Table 2.
CAI comparison between original sequences and optimized sequences.
| DNA | bp | Original | Genewiz | ThermoFisher | BiLSTM-CRF(a) | BiLSTM-CRF(b) |
|---|---|---|---|---|---|---|
| HPDF | 615 | 0.70 | 0.85 | 0.92 | 0.96 | 0.98 |
| PAE | 1839 | 0.76 | 0.81 | 0.92 | 0.96 | 0.98 |
| MMPL3 | 2835 | 0.67 | 0.79 | 0.93 | 0.96 | 0.98 |
| FALVAC-1 | 972 | 0.67 | 0.85 | 0.93 | 0.95 | 0.96 |
| PA | 561 | 0.60 | 0.83 | 0.93 | 0.97 | 0.98 |
| PTP4A3 | 564 | 0.70 | 0.83 | 0.93 | 0.96 | 0.98 |
| Average | 1231 | 0.68 | 0.83 | 0.93 | 0.96 | 0.98 |
This table shows the CAIs of the sequences optimized by different optimization tools, among which the values for Genewiz and ThermoFisher are provided on their official websites (ThermoFisher: www.thermofisher.com, Genewiz: www.genewiz.com). BiLSTM-CRF(b) has the highest average CAI, showing that it has great potential to enhance protein expression.