First prize in CRM Causal Inference Challenge
An interdisciplinary team of researchers led by Joris Mooij of the Informatics Institute won the first prize in the CRM Causal Inference Challenge 2016.
The task of the challenge was to predict the effects of gene knockouts on several phenotypes for mice, without knowing which gene would be knocked out.
The team used two approaches: one based on human expert knowledge (consulting a biologist and a medical researcher specialized in inflammation associated anemia), the other using an automatic, purely data-driven, causal prediction algorithm that was specifically developed for this challenge by the team. Both approaches were awarded the first prize for their scientific novelty and because they gave the most accurate predictions.
This event took place at Montreal, Canada, 25 July - 9 August 2016. The price was part of the Workshop Statistical Causal Inference and its Applications to Genetics.
Statistical Causal Inference and its Application to Genetics
There are a wide variety of approaches which have been developed for estimation and causal interpretation when both observational and experimental data are available. Although the performance of each approach will vary from setting to setting, the organisers hoped this competition would provide a realistic setting in which to broadly compare a wide variety of methods. The participants were encouraged to use either existing or novel methods.
The interdisciplinary team
The team members are Tom Claassen, Sara Magliacane, Philip Versteeg, Stephan Bongers, Thijs van Ommen, Patrick Forre, Renée van Amerongen (Swammerdam Institute for Life Sciences) and Lucas van Eijk (Radboud University Medical Center).