17-19 October the Netherlands Cancer Institute hosted the 2015 tranSMART Foundation Annual Meeting. With three days of talks, trainings, discussions, poster sessions and over 160 attendees this year's meeting has been the largest meeting up until now.
Together with CTMM TraIT and IMI TRANSLOCATION we gave 4 talks in the science and technology tracks, displayed 3 posters, organized 2 very well attended trainings and organized the 1 and only hackathon.
Amongst the +25 talks where many contributions from The Hyve. As leader of the architecture working group our CEO Kees van Bochove gave multiple presentations, introducing the hackathon introduction (below) and on the architecture of tranSMART:
Our front-end developer Riza Nugraha gave the following talk on the new TranSMART UI concept we have developed in the IMI TRANSLOCATION project:
More information can be found here.
Besides these two recorded sessions Jan Hudecek (NKI, CTMM TraIT) and Peter Kok (The Hyve, CTMM TraIT) gave their talk "Improved visualisation of molecular data from tranSMART: the cBioPortal connection" on the developments we made to integrate TranSMART with the recently open sourced cBioPortal tool.
Also our big-data architect Sjoerd van Hagen (The Hyve, TRANSLOCATION) presented "Data Integration: Combining compound-centric data with subject-centric data in R by linking data in LSP and TranSMART" on how we combined the R libraries from tranSMART and the the LSP tool (by gritsystems) to allow the user to perform compound centric queries on tranSMART data.
During this meeting we displayed three posters, of which we already told you about two of them:
- Sjoerd van Hagen presented the "Evaluation of Spark and ADAM for large scale genomics data" poster, in which we demonstrate our experiences with Spark and ADAM for storing and doing calculations on big genomics data.
- Ward Weistra presented the "Multi-omics data analysis in tranSMART using the Cell Line Use Case dataset" poster we created with CTMM TraIT, which recently won the Bio-IT World 2015 Best Poster award.
New for this conference was the poster on the "Data Download Portal for The Maastricht Study Project" presented by our Piotr Zakrzewski on the data download portal extension we build on top of the TranSMART REST API, to allow clients to use tranSMART a controlled data distribution portal.
With CTMM TraIT we also organised two advanced trainings on data loading, TranSMART ETL with transmart-data (and transmart-batch) using the TraIT Cell Line Use Case, and on data science with tranSMART REST API in Python and R. Both were very well attended and well received. The materials for these trainings will be available online later.
Both trainings will also be included in the 2016 training schedule of the TranSMART Foundation, which will be published soon!
Finally we also organized the hackathon, the returning element for developers on the TranSMART Foundation Annual Meeting schedule. This year the team of both onside and offside developers focused on two topics:
- Extending the SmartR visualisation extension with more interactive analytics workflows.
- Building a proof of concept around using Apache Spark as the analytical backend for tranSMART
Here is the recording where Kees introduces the hackathon goals:
This Lanyrd listing gives a more detailed summary of the goals for the hackathon.
On the last day our developer Gijs Kant and Sascha Herzinger (Luxembourg Centre for Systems Biomedicine) gave the following summary the achievements of this successful hackathon:
The 2015 annual meeting gave a great overview of the growth of the community and the ongoing developments on the data integration platform. With the constructive discussions on the future of the platform I am sure 2016 will look even brighter.