A stacking ensemble learning framework for genomic prediction

Liang, Mang; Chang, Tianpeng; An, Bingxing; Duan, Xinghai; Du, Lili; Wang, Xiaoqiao; Jiang, Miao; Xu, Lingyang; Gao, Xue; Zhang, Lupei; Li, Junya; Gao, Huijiang

doi:10.21203/rs.3.rs-52592/v1

Cited by 3 publications

(3 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A tenfold cross-validation for training the model was used which was also reported to be the most appropriate by 15 . However, 23 used 20-fold cross-validation for their study to predict breeding values.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation and Ranking of Artificial Intelligence Algorithms for Performance Prediction in Sheep

Hamadani

Ganai

2023

Preprint

View full text Add to dashboard Cite

In a rapidly transforming world, farm data is growing exponentially. Realizing the importance of this data, researchers are looking for new solutions to analyse this data and make farming predictions. Considering all this, this research was undertaken to evaluate machine learning (ML) algorithms as potential tools for evaluating 52-year data for sheep. Data was appropriately prepared was done before analysis. Winsorization was done for outlier removal. Principal component regression (PCA) and feature selection (FS) were done and based on that, three datasets were created viz. PCA, PCA+ FS, and FS bodyweight prediction. Among the 11 ML algorithms that were evaluated, the correlations between true and predicted values for MARS algorithm, Bayesian ridge regression, Ridge regression, Support Vector Machines, Gradient boosting algorithm, Random forests, XgBoost algorithm, Artificial neural networks, Classification and regression trees, Polynomial regression, K nearest neighbours and Genetic Algorithms were 0.993, 0.992, 0.991,0.991,0.991, 0.99, 0.99,0.984,0.984,0.957, 0.949.0.734 respectively for bodyweights. The top five algorithms for the prediction of bodyweights, were MARS, Bayesian ridge regression, Ridge regression, Support Vector Machines and Gradient boosting algorithm. A total of 12 machine learning models were developed for the prediction of bodyweights in sheep in the present study. It may be said that machine learning techniques can perform predictions with reasonable accuracies and can thus be viable alternatives to conventional strategies.

show abstract

Section: Discussionmentioning

confidence: 99%

“…It means that we need to find new and innovative approaches to produce more food. This is a huge challenge for animal scientists despite a vast genetic wealth 23 . To address this, new technologies are seeping into the farms which are evolving from traditional to high-tech 4 .…”

Section: Introductionmentioning

confidence: 99%

Evaluation and Ranking of Artificial Intelligence Algorithms for Performance Prediction in Sheep

Hamadani

Ganai

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…After the first application of laser ablation inductively coupled plasma-mass spectrometry (LA-ICP-MS) to U-Pb geochronology (Fryer et al 1993), the technique has been mainly used to determine the U-Pb ages of zircons (Crowley et al 2014). However, other U-rich minerals such as monazite (Chang et al 2006), apatite (Chew et al 2014), allanite (Burn et al 2017), xenotime (Liu et al 2011), titanite (Sun et al 2012), davidite (Chipley and Polito 2007), perovskite (Cox and Wilton 2006), garnet (Duan et al 2020), cassiterite (Neymark et al 2018), wolframite (Luo et al 2019), baddeleyite (Wohlgemuth-Ueberwasser et al 2018), rutile (Xia et al 2013), columbite-tantalite (Che et al 2015, calcite (Yokoyama et al 2018), bastnaesite (Yang et al 2014), zirconolite (Wu et al 2010) and haematite (Courtney-Davies et al 2019) have also been dated.…”

mentioning

confidence: 99%

Investigation of the Ablation Behaviour of Andradite‐Grossular Garnets and Rutile with Implications for U‐Pb Geochronology

