Fig. 6. Comparison of disease candidate genes prediction results for different number of discreate points (NC) in gene expression quantization. 4.3.3. Learning and predicting results per disease We employ 10-fold cross validation technique to evaluate the performance of the proposed method. Using th...
This study presents a feature selection-based method for drug response prediction, named Auto-HMM-LMF, to efficiently predict cell line-drug associations. Gene expression profile, copy number alteration, single-nucleotide mutation, tissue type information of the cell line, and drugs’ chemical structu...
(a) The network structure of VGAE. For the input graph, the adjacency matrix and the gene expression matrix are aggregated through the single-layer GCN, and then the parameters of the distribution satisfied by the spots are learned through two GCN network modules. The hidden representation is ...