Building and Exploring Knowledge Network in Genestack
Building and Exploring Knowledge Network in Genestack
Date: Tuesday, January 17, 2017
Time: 2:30 PM
Time: 2:30 PM
Room: Sunset - Meeting House
Outside Genestack, building and exploring knowledge networks is a slow and challenging process. First, users have to identify and collect many large heterogeneous data, from various sources, in the right format to be integrated by Ondex. Then, users will face a steep learning curve of using Ondex to build networks: dealing with the technicalities of parsers, mappers, collapsers, etc. To explore a network, users would then need to manually configure, build, and set up a KnetMiner server. Furthermore, both the building and exploration stages demand high memory, which presents barriers for many users.
Genestack aims to help users with all these issues. We provide a flexible data management environment to simplify the data acquisition process: public datasets are immediately accessible on our platform, and custom datasets can be imported into the right data types and immediately shared and processed by Genestack’s Knowledge Network Builder application. The application enables users to run Ondex without dealing with low-level technical details: based only on the supplied datasets and minimal user specifications, the application automatically constructs an appropriate network that ties these datasets together. The created knowledge network can then be immediately explored using Genesack’s KnetMiner application, which also allows results to be saved and shared. Furthermore, users benefit from Genestack analytical pipelines, such as differential expression analysis, to immediate process raw data into interesting genes and be used as input in the network mining stage. Lastly, we’ll present ideas of integration with NDEx (Network Data Exchange), which includes publication of networks.