Machine learning models outperform deep learning models, provide interpretation and facilitate feature selection for soybean trait prediction

Gill, Mitchell; Anderson, Robyn; Hu, Haifei; Bennamoun, Mohammed; Petereit, Jakob; Valliyodan, Babu; Nguyen, Henry T.; Batley, Jacqueline; Bayer, Philipp E.; Edwards, David

doi:10.1186/s12870-022-03559-z

Cited by 27 publications

(14 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There currently appears to be few advantages to using MLGP over LMMGP, with further improvements needed to deliver stepchange improvements in crop genetic gain. Other studies have reported similar findings where relative performance of MLGP and LMMGP has been (disappointingly) similar, highly susceptible to dataset structure and experimental design, and generally non-transferrable to new datasets (Aono et al, 2022;Danilevicz et al, 2022;Gill et al, 2022;Jubair and Domaratzki, 2023). This initial lack of progress may not yet rule out the capabilities of MLGP with further development.…”

Section: Discussionmentioning

confidence: 88%

The pursuit of genetic gain in agricultural crops through the application of machine-learning to genomic prediction

et al. 2023

View full text Add to dashboard Cite

Current practice in agriculture applies genomic prediction to assist crop breeding in the analysis of genetic marker data. Genomic selection methods typically use linear mixed models, but using machine-learning may provide further potential for improved selection accuracy, or may provide additional information. Here we describe SelectML, an automated pipeline for testing and comparing the performance of a range of linear mixed model and machine-learning-based genomic selection methods. We demonstrate the use of SelectML on an in silico-generated marker dataset which simulated a randomly-sampled (mixed) and an unevenly-sampled (unbalanced) population, comparing the relative performance of various methods included in SelectML on the two datasets. Although machine-learning based methods performed similarly overall to linear mixed models, they performed worse on the mixed dataset and marginally better on the unbalanced dataset, being more affected than linear mixed models by the imposed sampling bias. SelectML can assist in the training, comparison, and selection of genomic selection models, and is available from https://github.com/darcyabjones/selectml.

show abstract

Section: Discussionmentioning

confidence: 88%

The pursuit of genetic gain in agricultural crops through the application of machine-learning to genomic prediction

et al. 2023

View full text Add to dashboard Cite

show abstract

“…The problem of whole-genome predictions of continuous phenotypes proves to be particularly hard to tackle for deep learning, given that state-of-the-art penalized regression models already achieve nearly optimal-predictive ability. An interesting example was provided by Gill et al 39 , who showed that other machine learning methods (i.e. random forest and XGBoost) outperformed deep learning on a soybean dataset, with the extra benefit of model interpretability and data insights.…”

Section: Discussionmentioning

confidence: 99%

Stacked kinship CNN vs. GBLUP for genomic predictions of additive and complex continuous phenotypes

Biscarini

2022

Sci Rep

View full text Add to dashboard Cite

Deep learning is impacting many fields of data science with often spectacular results. However, its application to whole-genome predictions in plant and animal science or in human biology has been rather limited, with mostly underwhelming results. While most works focus on exploring alternative network architectures, in this study we propose an innovative representation of marker genotype data and tested it against the GBLUP (Genomic BLUP) benchmark with linear and nonlinear phenotypes. From publicly available cattle SNP genotype data, different types of genomic kinship matrices are stacked together in a 3D pile from where 2D grayscale slices are extracted and fed to a deep convolutional neural network (DNN). We simulated nine phenotype scenarios with combinations of additivity, dominance and epistasis, and compared the DNN to GBLUP-A (computed using only the additive kinship matrix) and GBLUP-optim (additive, dominance, and epistasis kinship matrices, as needed). Results varied depending on the accuracy metric employed, with DNN performing better in terms of root mean squared error (1–12% lower than GBLUP-A; 1–9% lower than GBLUP-optim) but worse in terms of Pearson’s correlation (0.505 for DNN compared to 0.672 and 0.669 of GBLUP-A and GBLUP-optim for fully additive case; 0.274 for DNN, 0.279 for GBLUP-A, and 0.477 for GBLUP-optim for fully dominant case). The proposed approach offers a basis to explore further the application of DNN to tabular data in whole-genome predictions.

show abstract

“…When lecturers discuss lesson plans, things they normally have to deal with are creating an atmosphere conducive to collaboration, for example by suggesting that all lecturers be given the opportunity to express their opinions instead of being asked for them (Wahyuni, 2015). Choosing the right implementation model can have a significant impact on the efficiency and effectiveness of learning (Mulyasa, 2011;Sedgewick et al, 2016;& Gill et al, 2022).…”

Section: Discussionmentioning

confidence: 99%

Implementation of Lesson Study in Learning Process: A Study of Biology Student Learning Activities

Akbar

Rohma²,

Ikalor³

2022

Journal Of Biology Education Research (JBER)

View full text Add to dashboard Cite

Prospective student educators can use Lesson Study to improve their ability to plan and design learning. However, understanding of the implementation of Lesson Study in the learning process by 1st semester students at the Postgraduate Biology Education Study Program, Universitas Negeri Malang are still lacking. Thus, a study or research is necessary for the implementation of Lesson Study by students. This study aims to determine student learning activities through the application of Lesson Study and as a form of training for students to transform their knowledge as prospective educators. Descriptive qualitative research is the type of research conducted here. There were three stages to this research, namely planning, implementing, and reflecting. The data obtained from this study are the implementation of Lesson Study activities with the application of the Jigsaw type cooperative learning model, the implementation of Lesson Study activities with the application of the STAD cooperative learning model combined with Snowball Throwing, student learning activities, and the results of reflection on the implementation of Lesson Study. According to the study's findings, Lesson Study can accurately reflect student activities and its implementation in the learning process is also able to provide prospective educators with an understanding of pedagogical competence.

show abstract

Machine learning models outperform deep learning models, provide interpretation and facilitate feature selection for soybean trait prediction

Cited by 27 publications

References 56 publications

The pursuit of genetic gain in agricultural crops through the application of machine-learning to genomic prediction

The pursuit of genetic gain in agricultural crops through the application of machine-learning to genomic prediction

Stacked kinship CNN vs. GBLUP for genomic predictions of additive and complex continuous phenotypes

Implementation of Lesson Study in Learning Process: A Study of Biology Student Learning Activities

Contact Info

Product

Resources

About