Skip to content

S-LDSC Analysis of DecodeME

Stratified Linkage Disequilibrium Score Regression (S-LDSC)1 was applied to summary statistics from GWAS-1 of DecodeME2.

Reference Data Sources

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

Results

GTEx and Franke lab tissue expression data

The plot and table below show the results of the application of S-LDSC to DecodeME 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. The table shows details of the cell/tissues types with the smallest p values.

s-ldsc-decodme-gene-expression

Name Coefficient Coefficient_P_value Reject Null
A08.186.211.Brain 4.72676e-09 2.47819e-07 True
A08.186.211.730.885.287.500.571.735.Visual.Cortex 4.53115e-09 1.56363e-06 True
A08.186.211.464.405.Hippocampus 4.49163e-09 2.19491e-06 True
Brain_Cortex 3.66873e-09 2.6085e-06 True
A08.186.211.730.885.287.500.270.Frontal.Lobe 4.63607e-09 3.13473e-06 True
Brain_Nucleus_accumbens_(basal_ganglia) 3.59733e-09 3.36967e-06 True
Brain_Frontal_Cortex_(BA9) 3.69246e-09 3.7678e-06 True
A08.186.211.730.885.287.500.Cerebral.Cortex 4.17782e-09 6.18484e-06 True
A08.186.211.464.710.225.Entorhinal.Cortex 4.14964e-09 6.71693e-06 True
A08.186.211.464.Limbic.System 4.20837e-09 7.01006e-06 True
Brain_Substantia_nigra 3.51471e-09 8.8185e-06 True
Brain_Anterior_cingulate_cortex_(BA24) 3.26788e-09 1.0814e-05 True
Brain_Amygdala 3.19633e-09 2.91008e-05 True
Brain_Hippocampus 3.25752e-09 3.53736e-05 True
Brain_Caudate_(basal_ganglia) 3.10888e-09 7.75616e-05 True
Brain_Putamen_(basal_ganglia) 3.10587e-09 0.000104439 True
A08.186.211.730.317.357.352.435.Hypothalamo.Hypophyseal.System 3.61124e-09 0.000166442 True
A08.186.211.730.317.357.Hypothalamus 3.24167e-09 0.00023368 True
A08.186.211.730.885.287.500.670.Parietal.Lobe 3.4577e-09 0.000240607 True
Brain_Hypothalamus 2.84899e-09 0.000356557 True
A08.186.211.730.317.Diencephalon 2.9906e-09 0.000463437 True
Brain_Cerebellar_Hemisphere 2.66524e-09 0.000644924 True
A08.186.211.730.885.287.249.Basal.Ganglia 2.69897e-09 0.0010573 True
A08.186.211.865.428.Metencephalon 2.44467e-09 0.0019363 False
A08.186.211.730.885.287.249.487.Corpus.Striatum 2.2671e-09 0.00453931 False
A08.186.211.653.Mesencephalon 2.44809e-09 0.00517217 False
Brain_Cerebellum 2.21062e-09 0.00541441 False
Brain_Spinal_cord_(cervical_c-1) 2.08238e-09 0.00603588 False

As we saw in our earlier MAGMA analysis using the GTEx dataset, the significant tissues are all CNS-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 and table below:

s-ldsc-roadmap-chromatin

Name Coefficient Coefficient_P_value Reject Null
Fetal_Brain_Female__DNase 1.02689e-07 4.68419e-10 True
Brain_Dorsolateral_Prefrontal_Cortex__H3K27ac 5.08217e-08 1.23002e-09 True
Fetal_Brain_Male__DNase 9.55596e-08 2.73287e-09 True
Fetal_Brain_Male__H3K4me1 3.78077e-08 5.22621e-09 True
Brain_Dorsolateral_Prefrontal_Cortex__H3K4me3 1.70485e-07 5.98839e-09 True
Fetal_Brain_Female__H3K4me1 6.20252e-08 1.37543e-08 True
Brain_Anterior_Caudate__H3K27ac 3.86611e-08 3.06201e-08 True
Brain_Inferior_Temporal_Lobe__H3K27ac 3.30624e-08 2.87328e-07 True
Brain_Cingulate_Gyrus__H3K9ac 8.70527e-08 3.38485e-07 True
Brain_Angular_Gyrus__H3K9ac 8.39789e-08 9.75539e-07 True
Brain_Dorsolateral_Prefrontal_Cortex__H3K4me1 5.21768e-08 1.72804e-06 True
Brain_Germinal_Matrix__H3K4me3 1.26424e-07 1.85932e-06 True
Brain_Inferior_Temporal_Lobe__H3K4me3 1.03363e-07 3.9248e-06 True
Brain_Anterior_Caudate__H3K4me3 1.02781e-07 4.68e-06 True
Fetal_Brain_Female__H3K4me3 1.26751e-07 5.12867e-06 True
Brain_Angular_Gyrus__H3K4me3 1.29751e-07 9.79109e-06 True
Brain_Dorsolateral_Prefrontal_Cortex__H3K9ac 9.32505e-08 9.83818e-06 True
Brain_Cingulate_Gyrus__H3K4me3 1.06995e-07 1.02402e-05 True
Brain_Angular_Gyrus__H3K4me1 4.08889e-08 1.39444e-05 True
Brain_Angular_Gyrus__H3K27ac 3.26885e-08 1.89114e-05 True
Brain_Anterior_Caudate__H3K4me1 3.54259e-08 3.15633e-05 True
Brain_Anterior_Caudate__H3K9ac 6.35599e-08 3.21619e-05 True
Brain_Cingulate_Gyrus__H3K27ac 2.92786e-08 3.245e-05 True
Brain_Cingulate_Gyrus__H3K4me1 3.2249e-08 4.08821e-05 True
Cortex_derived_primary_cultured_neurospheres__H3K4me3 1.30409e-07 5.87376e-05 True
Ganglion_Eminence_derived_primary_cultured_neurospheres__H3K4me3 9.46707e-08 8.30653e-05 True
Brain_Inferior_Temporal_Lobe__H3K9ac 5.55865e-08 8.69272e-05 True
Brain_Hippocampus_Middle__H3K4me3 6.50022e-08 0.000226226 True
Brain_Inferior_Temporal_Lobe__H3K4me1 3.38614e-08 0.00032817 True
Brain_Hippocampus_Middle__H3K4me1 1.98882e-08 0.000355901 True
Fetal_Brain_Female__H3K36me3 3.17864e-08 0.000550213 True
Brain_Germinal_Matrix__H3K4me1 4.83297e-08 0.000608534 True
skeletal_muscle_ENTEX__H3K4me1 1.17311e-08 0.000895252 False
Brain_Substantia_Nigra__H3K4me3 7.29864e-08 0.00121435 False
Placenta_Amnion__H3K36me3 2.5235e-08 0.00158191 False
Brain_Dorsolateral_Prefrontal_Cortex__H3K36me3 3.5113e-08 0.00163492 False
Brain_Hippocampus_Middle__H3K27ac 1.91901e-08 0.00173697 False
Ganglion_Eminence_derived_primary_cultured_neurospheres__H3K4me1 3.16805e-08 0.00215425 False

