Environmental covariate setting

[yan-cheng-lin]

Hello Deruncie,

First of all, thank you for developing such a powerful tool!

I have a question regarding the settings for environmental covariates in priors. I am conducting genomic prediction in multi-environment trials and have included environmental covariates like precipitation and temperature.

For some trials, I have irrigation management information but lack data from soil sensors. Can I provide this information as a factor to X_Env, or do I need to use a dummy variable (0, 1) and assign it to the same X_Env_groups?

Thank you in advance for your guidance.

Best regards,
Yan-Cheng

[deruncie]

Hi Yan-Cheng, Thanks for the question. I think partial environmental covariate data is a challenge that we haven’t fully explored yet. MegaLMM does require that each environmental covariate has a value for every trial. I think the best choice might be to try to model the soil data that you miss based on any other environment data that you have across the other trials, and impute the missing values that way. If that doesn’t work, then simply imputing these missing values with the mean of the remaining values is your best bet. Dan From: Yan-Cheng ***@***.***> Date: Tuesday, January 21, 2025 at 3:13 AM To: deruncie/MegaLMM ***@***.***> Cc: Subscribed ***@***.***> Subject: [deruncie/MegaLMM] Environmental covariate setting (Issue #12) Hello Deruncie, First of all, thank you for developing such a powerful tool! I have a question regarding the settings for environmental covariates in priors. I am conducting genomic prediction in multi-environment trials and have included environmental covariates like precipitation and temperature. For some trials, I have irrigation management information but lack data from soil sensors. Can I provide this information as a factor to X_Env, or do I need to use a dummy variable (0, 1) and assign it to the same X_Env_groups? Thank you in advance for your guidance. Best regards, Yan-Cheng — Reply to this email directly, view it on GitHub<#12>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AB2IDTNNC4ZL2K5MHB37LMT2LYTWLAVCNFSM6AAAAABVSIIKQGVHI2DSMVQWIX3LMV43ASLTON2WKOZSHAYDCNJQGE3DMMI>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>