PDSclassifier

PDSclassifier R package provides a pathway-based molecular classification system for colorectal cancer (CRC), which can be applied to gene expression profiles to stratify into three Pathway-Derived Subtype (PDS): PDS1, PDS2 and PDS3, with distinct molecular biology.

Installation

You can install the development version of PDSclassifier like so:

if(!requireNamespace("devtools", quietly = TRUE))
  install.packages("devtools")
devtools::install_github('sidmall/PDSclassifier')

Usage

An example where using the testData from the R package, PDS classification can be made with PDSpredict() function.

library(PDSclassifier)
pds_calls <- PDSpredict(testData, species = 'human', threshold = 0.6)

PDSclassifier can be applied to both human and mouse transcriptomic data with parameter: species = c("human", "mouse"). The default prediction probability threshold = 0.6 has been set. The PDS prediction probability ranges between 0 to 1. The recommendation is be stay between 0.5-0.7 to retain enough samples without losing underlying biology that defines PDS.

Additionally, calculateSMI() function enable users to determine if the given sample (bulk tissue or single-cell) is transcriptionally stem-like or differentiated-like with Stem Maturation Index (SMI). The outcome provides single sample gene set enrichment analysis (ssGSEA) score for MYC targets and PRC targets, and from these, SMI is calculate (provided unscaled and scaled (-1 to 1)).

smi_data <- calculateSMI(as.matrix(testdata[,-1]), datatype = "bulk", species = "human")

Citation

The PDSclassifier R package is developed by the DunneLab. Please contact Sudhir Malla (Sid) for any package related issues, questions, or suggestions.

You can cite as such:

Malla, S.B., Byrne, R.M., Lafarge, M.W. et al., Pathway level subtyping identifies a slow-cycling biological phenotype associated with poor clinical outcomes in colorectal cancer. Nat Genet 56, 458–472 (2024). https://doi.org/10.1038/s41588-024-01654-5