Virginia Tech team wins 4th place in DREAM 9.0
Our team, from Prof Heath lab, won the fourth place in BROAD-DREAM challenge 9.0. The goal of this challenge is to develop predictive models that can infer gene dependencies in cancer cells using features of the cell lines.
Participants are provided with gene essentiality scores from a set of cancer cell lines. The challenge is to build predictive models that can infer these gene dependencies from the molecular characterization of these cell lines. Gene expression data, copy number data, and mutation data obtained from hybrid sequencing for the same set of cell lines are provided to the participants.
Given the gene expression and copy number measured over 149 cell lines (105 for training and 44 for testing), we developed two systems to predict the gene essentiality score for 14,780 genes. Our best model was ranked fourth in the competition.
check the final scores here,
https://www.synapse.org/#!Synapse:syn2384331/wiki/64760
Participants are provided with gene essentiality scores from a set of cancer cell lines. The challenge is to build predictive models that can infer these gene dependencies from the molecular characterization of these cell lines. Gene expression data, copy number data, and mutation data obtained from hybrid sequencing for the same set of cell lines are provided to the participants.
Given the gene expression and copy number measured over 149 cell lines (105 for training and 44 for testing), we developed two systems to predict the gene essentiality score for 14,780 genes. Our best model was ranked fourth in the competition.
check the final scores here,
https://www.synapse.org/#!Synapse:syn2384331/wiki/64760