Skip to content

Latest commit

 

History

History
64 lines (55 loc) · 2.64 KB

README.md

File metadata and controls

64 lines (55 loc) · 2.64 KB
Raw

IMPORTANT

ALLCatchR was further developed to predict Ph-pos (BCR:ABL1) subcluster

please see https://github.com/ThomasBeder/ALLCatchR_bcrabl1

old ALLCatchR

The gene expression classifier ALLCatchR identifies B-precursor ALL subtypes and underlying developmental trajectories

ALLCatchR was developed to predict:

  • 21 BCP-ALL molecular subtypes (BCL2/MYC, CDX2/UBTF, CEBP, DUX4, ETV6::RUNX1, ETV6::RUNX1-like, HLF, Hyperdiploid, iAMP21, IKZF1 N159Y, KMT2A, Low hypodiploid, MEF2D, Near haploid, NUTM1 PAX5 P80R, PAX5alt, Ph-like, Ph-pos, TCF3::PBX1, ZNF384)
  • Associations to B lymphopoiesis stages based on gene set enrichment analyses
  • Blast Count percentage
  • Immunophenotype
  • Patient's sex

image

Publication

Beder T, Hansen BT, Hartmann AM, et al. A Multi-Cohort Gene Expression Classifier (ALLCatchR) Identifies B-Precursor ALL Subtypes and Their Developmental Trajectories across Age Groups. Blood. 2022;140(Supplement 1):1053–1055. (https://doi.org/10.1182/blood-2022-163311)

Installation

open RStudio install devtools and follow the installion guide https://github.com/r-lib/devtools

if (!require("devtools", quietly = TRUE))
  install.packages("devtools")

install ALLCatchR

devtools::install_github("ThomasBeder/ALLCatchR")

Quickstart

If Counts.file is left NA ten test samples are predicted

library(ALLCatchR)
allcatch()

Run ALLCatchR

As input ALLCatchR requires a single text file in which the first column represent the genes and the other columns the count data for each sample

library(ALLCatchR)
allcatch(Counts.file = NA, ID_class = "symbol", sep = "\t")
# Counts.file: /path/to/your/count/data, if left empty a test
# ID_class: gene names can be either "symbol", "ensemble_ID" or	"entrez_ID"
# sep: seperator of the text file usually "\t", "," or ";"

output

ALLCatchR writes a predictions.tsv file to your current working directory with the following columns:

  • sample: Sample ID
  • Score: BCP-ALL subtype prediction score
  • Prediction: Predicted subtype
  • Confidence: Confidence of subtype predictions, IMPORTANT predictions of samples labeled as unclassified here should be considered unclassified
  • BlastCounts: Predicted Blast Count percentage
  • Sex: Patient's sex predition
  • Score_sex: Sex prediciton probability
  • Immuno: Immunophenotype prediction
  • ScoreImmuno: Immunophenotype prediciton probability
  • 10-16: Enrichment analysis results (singscore, https://github.com/DavisLaboratory/singscore) to gene sets defined for B-cell progenitors
  • 37-37: SVM linear predictions for individual subtypes
  • 38-59: Results from gene set based nearest neighbor analysis