Using Mendelian Randomization and Colocalization to Elucidate the Causal Relationship between Graves' diseases and Neurodenegrative diseases
Integrative Analysis to explore the shared genetic architecture between Graves’ Disease (GD) and Multiple Sclerosis (MS)
This work implements an integrative genomic framework to investigate shared and distinct genetic architecture between Graves’ disease (GD) and Multiple Sclerosis (MS). Using large-scale GWAS summary statistics, the pipeline combines genetic correlation analysis (LDSC), Mendelian randomization, and Bayesian colocalization and fine mapping analysis to disentangle global genetic correlation, test potential causal relationships, and identify shared or disease-specific risk loci.
The workflow is designed to support reproducible cross-trait inference in autoimmune diseases and can be adapted to other complex traits.
- Determine genome-wide genetic correlation between GD and MS
- Test potential causal relationships using two-sample Mendelian randomization
- Identify shared versus distinct causal variants via fine-mapping–based colocalization
- Distinguish pleiotropy from confounding in cross-disease genetic associations
- GWAS summary statistics
- Linkage Disequilibrium Score Regression (LDSC)
- Two-sample Mendelian Randomization (IVW and MR-PRESSO)
- Bayesian colocalization using coloc.abf
- Fine-mapping using SuSiER -
coloc.susie - Visualization in R
- Biological interpretation of shared loci
All analyses are fully reproducible using the scripts provided in this repository. The workflow used in this study can be executed stepwise once appropriate GWAS summary statistics are supplied by the user.
Software versions, and parameter settings are specified in the article to ensure consistent results across computational environment.
We used publicly available GWAS summary statistics for Graves’ disease (https://opengwas.io/datasets/ebi-a-GCST90018847) and Multiple Sclerosis (https://opengwas.io/datasets/finn-b-G6_MS). All data sources are cited within the manuscript and can be accessed from their original repositories.
The pipeline generates:
- Genome-wide genetic correlation estimates and heritability
- Mendelian randomization effect estimates (primarily - IVW) with sensitivity diagnostics
- Locus-level posterior probabilities from Bayesian colocalization
- Locus level finemapping from credible sets
- Summary tables.
This project is licensed under the MIT License. See the LICENSE file for details.
For questions or collaboration inquiries, please contact:
Adeyemi Timothy Akinade
Email: [akinadeadeyemi1610@gmail.com; aakinad@clemson.edu]