Frequently asked questions
- How is ONCOchannelome structured ?
ONCOchannelome is structured into four modules: Browse by tumor type, ion channel, ion channel EMT correlation and ion channel co-expression network. The browse by tumor type module consists of sample details, overall expression distribution across ion channels in the specific tumor type, number of differentially expressed (DE) ion channels, overlaps in the ion channels DE across tumor and metastatic states and list of biological processes, molecular functions and cellular compartments enriched by the DE ion channels in the tumor type across states. The ion channel retrieval module consists of details on the expression distribution of the specific ion channel across tumor types in TCGA, MET500 and GTEx datasets. The details on the fold change values of the specific ion channel in terms of differential expression can also be obtained through this module. The ion channel EMT module includes information on the correlation of the specific ion channel with epithelial and mesenchymal gene signatures in normal, tumor and metastatic samples. The co-expression of the ion channels identified to be DE with the other DE ion channels in a specific tumor type and state can be procured along with their potential transcription factors from the ion channel co-expression network module.
- How were the samples from TCGA, MET500 and GTEx merged for combined analysis ?
Based on the number of tumor samples available through TCGA, GTEx samples were filtered in terms of age, gender and tumor sample ratio. The MET500 samples were filtered to include only the samples processed using the polyadenylation method of sample preparation in order to match the methodology usied in TCGA. Batch effects across TCGA, MET500 and GTEx samples were visualized by performing principal component analysis and thereafter batch correction in terms of both sample group and sample type was performed using removeBatchEffect function from R Bioconductor limma package.
- What does NT, TM and NM stand for ?
It stands for the comparisons performed during the identification of DE ion channels.
NT – Normal and Tumor samples
TM – Tumor and Metastatic samples
NM – Normal and Metastatic samples
- How were the molecular functions, biological processes and cellular compartments for the altered ion channels identified?
Gene set enrichment analysis (GSEA) was performed using R Bioconductor clusterProfiler library with a minimum gene size set to 5, maximum gene size set to 300 and p-value cut-off set to 0.05.
- How is the present analysis different from the work presented in projects provided through external links tab ?
ONCOchannelome is a comprehensive computational framework focussed on identification of alterations in ion channels in cancer. Unlike other published resources it not only allows users to identify the differentially expressed ion channels but also:
Provides state specific alterations
Provides potential molecular functions and biological processes that could be activated or suppressed
Calculates correlation of altered ion channels with epithelial and mesenchymal gene signatures
Generates co-expression networks of differentially expressed ion channels and
Provides details on the transcription factors that could potentially regulate the ion channel co-expression network
- How was EMT correlation calculated ?
Epithelial and mesenchymal gene signatures were procured from published literature. Single sample gene set enrichment analysis was performed on the transcriptome profiles across tumors with availability of metastatic samples. Thereafter, gene set enrichment quantified and Spearman correlation was calculated.
- How was co-expression network generated ?
Weighted gene co-expression network analysis was performed to generate the co-expression networks.
- What do the colours of the nodes in the Co-expression network indicate ?
Cyan represents the ion channel to be downregulated and yellow represents the ion channel to be upregulated.
- What is the co-expression network summary table ?
The co-expression network summary table are the values generated in the intermediate steps of weighted gene co-expression network analysis.
- How were the transcription factors identified ?
Transcription factors were identified using Enrichr.
- How was the correlation of the transcription factors with the network calculated ?
The module eigen genes obtained through intermediate steps of WGCNA and the expression values of the identified transcription factors were used to determine Pearson correlation with t-distribution based p-value.
- What are the sources of the identified transcription factors ?
The following are the sources of the identified transcription factors
ChEA - ChIP enrichment analysis
PPIs - Enrichr for protein-protein interactions
ENCODE - Encyclopedia of DNA elements transcription factors ChIP-seq 2015
TRANSFAC - TRANScription FACtor database
JASPAR - Position weight matrix (PWMs)
TRRUST - Transcriptional regulatory relationships unraveled by sentence-based text mining