Fig. 8: Predictive performance of crisprHAL. | Nature Communications

Fig. 8: Predictive performance of crisprHAL.

From: A generalizable Cas9/sgRNA prediction model using machine transfer learning with small high-quality datasets

Fig. 8: Predictive performance of crisprHAL.The alternative text for this image may have been generated using AI.

A Correlation between the TevSpCas9 dataset (n = 279) and the TevSpCas9 model 5-fold cross validation predictions (mean rank correlation of 0.630 across 5-folds). B Correlation between the SpCas9 dataset (n = 302) and the SpCas9 model 5-fold cross validation predictions (mean rank correlation of 0.627 across 5-folds). For (A) and (B) the line of best fit is indicated and the 95% confidence interval is represented by a gray shaded area. C TevSpCas9 model and (D) SpCas9 model prediction correlations with the original Z-scores from the unique Guo SpCas9 dataset (n = 7821). E TevSpCas9 model and (F) SpCas9 model prediction correlations with the Citrobacter rodentium TevSpCas9 dataset (n = 30138). G TevSpCas9 model and (H) SpCas9 model prediction correlations with the S. entericakatG fragment TevSpCas9 dataset (n = 228). Source data are provided as a Source Data file.

Back to article page