Predicting phenotypes from genetic, environment, management, and historical data using CNNs
Open accessPublished version of a preprint
The DOI matches its Crossref record; 79 of the 79 references it cites were found in a registry or an index.
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. 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.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“The methodology is also flexible with respect to data type, allowing the combination of variables that are static over a growing season (e.g.genotype) and those that are dynamic (e.g.temperature) in a single model (Washburn et al. 2021).”
From Yield Prediction Through Integration of Genetic, Environment, and Management Data Through Deep Learning (2023) · cites this paper as Washburn et al. 2021
In the citing paper
“Several recent studies have applied these methods with a relatively large dataset either with (Washburn et al. 2021) or without (Khaki et al. 2020) a genetic component into the model, with little feature engineering performed.”
From Yield Prediction Through Integration of Genetic, Environment, and Management Data Through Deep Learning (2023) · cites this paper as Washburn et al. 2021
In the citing paper
“Although we aimed to broaden the range of possible architectures relative to previous modeling on G2F data (Washburn et al. 2021), we constrained the overall structure to process each tensor individually then allowing for interactions between the final layer of each module.”
From Yield Prediction Through Integration of Genetic, Environment, and Management Data Through Deep Learning (2023) · cites this paper as Washburn et al. 2021
In the citing paper
“A strong peak in salience shortly after planting is seen in both saliency maps which agrees with previously reported results (Washburn et al. 2021).”
From Yield Prediction Through Integration of Genetic, Environment, and Management Data Through Deep Learning (2023) · cites this paper as Washburn et al. 2021
In the citing paper
“Similar to the results of previous modeling (Washburn et al. 2021), we find that no single data grouping provides sufficient information to disregard all others.”
From Yield Prediction Through Integration of Genetic, Environment, and Management Data Through Deep Learning (2023) · cites this paper as Washburn et al. 2021
1 more claim about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Book Review: Pattern Recognition and Machine Learning
Nasrabadi, N. M. (2007) · Journal of Electronic Imaging · doi:10.1117/1.2819119 · Cited by 22,066 · cited 3×
“The choice of ReLU activation functions as opposed to sigmoid or hyperbolic tangent (tanh) was based on its popularity in the field and evidence of its ability to converge more quickly than tanh activation functions (Bishop 2006; Xavier Glorot et al. 2011; LeCun et al. 2015).”
BackgroundMethodPattern Recognition and Machine Learning
(2006) · doi:10.1007/978-0-387-45528-0 · Cited by 9,824
TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G. S., Davis, A., Dean, J., Devin, M., et al. (2016) · arXiv · doi:10.48550/arxiv.1603.04467 · Cited by 9,408 · cited 1×
“The CNN models were built and run using the tensorflow.keras module version 2.1.0 in python 3.6.9 (Chollet 2015; Abadi et al. 2016a, b).”
MethodInfluentialPrediction of Total Genetic Value Using Genome-Wide Dense Marker Maps
Meuwissen, T. H. E., Hayes, B. J., & Goddard, M. E. (2001) · Genetics · doi:10.1093/genetics/157.4.1819 · Cited by 7,917
TASSEL: software for association mapping of complex traits in diverse samples
Bradbury, P. J., Zhang, Z., Kroon, D. E., Casstevens, T. M., Ramdoss, Y., & Buckler, E. S. (2007) · Bioinformatics · doi:10.1093/bioinformatics/btm308 · Cited by 7,759
Self-citation(2011) · Cited by 6,993
The Pfam protein families database: towards a more sustainable future
Finn, R. D., Coggill, P., Eberhardt, R. Y., Eddy, S. R., Mistry, J., Mitchell, A. L., Potter, S. C., Punta, M., Qureshi, M., Sangrador-Vegas, A., et al. (2016) · Nucleic Acids Research · doi:10.1093/nar/gkv1344 · Cited by 5,975
Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
Simonyan, K., Vedaldi, A., & Zisserman, A. (2013) · arXiv · doi:10.48550/arxiv.1312.6034 · Cited by 5,370 · cited 2×
“Additionally, while CNNs are often referred to as a “black box,” there are now welldemonstrated methods for dissecting and interpreting the processes within a given CNN model, and the features the model determines to be most important to prediction (Simonyan et al. 2013; Samek et al. 2019).”
MethodThe DSSAT cropping system model
Jones, J. W., Hoogenboom, G., Porter, C. H., Boote, K. J., Batchelor, W. D., Hunt, L. A., Wilkens, P. W., Singh, U., Gijsman, A. J., & Ritchie, J. T. (2003) · European Journal of Agronomy · doi:10.1016/s1161-0301(02)00107-7 · Cited by 4,465
Research works printed but not linked: 1
- Abadi M, Barham P, Chen J et al (2016b) Tensorflow: a system for large-scale machine learning. In: 12th symposium on operating systems design and implementation, pp 265–283
Cited by
Breeding crops for drought-affected environments and improved climate resilience
Cooper, M., & Messina, C. D. (2023) · The Plant Cell · doi:10.1093/plcell/koac321 · Cited by 195
“…the crop lifecycle (Figure 4; Cooper et al., 2014a, 2014b, 2020; Costa-Neto et al., 2021; Gage et al., 2021; Li et al., 2021; Smith et al., 2021a; Washburn et al., 2021; Diepenbrock et al., 2022; Messina Breeding for drought resistance THE PLANT CELL 2023: 35; 162–186 | 171 et al., 2022a, 2022c;…”
BackgroundInfluentialSelf-citationMachine learning for predicting phenotype from genotype and environment
Guo, T., & Li, X. (2023) · Current Opinion in Biotechnology · doi:10.1016/j.copbio.2022.102853 · Cited by 61
Kick, D. R., Wallace, J. G., Schnable, J. C., Kolkman, J. M., Alaca, B., Beissinger, T. M., Edwards, J., Ertl, D., Flint-Garcia, S., Gage, J. L., et al. (2023) · G3: Genes|Genomes|Genetics · doi:10.1093/g3journal/jkad006 · Cited by 53
Self-citationCosta‐Neto, G., Crespo‐Herrera, L., Fradgley, N. S., Gardner, K. A., Bentley, A. R., Dreisigacker, S., Fritsche‐Neto, R., Montesinos-López, O. A., & Crossa, J. L. (2022) · G3 Genes Genomes Genetics · doi:10.1093/g3journal/jkac313 · Cited by 67
Genomic selection: Essence, applications, and prospects
Escamilla, D. M., Li, D., Negus, K. L., Kappelmann, K. L., Kusmec, A., Vanous, A. E., Schnable, P. S., Li, X., & Yu, J. (2025) · The Plant Genome · doi:10.1002/tpg2.70053 · Cited by 76
“However, non-parametric models may perform better in some scenarios, such as categorical phenotypes, different data types, complex interactions, missing data, outliers, and correlated variables (Negus et al., 2024; Ogutu et al., 2011; Washburn et al., 2020, 2021).”
Metadata
- Authors
- Washburn, J. D., Cimen, E., Ramstein, G., Reeves, T., O’Briant, P., McLean, G., Cooper, M., Hammer, G., & Buckler, E. S.
- Affiliations
- Agricultural Research Service (US)
- Aarhus University (DK)
- Cornell University (US)
- The University of Queensland (AU)
- Corresponding author
- Washburn, Jacob D.
- Venue
- Theoretical and Applied Genetics
- Published
- 2021-08-27
- Volume
- 134
- Issue
- 12
- Pages
- 3997-4011
- ISSN
- 0040-5752 (ISSN-L)
- 1432-2242
- OpenAlex
- W3193827159
- Cited by
- 57
- References
- 79
- Citations by year
- 2026: 15 · 2025: 16 · 2024: 10 · 2023: 11 · 2022: 5
- FWCI
- 3.90
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Licence
- www.springer.com/tdm (tdm)
- www.springer.com/tdm (vor)
- Locations
- Theoretical and Applied Genetics (publishedVersion)
- bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion)
- PubMed (publishedVersion)
- Scopus (Elsevier) (submittedVersion)
- Funders
- Agricultural Research Service
- Directorate for Biological Sciences (1710618)
- Topics
- Life Sciences › Agricultural and Biological Sciences › Plant Science › Smart Agriculture and AI
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic and phenotypic traits in livestock
- Life Sciences › Agricultural and Biological Sciences › Plant Science › Genetics and Plant Breeding
- Keywords
- phenotype prediction, convolutional neural networks, genomic prediction, environmental effects, management practices, pre-training, saliency maps, feature importance, deep learning
- MeSH terms
- Machine Learning, Zea mays, Phenotype, Neural Networks, Computer, Crops, Agricultural, Genomics, Data Mining
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Predicting phenotypes from genetic, environment, management, and historical data using CNNs. https://app.cochat.ai/papers/10.1007/s00122-021-03943-7
- The paper
- 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. https://doi.org/10.1007/s00122-021-03943-7
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.