OmicsDI

Omics Discovery Interface

Overview

Link to the presentation by Yasset Perez-Riverol.

Building a system to discover related data sets, e.g. finding most similar data sets. This does however, open the door to privacy issues, e.g. to link longitudinal data from a single individual (see also the privacy in proteomics topic).

The system should be scaled up to multiple databases.

In order to link data sets based on their quality, we'll first need to be able to query the data at a very detailed level, and this level of granularity is currently not exposed on the web. Genomics is already doing this.

In the future, privacy/ethics constraints may also require that some of the data cannot travel from the host institute, which leads to more distributed systems.

The basic function of discovering whether an interesting data set exists, is actually already very useful.