Enhancing protein expression in humans through codon optimization with transformer and contrastive learning
Juseong Kim, Jeongmu Kim, Jae-Wook Lee, Qian Qi, Caroline Danehy, Ho Young Kang, Yong Cheng, Giltae Song · 2026
Read Enhancing protein expression in humans through codon optimization with transformer and contrastive learning by Juseong Kim, Jeongmu Kim, Jae-Wook Lee, Qian Qi, Caroline Danehy, Ho Young Kang, Yong Cheng, Giltae Song (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13330525.txt.
Description
Juseong Kim and colleagues present a deep learning tool for codon choice in human gene designs in a 2026 study. Messenger RNA drugs need codon patterns that boost protein output and keep the RNA stable. Older methods lean on frequency tables and miss wider sequence context. The new tool, COformer, blends convolutional and transformer networks.
Contrastive learning helps the model tie codon patterns to matching amino acids while reading both local and global context. Training used sequences tuned by commercial tools. In lab tests, designs from COformer raised protein output through better codon use and stronger RNA stability, the authors report. The approach may aid vaccine and drug design where output levels matter.
Catalog record
| CREATOR | Juseong Kim, Jeongmu Kim, Jae-Wook Lee, Qian Qi, Caroline Danehy, Ho Young Kang, Yong Cheng, Giltae Song |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51626 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |