Prediction of evolutionary constraint by genomic annotations improves functional prioritization of genomic variants in maize
Problem
Identifying which maize genetic variants are functionally important, at high resolution, to support crop improvement via genomic prediction and genome editing.
Why it matters
Finding causal variants precisely is a bottleneck for improving crops and food production.
What the AI did
Deep learning combined with bioinformatics annotations was used to predict nucleotide conservation across angiosperms as a proxy for mutation fitness effects.
Results
Annotated nonsynonymous mutations in 25,824 maize gene models; predictions validated by within-species conservation, chromatin accessibility, and gene expression.
Impact
The evidence supports improved functional prioritization of maize variants, validated by experimental data.
Record
| Venue | Genome biology |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Guillaume P. Ramstein, Edward S. Buckler |
| Citations | 34 (OpenAlex, at import time) |
| References | 92 |
| Model or method | deep learning (specific architecture not stated in source) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1186/s13059-022-02747-2 |
| Open access | gold |
| Keywords | maize, genomic annotations, deep learning, variant prioritization, nucleotide conservation |
Importance rating
significant — thousands to millions affected
Abstract states causal variant identification is required for crop improvement via genomic prediction and genome editing, and the method improves prediction of fitness-related traits like grain yield.