The predictive accuracies of cor of prediction and ture values are different between two R packages

YinLiLin / hibayes

:swimmer: Individual-Level, Summary-Level and Single-Step Bayesian Regression Models for Genomic Prediction and Genome-Wide Association Studies

GNU General Public License v3.0

46 stars 7 forks source link

The predictive accuracies of cor of prediction and ture values are different between two R packages #8

Closed Xinin-loveScience closed 2 years ago

Xinin-loveScience commented 3 years ago

Hello! The packages "hibayes" and "G2P" were performed to GS of weight trait, and the predictive accuracies were significantly different. The accuracies were 0.82 and 0.15 by BayesB in hibayes and G2P, respectively. I don't know the reasons.

YinLiLin commented 3 years ago

Hi,

Thank you for using hibayes. I am not familiar with G2P package you mentioned, I suggest to try BGLR package to fit Bayes model if you want to make a comparison.

Xinin-loveScience commented 3 years ago

Hi, Thanks for your reply. Do you call the BGLR package? I use the G2P is called BGLR package to calculate bayes.

YinLiLin commented 3 years ago

I just quickly searched the G2P package, it seems to be a tool enriched with lots of models for genomic selection by integrating some widely-used packages. Except for BGLR, you can also try GCTB to fit varies Bayes models for a comparison.

Xinin-loveScience commented 3 years ago

谢谢，我用了你说的这个GCTB做了BayesB，但是输出文件没有发现育种值？

YinLiLin commented 3 years ago

GCTB只输出标记效应值，需要结合plink计算个体育种值

Xinin-loveScience commented 3 years ago

谢谢，有具体的plink计算育种值的方法吗？

YinLiLin commented 3 years ago

请参考plink --score功能