Effective gene expression prediction from sequence by integrating long-range interactions.
Avsec Ž, Agarwal V, Visentin D, Ledsam JR, Grabska-Barwinska A, Taylor KR, Assael Y, Jumper J, Kohli P, Kelley DR. · 2021
Read Effective gene expression prediction from sequence by integrating long-range interactions. by Avsec Ž, Agarwal V, Visentin D, Ledsam JR, Grabska-Barwinska A, Taylor KR, Assael Y, Jumper J, Kohli P, Kelley DR. (2021) 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/europmc-pmc8490152.txt.
Description
Avsec and co-authors build Enformer, a deep learning model that predicts gene expression from DNA sequence. How noncoding DNA sets gene activity in each cell type is a big open problem. Human genetics work depends on better answers.
Enformer reads long-range contacts up to 100 kilobases away in the genome. That reach sharply raised prediction accuracy over past models. Variant effect calls improved for both natural variants and lab-made changes. The model also learned enhancer-promoter contacts straight from sequence. It matched methods that need direct lab data as input. The authors expect gains for disease mapping and for reading regulatory evolution.
Catalog record
| CREATOR | Avsec Ž, Agarwal V, Visentin D, Ledsam JR, Grabska-Barwinska A, Taylor KR, Assael Y, Jumper J, Kohli P, Kelley DR. |
|---|---|
| YEAR | 2021 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |