A deep learning model captures position-specific effects of plant regulatory sequences and suggests genes under complex regulation
Kevin C Rockenbach, Silvia F Zanini, Alison C Tidy, Richard J Morris, Rachel Wells, Agnieszka A Golicz · 2026
Read A deep learning model captures position-specific effects of plant regulatory sequences and suggests genes under complex regulation by Kevin C Rockenbach, Silvia F Zanini, Alison C Tidy, Richard J Morris, Rachel Wells, Agnieszka A Golicz (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/PMC13436691.txt.
Description
Rockenbach and colleagues trained a deep learning model in 2026 to predict plant gene activity straight from DNA sequence, then probed what the model had learned. Trained on the start and end control regions of genes in oilseed rape and its relative thale cress, the network forecast activity levels for each gene. The aim was to read gene control rules out of the model rather than assume them in advance.
Predictions matched measured activity across plant varieties. Without ever being shown binding data, the model had picked up the preferred positions of some transcription factor families, the proteins that dock on control regions. It judged a pattern's effect by where the pattern sat, not just its presence. Genes whose activity the model missed most tended to receive extra input from distant or chemical-mark control layers. The authors say such models can guide crop gene engineering.
Catalog record
| CREATOR | Kevin C Rockenbach, Silvia F Zanini, Alison C Tidy, Richard J Morris, Rachel Wells, Agnieszka A Golicz |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 106 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |