Back to Search
Start Over
Predicting phenotypes from genetic, environment, management, and historical data using CNNs
- Source :
- Washburn, J D, Cimen, E, Ramstein, G, Reeves, T, O’Briant, P, McLean, G, Cooper, M, Hammer, G & Buckler, E S 2021, ' Predicting phenotypes from genetic, environment, management, and historical data using CNNs ', Theoretical and Applied Genetics, vol. 134, no. 12, pp. 3997-4011 . https://doi.org/10.1007/s00122-021-03943-7
- Publication Year :
- 2021
- Publisher :
- Springer Science and Business Media LLC, 2021.
-
Abstract
- Key Message: Convolutional Neural Networks (CNNs) can perform similarly or better than standard genomic prediction methods when sufficient genetic, environmental, and management data are provided. Abstract: Predicting phenotypes from genetic (G), environmental (E), and management (M) conditions is a long-standing challenge with implications to agriculture, medicine, and conservation. Most methods reduce the factors in a dataset (feature engineering) in a subjective and potentially oversimplified manner. Deep neural networks such as Multilayer Perceptrons (MPL) and Convolutional Neural Networks (CNN) can overcome this by allowing the data itself to determine which factors are most important. CNN models were developed for predicting agronomic yield from a combination of replicated trials and historical yield survey data. The results were more accurate than standard methods when tested on held-out G, E, and M data (r = 0.50 vs. r = 0.43), and performed slightly worse than standard methods when only G was held out (r = 0.74 vs. r = 0.80). Pre-training on historical data increased accuracy compared to trial data alone. Saliency map analysis indicated the CNN has “learned” to prioritize many factors of known agricultural importance.
- Subjects :
- Feature engineering
Gene by environment
Computer science
business.industry
Yield (finance)
General Medicine
Standard methods
Biology
Perceptron
Machine learning
computer.software_genre
Convolutional neural network
Prediction methods
Genetics
Deep neural networks
Survey data collection
Saliency map
Artificial intelligence
business
Agronomy and Crop Science
computer
Biotechnology
Slightly worse
Subjects
Details
- ISSN :
- 14322242 and 00405752
- Volume :
- 134
- Database :
- OpenAIRE
- Journal :
- Theoretical and Applied Genetics
- Accession number :
- edsair.doi.dedup.....edbb83766b2d24e54b228d903b7e2be5