This repository contains scripts of the R code used to generate the results in our paper "Biological heterogeneity in idiopathic pulmonary arterial hypertension identified through unsupervised transcriptomic profiling of whole blood".
- An Intel-compatible platform running Windows 10 /8.1/8 /7 /Vista /XP /2000 Windows Server 2019 /2016 /2012 /2008 /2003
- At least 256 MB of RAM, a mouse, and enough disk space for recovered files, image files, etc.
- The administrative privileges are required to install and run R‑Studio utilities.
- A network connection for data recovering over network.
- Tested on RStudio, Version 1.2.5042
This step includes the specific preprocessing for our paper "Biological heterogeneity in idiopathic pulmonary arterial hypertension identified through unsupervised transcriptomic profiling of whole blood".
Inputs:
- RNA-sequencing file (genes x patients) :
rnaseq_data.xlsx - Clinical variable file (patients x variables) :
clinical_data.xlsx
Outputs:
- Pre clustering ready file :
pre_clustering_p_all_tpm.RDS
Run-time for 300 genes and 359 patients: <= 5 seconds
For thi step you can use the demo dataset demo_pre_clustering_p_all_tpm.RDS and gene list demo_sorted_variant_genes.RDS, provided in this repository.
Inputs:
- Pre clustering ready file :
pre_clustering_p_all_tpm.RDS - Number of subgroups (k)
- Number of most variant genes to be used: (top_genes)
Outputs:
- Gene list sorted based on variance(descending order) :
sorted_variant_genes.RDS - Subgroup memberships for patients :
memberships_k5.RDS&memberships_k5.csv
Run-time for 300 genes and 359 patients: <= 10 sec
Please contact Sokratis Kariotis (Biosok) through Github for queries relating to this code.
Shield:
This work is licensed under a Creative Commons Attribution 4.0 International License.
