Towards AI-designed genomes using a variational autoencoder
Natasha K. Dudek, Doina Precup · 2024
Read Towards AI-designed genomes using a variational autoencoder by Natasha K. Dudek, Doina Precup (2024) 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/pmc-11631412.txt.
Description
Natasha Dudek and Doina Precup train a smart model to learn the gene makeup of bacterial genomes. Genomes are vast webs of genes that must work as one. They code each genome as a long list of gene on or off marks and mask most marks at train time.
The model learns to rebuild the full list from the masked one, which forces it to learn gene links. One made up genome showed near whole, well linked life paths that fit one habitat. Scores on held out genomes were strong. The work hints AI may one day draft working genomes.
Catalog record
| CREATOR | Natasha K. Dudek, Doina Precup |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |