6533b7cffe1ef96bd1258ff9

RESEARCH PRODUCT

MOESM5 of 7C: Computational Chromosome Conformation Capture by Correlation of ChIP-seq at CTCF motifs

subject

description

Additional file 5: Figure S2. 7C model parameters and optimal cut-offs for binary prediction. (A) Parameter values of the logistic regression model in 7C for different features (columns), separated for different models (rows). Average of model parameters of model training in 10-fold cross-validation is shown with error bars indicating the standard deviations. While the first six rows represent the models with the indicated TF ChIP-seq data and the genomic features, “Avg. all TF” is the average across all 124 TFs analyzed and “Avg. best 10 TF” is the average across the best ten performing TF models. (B) Prediction performance as f1 score (y-axis) for different cutoffs on the prediction probability p for the six selected models.