Comments (5)
Hi, thank you for your report.
Could you please pass the column BLOCK to factor with dataset <- to_factor(dataset, BLOCK)
,and report if any warning is generated when you run inspect(data)
?
Is there any missing value in the data? Please, use has_na(data)
.
Based on the levels of your experiment the expected number of rows would be 1170 (13 x 30 x 3), Am I right? Maybe have you unbalanced data? Try to make a two-way table with make_mat(dataset, ENV, GEN, GY)
and see if any cell has NA
.
You can use the package reprex to report the results here
Regards!
from metan.
@billh2006 I closed accidentally this issue. I added a commit that improves the check of data before computing the cross-validation. Please, install the development version and check with your data.
from metan.
@TiagoOlivoto- Thanks for your quick response. Yes-my data set is unbalanced. Certain genotypes are not represented in all environments and 8 (thank you for the new error message) of the genotype-environment interactions are missing data for one or more of the reps. Is a complete, balanced dataset a necessary condition for cv_blup and downstream functions (e.g. waasb)?
from metan.
@billh2006 - You can have unbalanced data but it is necessary that all genotype-environment combinations have the same number of replicates (complete dataset). If you assume to have three replicates, for each cross-validation procedure the function will select randomly two replicates for each genotype-environment combination to serve as training set. The other replicate will serve as validation set. If for some GEI combinations you have only two reps, the model cannot be computed because there will be no validation data for such a combination. Please, note a toy example bellow with the example data data_ge
library(metan)
#> Registered S3 method overwritten by 'GGally':
#> method from
#> +.gg ggplot2
#> []=====================================================[]
#> [] Multi-Environment Trial Analysis (metan) v1.4.0.9000[]
#> [] Author: Tiago Olivoto []
#> [] Type citation('metan') to know how to cite metan []
#> [] Type vignette('metan_start') for a short tutorial []
#> [] Visit http://bit.ly/2TIq6JE for a complete tutorial []
#> []=====================================================[]
df <- data_ge
df[1, 4] <- NA
df
#> # A tibble: 420 x 5
#> ENV GEN REP GY HM
#> <fct> <fct> <fct> <dbl> <dbl>
#> 1 E1 G1 1 NA 44.9
#> 2 E1 G1 2 2.50 46.9
#> 3 E1 G1 3 2.43 47.8
#> 4 E1 G2 1 3.21 45.2
#> 5 E1 G2 2 2.93 45.3
#> 6 E1 G2 3 2.56 45.5
#> 7 E1 G3 1 2.77 46.7
#> 8 E1 G3 2 3.62 43.2
#> 9 E1 G3 3 2.28 47.8
#> 10 E1 G4 1 2.36 47.9
#> # ... with 410 more rows
cv_blup(df, env = ENV, gen = GEN, rep = REP, resp = GY, nboot = 10)
#> ENV GEN n
#> 1 E1 G1 2
#> Error: Combinations of genotype and environment with different number of replication than observed in the trial (3)
# An unbalanced data
df2 <- data_ge
df2[1:3, 4] <- NA
df2
#> # A tibble: 420 x 5
#> ENV GEN REP GY HM
#> <fct> <fct> <fct> <dbl> <dbl>
#> 1 E1 G1 1 NA 44.9
#> 2 E1 G1 2 NA 46.9
#> 3 E1 G1 3 NA 47.8
#> 4 E1 G2 1 3.21 45.2
#> 5 E1 G2 2 2.93 45.3
#> 6 E1 G2 3 2.56 45.5
#> 7 E1 G3 1 2.77 46.7
#> 8 E1 G3 2 3.62 43.2
#> 9 E1 G3 3 2.28 47.8
#> 10 E1 G4 1 2.36 47.9
#> # ... with 410 more rows
cv_blup(df2, env = ENV, gen = GEN, rep = REP, resp = GY, nboot = 10)
#> $RMSPD
#> MODEL RMSPD
#> 1 BLUP_g_RCBD 0.3552628
#> 2 BLUP_g_RCBD 0.3763954
#> 3 BLUP_g_RCBD 0.4266109
#> 4 BLUP_g_RCBD 0.4532448
#> 5 BLUP_g_RCBD 0.4043090
#> 6 BLUP_g_RCBD 0.4221250
#> 7 BLUP_g_RCBD 0.4243280
#> 8 BLUP_g_RCBD 0.3874620
#> 9 BLUP_g_RCBD 0.3697422
#> 10 BLUP_g_RCBD 0.3845615
#>
#> $RMSPDmean
#> # A tibble: 1 x 6
#> MODEL mean sd se Q2.5 Q97.5
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 BLUP_g_RCBD 0.400 0.0308 0.00972 0.359 0.447
#>
#> attr(,"class")
#> [1] "cvalidation"
Created on 2020-04-07 by the reprex package (v0.3.0)
You can use both unbalanced and incomplete data with waasb()
and gamem_met()
. Since we don't need a validation set, the function will simply ignore the missing data
library(metan)
#> Registered S3 method overwritten by 'GGally':
#> method from
#> +.gg ggplot2
#> []=====================================================[]
#> [] Multi-Environment Trial Analysis (metan) v1.4.0.9000[]
#> [] Author: Tiago Olivoto []
#> [] Type citation('metan') to know how to cite metan []
#> [] Type vignette('metan_start') for a short tutorial []
#> [] Visit http://bit.ly/2TIq6JE for a complete tutorial []
#> []=====================================================[]
df <- data_ge
df[1, 4] <- NA
df
#> # A tibble: 420 x 5
#> ENV GEN REP GY HM
#> <fct> <fct> <fct> <dbl> <dbl>
#> 1 E1 G1 1 NA 44.9
#> 2 E1 G1 2 2.50 46.9
#> 3 E1 G1 3 2.43 47.8
#> 4 E1 G2 1 3.21 45.2
#> 5 E1 G2 2 2.93 45.3
#> 6 E1 G2 3 2.56 45.5
#> 7 E1 G3 1 2.77 46.7
#> 8 E1 G3 2 3.62 43.2
#> 9 E1 G3 3 2.28 47.8
#> 10 E1 G4 1 2.36 47.9
#> # ... with 410 more rows
mod <- waasb(df, ENV, GEN, REP, GY)
#> Warning: Row(s) 1 with NA values deleted.
#> Method: REML/BLUP
#> Random effects: GEN, GEN:ENV
#> Fixed effects: ENV, REP(ENV)
#> Denominador DF: Satterthwaite's method
#> ---------------------------------------------------------------------------
#> P-values for Likelihood Ratio Test of the analyzed traits
#> ---------------------------------------------------------------------------
#> model GY
#> COMPLETE NA
#> GEN 1.19e-05
#> GEN:ENV 2.02e-11
#> ---------------------------------------------------------------------------
#> All variables with significant (p < 0.05) genotype-vs-environment interaction
gmd(mod)
#> Class of the model: waasb
#> Variable extracted: genpar
#> # A tibble: 9 x 2
#> Parameters GY
#> <chr> <dbl>
#> 1 Phenotypic variance 0.181
#> 2 Heritability 0.154
#> 3 GEIr2 0.313
#> 4 Heribatility of means 0.814
#> 5 Accuracy 0.902
#> 6 rge 0.370
#> 7 CVg 6.24
#> 8 CVr 11.6
#> 9 CV ratio 0.537
Created on 2020-04-07 by the reprex package (v0.3.0)
from metan.
Thank you!
from metan.
Related Issues (20)
- Labels of plot_scores(AMMI_model, type = 2) HOT 1
- Release metan 1.16.0 HOT 1
- error while analysis path analysis HOT 7
- Problem with ge_stats HOT 3
- AMMI Analysis HOT 2
- Shukla Stablility with wrong length of levels HOT 1
- print_table(.) : could not find function "print_table" HOT 1
- Custom names of function arguments do not work HOT 1
- MGIDI selection differential is false if averaged data is negative HOT 2
- error when I run ge_stat() code of metan package HOT 8
- Problem with plot.gge or plot function HOT 2
- Error in corr_coef(data, ..., type = c("linear", "partial"), method = c("pearson", : could not find function "corr_coef" HOT 2
- Error in path_coeff(datt, resp = SYPPI) : could not find function "path_coeff" HOT 2
- Same code of "metan" running on on one dataset, but not on similar second dataset HOT 2
- Hello HOT 2
- metan temporarily out of CRAN HOT 1
- Mislabeling of ENV and GEN in Which Won Where Biplot HOT 6
- rep and block arguments for gge HOT 1
- problem in running your example code
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from metan.