Skip to content

Latest commit

 

History

9 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 

Repository files navigation

Mendelian Randomization Analysis of Periodontal Disease

Causal inference study identifying genetic risk factors for chronic and acute periodontitis — published as a Master's thesis at the University of South Florida (2024).

📄 Read the full thesis


Overview

This study uses two-sample Mendelian Randomization (MR) to evaluate whether five modifiable exposures causally influence periodontal disease risk. By using SNPs as genetic instrumental variables, MR sidesteps the confounding and reverse causation problems that plague observational studies.

Exposures tested:

  • Smoking status
  • Maternal smoking after birth
  • Plasminogen levels
  • Creatinine levels
  • Absence of psychosocial stress

Outcomes:

  • Chronic periodontitis
  • Acute periodontitis

Key Findings

Exposure Outcome Direction Significance
Creatinine levels Acute periodontitis Positive ↑ Significant
Maternal smoking Periodontitis risk Positive ↑ Significant
Plasminogen levels Periodontitis risk Negative ↓ Significant
Smoking status Chronic periodontitis Positive ↑ Borderline

Highlights:

  • Creatinine → acute periodontitis link supports the hypothesis that renal dysfunction amplifies systemic inflammation
  • Maternal smoking effect aligns with prenatal nicotine exposure literature
  • Plasminogen's negative association diverges from prior pro-inflammatory models — flagged as a priority for follow-up research

Methods

All analyses were conducted in R using a two-sample MR framework.

MR methods:

  • Inverse Variance Weighted (IVW) — primary estimate
  • MR-Egger — tests and corrects for directional pleiotropy
  • Weighted Median — robust when up to 50% of IVs are invalid

Instrumental variable selection:

  • Genome-wide significance threshold (P < 5×10⁻⁸)
  • LD pruning (r² threshold applied)
  • Allele harmonization across exposure and outcome GWAS

Sensitivity analyses:

  • Leave-one-out analysis
  • Cochran's Q heterogeneity test
  • Funnel plots and scatter plots for visual diagnostics

Repository Structure

Mendelian-Randomization/
└── TwoSampleMR.R       # Full MR pipeline: IV selection, harmonization, analysis, sensitivity checks

Requirements

install.packages(c("TwoSampleMR", "MRInstruments", "tidyverse"))

Citation

Charles, R. (2024). Unravelling the Impact of Blood Metabolites and Lifestyle Factors on Periodontal Disease Using Mendelian Randomization. Master's Thesis, University of South Florida. https://digitalcommons.usf.edu/etd/10605


Contact

Rhea Charles · riocx1997@gmail.com · LinkedIn · GitHub

About

The codes for my Mendelian Randomization thesis

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages