Data and Knowledge Management in Translational Research: Implementation of the eTRIKS Platform for the IMI OncoTrack Consortium

This is a demo website for reviewers/readers to better test the concept/implementation reported in our manuscript “Data and Knowledge Management in Translational Research: Implementation of the eTRIKS Platform for the IMI OncoTrack Consortium”.

The example presented here is using curated (public) data and tranSMART instance resulted from the IMI eTRIKS Consortium work that will be reported in a separate paper.

Users of this demo server are encouraged to follow the use case first and then feel free to test any other hypotheses they come up with the datasets provided in the demo server.

A Demo Use Case

This use case is to guide the user to walk through the main function of the system. For more detailed user guide, please visit the documentation page from the tranSMART foundation.

A screen recording is available for more straightforward understanding of the steps. Please choose higher quality (recommend 1080p50) in your youtube settings for better visualisation.

How to browse datasets

To test the steps yourself, please visit the eTRIKS Public Server.

As show in Figure 1, the user interface is composed of three major components:

  1. The data Tree: datasets in the system are organised in a tree structure. User can click on the “+” or “-“ in front of each node of the tree to expand/collapse the branches.
  2. The Cohort Selection Boxes: user can define their own sub-cohorts here by filtering the datasets (see below).
  3. Function Tabs: after cohort selection user can perform certain actions (visualisation, analysis, export etc.) on the sub-cohorts.

Figure1: Figure1

How to setup your customised sub-cohorts

User can build customised sub-cohorts to test their hypothesis. For example, one might want to compare subjects in the TCGA Breast Invasive Carcinoma (BRCA_MERCK) study that deceased earlier (here “days to death” < 2 years or 730 days) to those that survived longer (here “days to death” >= 5 years or 1825 days). To do that, one can simply drag the “days to death” node into the “Subset 1” and “Subset 2” boxes, and in the popup windows define the criteria (see Figure 2 and 3).

For simplicity we denote the sub-cohorts as "short survivors" and "long survivors".

Figure2: Figure2

Figure3: Figure3

Testing hypotheses

To test any hypotheses (i.e. visualisation and analysis), one can click the “Summary Statistics” button in the Function Tabs (see Figure4). Here user will be able to get information on the sub-cohorts he/she has just setup.

Figure4: Figure4

As examples, we would like to know:

Hypothesis 1: If there are differences in terms of “radiation therapy” between “short survivors” and “long survivors”.

Hypothesis 2: If there are differences in “ABHD10 expression” between samples from “short survivors” and “long survivors”.

To do that, one can simply drag the “radiation therapy” node and “Agilent 244K Custom Gene Expression G4502A-07-3” node in to the “Summary Statistics” window (see Figure 5).

Figure5: Figure5

For Hypothesis 1: user will immediately get the statistics displayed as in Figure 6, in which a “non-significant” result is found.

Figure6: Figure6

For Hypothesis 2: user will have the interface to choose the gene (here ABHD10) as shown in Figure7.

Figure7: Figure7

Afterwards the results and statistics is displayed as in Figure 8, in which a “significant” result is found.

Figure8: Figure8

To test the use case, please visit the eTRIKS Public Server.