Jenkins

Goemann

Belousov

et al. 2023

Geostandard Geoanalytic Res

View full text Add to dashboard Cite

U-Pb dating of andradite-grossular garnet (grandite) and rutile by LA-ICP-MS can be used to constrain various metamorphic, metasomatic and igneous geological processes. In this study, we examine and compare the impact of different analytical conditions (fluence, pulse width, laser beam size and ablation frequency) on the ablation crater morphology, ablation rates, down-hole fractionation and U-Pb ages of grandite and rutile samples of different compositions. The shapes of grandite ablation craters suggest the mineral ablates by classical evaporation with significant melting that cannot be eliminated even at fluences just above the ablation threshold. Grandite garnets with higher andradite proportions have faster ablation rates. The overall low U contents of grandite require using large laser beam sizes to obtain acceptable precision of U-Pb ages. At such conditions and crater depths \ 10 lm, fluences of 2.1 and 3.5 J cm -2 , laser pulse width of 5 ns and 20 ns, and ablation frequencies between 3.5 and 6.5 Hz, obtain similar and reproducible ages when the proportion of grossular is \ 35%. Rutile ablation crater morphology shows evidence of melt splashing and thermal stress cracking. They have significant crater bottom features, which increase in relief with lower fluences and a higher number of laser shots, indicating the features are probably energy-related and making higher fluences, such as 5 J cm -2 , necessary for uniform ablation when using 193 nm excimer lasers. The slow ablation rate at low fluences and then steep increase at around 2.0 J cm -2 suggests a transition in the ablation mechanism from exfoliation to classical vaporisation. Crater bottom features and other ablation behaviour vary between samples, which could be related to their difference in colour. Although the down-hole fractionation patterns of the samples are similar at 5 J cm -2 , the U-Pb ages of some samples vary significantly with different analytical conditions and/or measurement sessions, particularly when using laser beam sizes of 30 lm, suggesting differences in mass bias and more variable ablation behaviour. A laser beam size of at least 60 lm is recommended for reproducible U-Pb dating of rutile.

show abstract

Artificial intelligence algorithm comparison and ranking for weight prediction in sheep

Hamadani

Ganai

2023

Sci Rep

View full text Add to dashboard Cite

In a rapidly transforming world, farm data is growing exponentially. Realizing the importance of this data, researchers are looking for new solutions to analyse this data and make farming predictions. Artificial Intelligence, with its capacity to handle big data is rapidly becoming popular. In addition, it can also handle non-linear, noisy data and is not limited by the conditions required for conventional data analysis. This study was therefore undertaken to compare the most popular machine learning (ML) algorithms and rank them as per their ability to make predictions on sheep farm data spanning 11 years. Data was cleaned and prepared was done before analysis. Winsorization was done for outlier removal. Principal component analysis (PCA) and feature selection (FS) were done and based on that, three datasets were created viz. PCA (wherein only PCA was used), PCA+ FS (both techniques used for dimensionality reduction), and FS (only feature selection used) bodyweight prediction. Among the 11 ML algorithms that were evaluated, the correlations between true and predicted values for MARS algorithm, Bayesian ridge regression, Ridge regression, Support Vector Machines, Gradient boosting algorithm, Random forests, XgBoost algorithm, Artificial neural networks, Classification and regression trees, Polynomial regression, K nearest neighbours and Genetic Algorithms were 0.993, 0.992, 0.991, 0.991, 0.991, 0.99, 0.99, 0.984, 0.984, 0.957, 0.949, 0.734 respectively for bodyweights. The top five algorithms for the prediction of bodyweights, were MARS, Bayesian ridge regression, Ridge regression, Support Vector Machines and Gradient boosting algorithm. A total of 12 machine learning models were developed for the prediction of bodyweights in sheep in the present study. It may be said that machine learning techniques can perform predictions with reasonable accuracies and can thus help in drawing inferences and making futuristic predictions on farms for their economic prosperity, performance improvement and subsequently food security.

show abstract

A stacking ensemble learning framework for genomic prediction

Cited by 3 publications

References 32 publications

Evaluation and Ranking of Artificial Intelligence Algorithms for Performance Prediction in Sheep

Evaluation and Ranking of Artificial Intelligence Algorithms for Performance Prediction in Sheep

Investigation of the Ablation Behaviour of Andradite‐Grossular Garnets and Rutile with Implications for U‐Pb Geochronology

Artificial intelligence algorithm comparison and ranking for weight prediction in sheep

Contact Info

Product

Resources

About