Skip to main content
. 2020 Jan 8;10:1303. doi: 10.3389/fgene.2019.01303

Table 1.

Performance comparison between CNN-SVR and CNN models for gRNA activity prediction on four cell-line datasets under 10-time 10-fold cross-validation.

Model CNN-SVR CNN CNN-SVR CNN
Spearman AUROC
HCT116 0.719 ± 0.008 0.661 ± 0.030 0.933 ± 0.001 0.932 ± 0.001
HEK293T 0.807 ± 0.016 0.725 ± 0.029 0.983 ± 0.002 0.972 ± 0.001
HELA 0.699 ± 0.006 0.702 ± 0.007 0.933 ± 0.001 0.916 ± 0.001
HL60 0.589 ± 0.006 0.576 ± 0.040 0.934 ± 0.003 0.914 ± 0.003

Performance is shown as mean ± standard deviation. This representation also applies to Table 2. The best performance across different folds cross-validation method is highlighted in bold for clarification. These highlights also apply to Tables 2 to 4 and Tables S3 to S5.