Epigenetic Aging Clocks for Canines

Using DNA methylation (DNAm) data and regression models to develop epigenetic clocks that can estimate canine age.
This was performed as part of a research training grant challenge, led by UCLA’s Institute for Quantitative and Computational Biosciences (QCBio) and supported by NIH’s Big Data to Knowledge (BD2K) initiative

Content

• canineAge_dataMatrix.txt.gz Input file, with sample methylation levels at CpG sites
• canineAge_regressionModeling.ipynb Jupyter notebook with data analysis workflow

  1. Data Preparation
  2. PCA Feature Reduction
  3. Linear Regression
  4. Hyperparameter Tuning
  5. Elastic Net Regression
  6. Bootstrapping for Enhanced Variable Selection
  7. Distance-Weighted Prediction Interpolation
  8. Age Estimates Output