Skip to content

Latest commit

 

History

History
49 lines (37 loc) · 1.31 KB

README.md

File metadata and controls

49 lines (37 loc) · 1.31 KB

Files

sampleAligner

Use the helper function created in this file to align the methylation data with the patient information and cell fractions.

groupProbes

Finds the most significant probes for different group stratification.

dmctFinder

Calculates the differentially methylated cell-type specific probes based on cell fraction using CellDMC.

cellAge

Uses the DMCTs to calculate the epigenetic ages of each cell type across all samples.

Workflow

1. Load Data

  • Call the fileAligner to get both the patient information file and the methylation data for the respective patients.

2. Distinguish Phentotype Groups

  • Separate into the the groups of each phenotype you wish to investigate the cell type fractions for.

3. Run CellDMC

  • Input the methylation data (n_samples, n_features), list of phenotypes (n_samples), and fraction (n_samples x n_features).

4. Extract Cell-specific CpG Sites

  • For each cell type collect their corresponding DMCTs if present.

5. Calculate Coefficients

  • Run through elastic net regularization model to find coefficients.

6. Calculate Cell Age

  • Run coefficients & methylation data through Horvath Clock.

Requirements

  • ggplot2
  • magrittr
  • dplyr
  • ggbeeswarm
  • broom
  • ggpubr
  • patchwork
  • data.table
  • EpiDISH
  • readr
  • data.table
  • ISLR
  • caret
  • glmnet