TMZEP
TMZEP is developed by Wang-Lab@HKUST.
* These fields are required.
Welcome to TMZep!
Overview
TMZ efficacy predictor (TMZep) is a tool for predicting GBM patient's response to Temozolomide (TMZ) based on their molecular features. TMZep was trained on IDH1 wild-type primary GBM samples. Users can input gene expression, somatic mutation and copy number alteration data to predict if a sample is sensitive or resistant to TMZ.
Predict sensitive or resistant?
Contact Us
| DISCLAIMER: TMZep is intended for research purposes only. It should not be used for medical or professional advice. |
What's New
TMZep (v1.0.0)
• January 6, 2019 4:23 PM
Created TMZEP.
• November 17, 2019 10:34 AM
Updated the imputation function.
• June 13, 2020 6:00 PM
Added the z-score button.
• August 30, 2021 10:37 AM
revised the predictor.
• September 7, 2022 3:42 PM
Last modified.
| © 2019-2022, Wang Genomics Laboratory@HKUST. All rights reserved. |
TMZ sensitivity prediction
Prepare the data:
Download sample data template.
The template includes several samples from TCGA. Replace them with your own data.
RNAseq expression features (EGR4_exp, PAPPA_exp, LRRC3_exp, ANXA3_exp, MGMT_exp) require Z-scores calculated by below formula.
Z-score calculation:
Z = (expression in single tumor sample) - (mean expression in all tumor samples in the cohort) / (standard deviation of expression in all tumor samples in the cohort)
If RNAseq is not available for a sample, put NA for these features.
The rest of the features takes 1 (sample has the feature), 0 (sample do not have the feature), or 0.5 (NA) for input values.
“MGMT_methylated” denotes MGMT promoter status.
“subtype_Mesenchymal”, “subtype_Classical”, “subtype_Proneural” are the three GBM subtypes.
Gene_name+(amp/del) stands for copy number amplification (amp) or deletion (del).
Features with only gene names are for somatic mutations. Mutations with moderate to high impact (e.g., missense, frameshift, nonsense mutation) should be considered.