Table 1 MCC of the four deep learning models on the selected temperatures. The results suggest that the models perform reasonably well even on temperatures they were not trained on.
From: Scaling up DNA digital data storage by efficiently predicting DNA hybridisation using deep learning
Model | 37.0 °C | 42.0 °C | 47.0 °C | 52.0 °C | 62.0 °C |
|---|---|---|---|---|---|
RNN | 0.862 | 0.872 | 0.888 | 0.916 | 0.833 |
CNN | 0.884 | 0.893 | 0.908 | 0.929 | 0.811 |
\(\text {CNN}_{\text {Lite}}\) | 0.893 | 0.901 | 0.914 | 0.929 | 0.801 |
RoBERTa | 0.925 | 0.930 | 0.930 | 0.914 | 0.762 |
