This repository contains codes for the manuscript titled Modeling Type 1 Diabetes progression using machine-learning and single-cell transcriptomic measurements in human islets.
A schematic workflow of ML-based XGBoost model was built for gene selection and classification.
R v 4.1.2 was used to develop the models. All the package versions can be found in the manuscript.
Some of the key packages include-
install.packages("Seurat")
install.packages("xgboost")
install.packages("e1071")
install.packages("foreach")
install.packages("doParallel")
and others.
The processed scRNA-seq dataset (Seurat .RDS object) is uploaded here: https://hpap.pmacs.upenn.edu/analysis.
The Machine Learning model scripts folder contains the scripts for cell types, unannotated, and LOOCV analysis.
The libraries folder contains the dependencies for the above scripts for generating the data either at cell type level or all cells (unannotated).
The Differential Expression Analysis folder contains the scripts for performing differential expression analysis on either all cells (unannotated) or Pseudobulk.
This folder contains all the scripts used to generate the figures and tables in the manuscript.
Patil et al. (2024) Modeling Type 1 Diabetes progression using machine-learning and single-cell transcriptomic measurements in human islets. Cell Reports Medicine 2024 (Accepted in principle). Link/DOI will be updated soon.