Causal Inference Enrichment (CIE)

CIE is a platform for identification of active transcriptional regulators from differential gene expression data. CIE provides directional (Ternary and Quaternary scoring statistic) and un-directional (Fisher's Exact) enrichment analysis on regulatory networks. Networks integrated within the platform include TRED, TRRUST, STRINGdb, ChIP-Atlas, and tissue corrected ChIP-Atlas. TRED and ChIP-Atlas are not annotated with information on the mode of regulation (activation vs. repression) and are suitable for use with Fisher's exact test only. Tissue corrected ChIP-Atlas is a network constructed by combining ChIP-Atlas with tissue-specific gene expression data from GTEx using a regularized Gaussian Graphical Model. This network is annotated with information on the mode of regulation and can be used with Fisher or Ternary enrichment methods. STRINGdb has mixed directionality and can be used with the Quaternary method.


For offline, local use, or to use with custom networks, please download the R-package from our GitHub page: GitHub page: umbibio/CIE-R-Package


All required network data, as well as test differential gene expression data, are available in the Downloads tab.


See the Help tab for a typical CIE analysis



Citations

1) Farahman, S., O'Connor, C., Macoska, J., Zarringhalam, K. 'Causal Inference Engine: A platform for directional gene set enrichment analysis and inference of active transcriptional regulators', NAR doi: https://doi.org/10.1093/nar/gkz1046


2) Fakhry CT, Choudhary P, Gutteridge A, Sidders B, Chen P, Ziemek D, Zarringhalam, K. 'Interpreting transcriptional changes using causal graphs: new methods and their practical utility on public networks'. BMC Bioinformatics. 2016;17(1):318


Questions and Comments

Please contact Corey O'Connor (corey.oconnor001@umb.edu)

Causal Inference and Directional Enrichment Methods on Biological Networks

File download

Download

Table of Contents

Getting Started

Interacting with Output

Getting Started

  1. Select a regulatory network from the drop-down menu. The default is tissue corrected ChIP Atlas
  2. Upload a differential gene expression profile. This is a tab-delimited file of deferentially expressed genes with three mandatory columns: entrez id, p-value, fold change. The columns must be named (entrez, p-value, and FC). You can download sample input from the Download tab.
  3. Select an enrichment method. By default, CIE selects Fisher as this is the quickest analysis. You may also select a more informed method of enrichment suitable for the network.

Interacting With Output

  1. To display the result graphically, click the ‘Display Graph’. You should click the button again after you have altered the parameters (rows selected or number of targets)
  2. To run enrichment on Reactome, select regulators and click ‘Run pathway enrichment'. A new table will be generated with enrichment results. You will be automatically switched to it as it is generated. You should click the button again if you have altered the parameters (rows selected or paths to display). Paths to display controls how many pathways are returned from Reactome. This defaults to 10, the number of rows in one page of the displayed table.