1 Sample Information

Exploratory Data Analysis comparing the original molecular microscope data available at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE36059 and the synthetic balancing approaches with SMOTE, ADASYN and SMOTE-Tomek :

1.1 Original Data Set

  • Antibody Mediated Rejection (AMR): n = 65
  • Negative comparator (Non-AMR): n = 344

1.2 SMOTE Data Set

  • Antibody Mediated Rejection (AMR): n = 344
  • Negative comparator (Non-AMR): n = 344

1.3 ADASYN Data Set

  • Antibody Mediated Rejection (AMR): n = 353
  • Negative comparator (Non-AMR): n = 344

1.4 SMOTE-Tomek Data Set

  • Antibody Mediated Rejection (AMR): n = 344
  • Negative comparator (Non-AMR): n = 344

2 Principal component analysis

Two-dimensional PCA plot on the three data sets.

2.1 Original Data Set

## Loading required package: ggplot2

2.2 SMOTE data set

2.3 ADASYN data set

2.4 SMOTE-Tomek data set