Dataset Information

A joint learning approach for genomic prediction in polyploid grasses.

ABSTRACT: Poaceae, among the most abundant plant families, includes many economically important polyploid species, such as forage grasses and sugarcane (Saccharum spp.). These species have elevated genomic complexities and limited genetic resources, hindering the application of marker-assisted selection strategies. Currently, the most promising approach for increasing genetic gains in plant breeding is genomic selection. However, due to the polyploidy nature of these polyploid species, more accurate models for incorporating genomic selection into breeding schemes are needed. This study aims to develop a machine learning method by using a joint learning approach to predict complex traits from genotypic data. Biparental populations of sugarcane and two species of forage grasses (Urochloa decumbens, Megathyrsus maximus) were genotyped, and several quantitative traits were measured. High-quality markers were used to predict several traits in different cross-validation scenarios. By combining classification and regression strategies, we developed a predictive system with promising results. Compared with traditional genomic prediction methods, the proposed strategy achieved accuracy improvements exceeding 50%. Our results suggest that the developed methodology could be implemented in breeding programs, helping reduce breeding cycles and increase genetic gains.

SUBMITTER: Aono AH

PROVIDER: S-EPMC9304331 | biostudies-literature | 2022 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A joint learning approach for genomic prediction in polyploid grasses.

Aono Alexandre Hild AH Ferreira Rebecca Caroline Ulbricht RCU Moraes Aline da Costa Lima ADCL Lara Letícia Aparecida de Castro LAC Pimenta Ricardo José Gonzaga RJG Costa Estela Araujo EA Pinto Luciana Rossini LR Landell Marcos Guimarães de Andrade MGA Santos Mateus Figueiredo MF Jank Liana L Barrios Sanzio Carvalho Lima SCL do Valle Cacilda Borges CB Chiari Lucimara L Garcia Antonio Augusto Franco AAF Kuroshu Reginaldo Massanobu RM Lorena Ana Carolina AC Gorjanc Gregor G de Souza Anete Pereira AP

Scientific reports 20220721 1

Poaceae, among the most abundant plant families, includes many economically important polyploid species, such as forage grasses and sugarcane (Saccharum spp.). These species have elevated genomic complexities and limited genetic resources, hindering the application of marker-assisted selection strategies. Currently, the most promising approach for increasing genetic gains in plant breeding is genomic selection. However, due to the polyploidy nature of these polyploid species, more accurate model ...[more]

PMID: 35864135

Similar Datasets

Project description:Genomic prediction (GP) is the procedure whereby the genetic merits of untested candidates are predicted using genome wide marker information. Although numerous examples of GP exist in plants and animals, applications to polyploid organisms are still scarce, partly due to limited genome resources and the complexity of this system. Deep learning (DL) techniques comprise a heterogeneous collection of machine learning algorithms that have excelled at many prediction tasks. A potential advantage of DL for GP over standard linear model methods is that DL can potentially take into account all genetic interactions, including dominance and epistasis, which are expected to be of special relevance in most polyploids. In this study, we evaluated the predictive accuracy of linear and DL techniques in two important small fruits or berries: strawberry and blueberry. The two datasets contained a total of 1,358 allopolyploid strawberry (2n=8x=112) and 1,802 autopolyploid blueberry (2n=4x=48) individuals, genotyped for 9,908 and 73,045 single nucleotide polymorphism (SNP) markers, respectively, and phenotyped for five agronomic traits each. DL depends on numerous parameters that influence performance and optimizing hyperparameter values can be a critical step. Here we show that interactions between hyperparameter combinations should be expected and that the number of convolutional filters and regularization in the first layers can have an important effect on model performance. In terms of genomic prediction, we did not find an advantage of DL over linear model methods, except when the epistasis component was important. Linear Bayesian models were better than convolutional neural networks for the full additive architecture, whereas the opposite was observed under strong epistasis. However, by using a parameterization capable of taking into account these non-linear effects, Bayesian linear models can match or exceed the predictive accuracy of DL. A semiautomatic implementation of the DL pipeline is available at https://github.com/lauzingaretti/deepGP/.

Project description:ObjectiveTo build a supervised machine learning-based classifier, which can accurately predict whether Tai Chi practitioners may experience knee pain after years of exercise.DesignA prospective approach was used. Data were collected using face-to-face through a self-designed questionnaire.SettingSingle centre in Shanghai, China.ParticipantsA total of 1750 Tai Chi practitioners with a course of Tai Chi exercise over 5 years were randomly selected.MeasuresAll participants were measured by a questionnaire survey including personal information, Tai Chi exercise pattern and Irrgang Knee Outcome Survey Activities of Daily Living Scale. The validity of the questionnaire was analysed by logical analysis and test, and the reliability of this questionnaire was mainly tested by a re-test method. Dataset 1 was established by whether the participant had knee pain, and dataset 2 by whether the participant's knee pain affected daily living function. Then both datasets were randomly assigned to a training and validating dataset and a test dataset in a ratio of 7:3. Six machine learning algorithms were selected and trained by our dataset. The area under the receiver operating characteristic curve was used to evaluate the performance of the trained models, which determined the best prediction model.ResultsA total of 1703 practitioners completed the questionnaire and 47 were eliminated for lack of information. The total reliability of the scale is 0.94 and the KMO (Kaiser-Meyer-Olkin measure of sampling adequacy) value of the scale validity was 0.949 (>0.7). The CatBoost algorithm-based machine-learning model achieved the best predictive performance in distinguishing practitioners with different degrees of knee pain after Tai Chi practice. 'Having knee pain before Tai Chi practice', 'knee joint warm-up' and 'duration of each exercise' are the top three factors associated with pain after Tai Chi exercise in the model. 'Having knee pain before Tai Chi practice', 'Having Instructor' and 'Duration of each exercise' were most relevant to whether pain interfered with daily life in the model.ConclusionCatBoost-based machine learning classifier accurately predicts knee pain symptoms after practicing Tai Chi. This study provides an essential reference for practicing Tai Chi scientifically to avoid knee pain.

Dataset Information

A joint learning approach for genomic prediction in polyploid grasses.

Publications

A joint learning approach for genomic prediction in polyploid grasses.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets