Skip to main content
Fig. 5 | Journal of Cheminformatics

Fig. 5

From: SESNet: sequence-structure feature-integrated deep learning method for data-efficient protein engineering

Fig. 5

Results of models trained on different datasets. A–B: The spearman correlation of fitness prediction on high-order mutants by finetuning on 37 experimental single-site mutation results from datasets of F7YBW8 and on 20 experimental single-site mutation results of CAPSD_AAV2S, respectively. Where the red and blue bars represent the results of the pre-trained model and the original model without pretraining. And the green bars correspond to the results of the unsupervised model, which is ProGen2 for F7YBW8 and ESM-IF1 for CAPSD_AAV2S, respectively

Back to article page