Again, the strongest and most significant associations are all with CNS cell-types.

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 shown in the table below:

Name Coefficient Coefficient_P_value Reject Null
DC.8-4-11b+.MLN 2.98171e-09 0.00459158 False
T.4.Pa.BDC 2.5939e-09 0.00506716 False
T.8Mem.LN 2.85485e-09 0.00730256 False
DC.8-4-11b-.SLN 2.65688e-09 0.0116649 False
preT.DN2-3.Th 2.41341e-09 0.0121964 False
T.4int8+.Th 2.53226e-09 0.012314 False
LN.TR.14w.B6 2.52753e-09 0.0135406 False
preT.ETP-2A.Th 2.50404e-09 0.0172496 False
MF.103-11b+.Lu 2.36504e-09 0.0197861 False
MF.Microglia.CNS 2.40608e-09 0.0224495 False
B.Mem.Sp.v2 2.06099e-09 0.026437 False
Tgd.vg5-.act.IEL 2.08167e-09 0.0328614 False
B.MZ.Sp 2.08844e-09 0.0348465 False
T.8Eff.Sp.OT1.d15.LisOva 1.80262e-09 0.0357771 False
T.4SP69+.Th 1.91234e-09 0.0394321 False

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. There are no significant cell types:

Name Coefficient Coefficient_P_value Reject Null
Erythro 1.83516e-08 0.154592 False
Bcell 4.60401e-09 0.351883 False
CLP 4.52542e-09 0.381204 False
Mono -1.51999e-09 0.530117 False
MEP -1.5488e-09 0.566514 False
CD4 -4.57403e-09 0.628276 False
CD8 -6.41332e-09 0.669675 False
LMPP -9.21299e-09 0.724051 False

Cahoy and GTEx-Brain data

The next two reference datasets pertain to the nervous system.

Surprisingly, when we analyze the DecodeME results using the Cahoy dataset, the neuron cell type is not even close to being significant. This is discordant with some of the results above, in which many CNS-related cell and tissue types were marked as significant. Moreover, the oligodendrocyte cell type is closer to being significant than the neuron cell type.

Name Coefficient Coefficient_P_value Reject Null
Oligodendrocyte 2.12389e-09 0.0294642 False
Neuron 1.20321e-09 0.104572 False
Astrocyte -9.45929e-10 0.825497 False

When we apply the S-LDSC using the GTEx brain dataset, we find the the cortex tissue type is significant:

Name Coefficient Coefficient_P_value Reject Null
Brain_Cortex 3.27938e-09 0.000106241 True
Brain_Frontal_Cortex_(BA9) 2.09994e-09 0.00675003 False
Brain_Anterior_cingulate_cortex_(BA24) 1.50154e-09 0.0256637 False
Brain_Nucleus_accumbens_(basal_ganglia) 1.14427e-09 0.112026 False
Brain_Cerebellum 8.97215e-10 0.182957 False
Brain_Hippocampus 3.3308e-10 0.341041 False
Brain_Putamen_(basal_ganglia) 3.2917e-10 0.359427 False
Brain_Cerebellar_Hemisphere 2.45686e-10 0.39188 False

How to Reproduce This

To reproduce, run the DecodeME Analysis Script.

  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. Genetics Delivery Team, Thibaud Boutin, Andrew D Bretherick, Joshua J Dibble, Esther Ewaoluwagbemiga, Emma Northwood, Gemma L Samms, Veronique Vitart, Project, Cohort Delivery Team, Øyvind Almelid, and others. Initial findings from the DecodeME genome-wide association study of myalgic encephalomyelitis/chronic fatigue syndrome. medRxiv, pages 2025–08, 2025. URL: https://www.medrxiv.org/content/10.1101/2025.08.06.25333109v1. ↩

  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. ↩