Organising useR!2017 was a challenge but a very rewarding experience. With about 1200 attendees of over 55 nationalities exploring an interesting program, we believe it is appropriate to call it a success – something the aftermovie only seems to confirm.

Behind the Scenes

To give you a glimpse behind the scenes of the conference organization, Maxim Nazarov held a lightning talk on ‘redmineR and the story of automating useR!2017 abstract review process’

You can find the R package on the Open Analytics Github and slides are available here.

Laure Cougnaud presented during the useR! Newbies session in a talk called ‘Making the most of useR!’ and assisted the newbies throughout the conference as a conference buddy. She also served as the chair of the Bioinformatics I session.

In spite of recent appearances Open Analytics does more than organizing useR! Conferences and, as a platinum sponsor, Tobias Verbeke had the opportunity to present Open Analytics in a sponsorship talk.

Open Analytics offers its services in four different service lines:

• statistical consulting,
• scientific programming,
• application development & integration and
• data analysis hardware & hosting.

The talks our consultants contributed can be nicely laid out along these service lines.

Statistical Consulting

On the methodological side (statistical consulting) Kathy Mutambanengwe held a talk on ‘A restricted composite likelihood approach to modelling Gaussian geostatistical data’.

Adriaan Blommaert and Nicolas Sauwen co-authored the lightning talk by Tor Maes on Multivariate statistics for PAT data analysis: short overview of existing R packages and methods. Finally, the poster session presented work by Machteld Varewyck on Analyzing Digital PCR Data in R and Rytis Bagdziunas presented a poster on BIGL: assessing and visualizing drug synergy.

Scientific Programming

In the scientific programming area, Nicolas Sauwen held a talk on the ‘Differentiation of brain tumor tissue using hierarchical non-negative matrix factorization’

The application to differentation of brain tumor issue is an interesting case, but the hNMF method is currently the fastest NMF implementation and can be put to use in many other contexts of unsupervised learning. If interested, the discussed hNMF package can be found on CRAN.

Kirsten Van Hoorde held a lightning talk on the ‘R (‘template’) package to automate analysis workflow and reporting’,

co-authored by Laure Cougnaud. For an example package demonstrating the approach, please see the Open Analytics Github – slides can be found here.

Application Development and Integration

Regarding application development and integration, Marvin Steijaert and Griet Laenen held a long talk on ‘Biosignature-Based Drug Design: from high dimensional data to business impact’ demonstrating how machine learning is put to use to design drugs (slides here).

Data Analysis Hardware and Hosting

Regarding hosting of data analysis applications, Tobias Verbeke held a well attended talk on ShinyProxy a fully open source product that allows to run Shiny apps at scale in an enterprise context.

All information can be found on the ShinyProxy website and sources are on Github.

We hope you enjoyed the conference as much as we did. Let us know if you have any questions or comments on these talks or on Open Analytics and its services offer.

On popular demand: here you can find the source code of the Poissontris game – that other Shiny app 😉