Seymour Lopez
Data scientist

My Top Projects

Project 1: Differential gene expression analysis that investigates if a polymer prevents retinal scarring

Here I use DeSeq2 to extract differential expressed (DE) genes in epithelial cells treated under four conditions: Media only control, Polymer alone, TNF-α and TGF-β1 (TNT) alone, TNT + Polymer. The purpose of this pipeline is to replicate the results published by Parikh et al., 2022, which showed that upregulation of genes that are important for the NRF2 signalling pathway is associated with the presence of the polymer and not necessarily TNT. Skills used here:

Project 2: Exploring if changes in cortical structures predict neurodegenerative diagnosis

This project was to perform exploratory analysis on the ADNI dataset. Skills used here:

  • Investigating the relationship between confounding variables and features
  • Removing confounding effects using regression
  • Predictive modelling using logistic regression and random forest

Project 3: Data preprocessing for uncovering epilepsy subtypes

Here I clean and preprocess data in order to perform disease subtyping and staging of patients. Skills used here:

  • Imputing missing values
  • Harmonising data across multiple sources
  • Feature engineering to reduce data dimensions