Personalized Prediction of Drug Response

We developed a novel data mining approach for predicting drug response using genetic and clinical information. This algorithm ranks top in the 2014 DREAM (Dialogue for Reverse Engineering Assessments and Methods) rheumatoid arthritis drug response challenge, for both genetics-only model and combined model, for both sub-challenges measuring different clinical outcomes, and for both the leaderboard and the final previously unseen test set.

Congratulations to all the other teams for their excellent performance and thank IBM, NIH Roadmap Initiative, and all the for-profit and non-profit sponsors of this and past challenges for giving us the opportunities to benchmark the accuracy of the GuanLab algorithms!





Read our story below...

On April 7th, our team was selected as the first interim winner for the top accuracy in both sub-challenges and for both the genetics-only and genetics+clinical models.



We managed to defend our position and selected as the second interim winner on May 7th.



To be the final winner is more challenging: an one-shot prediction is allowed for a previously unseen test population. YET WE WIN! (June 4 for final submission)