareau, Noam Shoresh, and others. Heritability enrichment of specifically expressed genes identifies disease-relevant tissues and cell types. Nature Genetics, 50(4):621–629, 2018. URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC5896795/. ↩↩

  2. James Bentham, David L Morris, Deborah S Cunninghame Graham, Christopher L Pinder, Philip Tombleson, Timothy W Behrens, Javier Martín, Benjamin P Fairfax, Julian C Knight, Lingyan Chen, and others. Genetic association analyses implicate aberrant regulation of innate and adaptive immunity genes in the pathogenesis of systemic lupus erythematosus. Nature Genetics, 47(12):1457–1464, 2015. URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC4668589/. ↩

  3. Yoav Benjamini and Yosef Hochberg. Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal statistical society: series B (Methodological), 57(1):289–300, 1995. URL: https://rss.onlinelibrary.wiley.com/doi/abs/10.1111/j.2517-6161.1995.tb02031.x. ↩