Here is the Github page that has the codes for the bioinformatics pipeline described in the paper COVID-Datathon: Biomarker identification for COVID-19 severity based on BALF scRNA-seq data.
The scBALF_Monocle.md file from the Monocle folder explains how one can extract the list of potential COVID-19 severity biomarkers using Monocle. The R markdown files are also available in the same folder.
In the second step of our proposed pipeline we run multiple linear and non-linear machine learning classification algorithms to perform cell classification ( One lable vs the rest). Here we explain how one can run these classification algorithms:
- Lindear Discriminant Aanlysis (LDA):
The LDA.md file from the LDA folder explains how one can run perform cell classification using LDA. The R code files are also available in the same folder.
- Quadratic Discriminant Aanlysis (QDA):
The QDA.md file from the QDA folder explains how one can run perform cell classification using QDA. The R code files are also available in the same folder.
- Fleaxible Discriminant Aanlysis (FDA):
The FDA.md file from the FDA folder explains how one can run perform cell classification using FDA. The R code files are also available in the same folder.
- Support Vector Machine (SVM) with RBF kernerl
The SVM.md file from the SVM folder explains how one can run perform cell classification using SVM. The R code files are also available in the same folder.
- Random Forest (RF)
The RF.md file from the RF folder explains how one can run perform cell classification using RF. The R code files are also available in the same folder.