Last updated: 2021-02-01
Checks: 7 0
Knit directory: PredictOutbredCrossVar/
This reproducible R Markdown analysis was created with workflowr (version 1.6.2). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.
Great! Since the R Markdown file has been committed to the Git repository, you know the exact version of the code that produced these results.
Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.
The command set.seed(20191123) was run prior to running the code in the R Markdown file. Setting a seed ensures that any results that rely on randomness, e.g. subsampling or permutations, are reproducible.
Great job! Recording the operating system, R version, and package versions is critical for reproducibility.
Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.
Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.
Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility.
The results in this page were generated with repository version 883b1d4. See the Past versions tab to see a history of the changes made to the R Markdown and HTML files.
Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated:
Ignored files:
Ignored: .DS_Store
Ignored: .Rhistory
Ignored: .Rproj.user/
Ignored: analysis/.DS_Store
Ignored: code/.DS_Store
Ignored: data/.DS_Store
Ignored: output/.DS_Store
Ignored: output/crossRealizations/.DS_Store
Untracked files:
Untracked: .gitignore
Untracked: Icon
Untracked: PredictOutbredCrossVar.Rproj
Untracked: manuscript/
Untracked: output/crossPredictions/
Untracked: output/gblups_DirectionalDom_parentwise_crossVal_folds.rds
Untracked: output/gblups_geneticgroups.rds
Untracked: output/gblups_parentwise_crossVal_folds.rds
Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.
These are the previous versions of the repository in which changes were made to the R Markdown (analysis/predictionAccuracy.Rmd) and HTML (docs/predictionAccuracy.html) files. If you’ve configured a remote Git repository (see ?wflow_git_remote), click on the hyperlinks in the table below to view the files as they were in that past version.
| File | Version | Author | Date | Message |
|---|---|---|---|---|
| Rmd | 883b1d4 | wolfemd | 2021-02-01 | Update the syntax highlighting and code-block formatting throughout for |
| Rmd | 6a10c30 | wolfemd | 2021-01-04 | Submission and GitHub ready version. |
| html | 6a10c30 | wolfemd | 2021-01-04 | Submission and GitHub ready version. |
Format predicted and observed values so prediction accuracy can be computed.
library(tidyverse); library(magrittr); library(predCrossVar)
predmeans<-readRDS(here::here("output/crossPredictions","predictedCrossMeans_tidy_withSelIndices.rds"))
predmeans_dd<-readRDS(here::here("output/crossPredictions","predictedCrossMeans_DirectionalDom_tidy_withSelIndices.rds"))
#predmeans %>% count(Model,predOf)
#predmeans_dd %>% count(predOf)
predmeans %<>%
bind_rows(predmeans_dd %>%
mutate(Model=ifelse(predOf=="MeanBV","DirDomBV","DirDomAD"))) %>%
#rename(VarComp=predOf) %>%
mutate(predOf=gsub("MeanGV","MeanTGV",predOf))
rm(predmeans_dd)
#predmeans %>% count(Model,VarComp)
obsMeans<-readRDS(here::here("output/crossRealizations","realizedCrossMeans.rds")) %>%
rename(predOf=obsOf) %>%
mutate(Model=ifelse(Model=="DirDom",
ifelse(predOf=="MeanBV","DirDomBV","DirDomAD"),
Model))
#obsMeans %>% count(Model,predOf)
obsMeanBLUPs<-readRDS(here::here("output/crossRealizations","realizedCrossMeans_BLUPs.rds"))
obsVSpredMeans<-bind_rows(left_join(predmeans,obsMeans) %>% mutate(ValidationData="GBLUPs"),
left_join(predmeans,obsMeanBLUPs) %>% mutate(ValidationData="iidBLUPs"))
# obsVSpredMeans %>% count(Model,ValidationData,VarComp) %>% spread(ValidationData,n)# Variances
predvars<-readRDS(here::here("output/crossPredictions","predictedCrossVars_tidy_withSelIndices.rds")) %>%
bind_rows(readRDS(here::here("output/crossPredictions","predictedCrossVars_DirectionalDom_tidy_withSelIndices.rds"))) %>%
select(-Nsegsnps,-totcomputetime) %>%
pivot_longer(cols=c(VPM,PMV),names_to = "VarMethod",values_to = "predVar") %>%
group_by(Repeat,Fold,Model,sireID,damID,Trait1,Trait2,VarMethod) %>%
# sum over VarComps (ModelA = VarA, ModelAD = VarA+VarD)
summarize(predVar=sum(predVar),.groups="drop") %>%
mutate(predOf=ifelse(Model %in% c("A","DirDomBV"),"VarBV",
ifelse(Model %in% c("AD","DirDomAD"),"VarTGV",NA)))
predvars %>%
count(Model,predOf)# A tibble: 4 x 3
Model predOf n
<chr> <chr> <int>
1 A VarBV 99648
2 AD VarTGV 99648
3 DirDomAD VarTGV 99648
4 DirDomBV VarBV 99648obsVars<-readRDS(here::here("output/crossRealizations","realizedCrossVars.rds")) %>%
rename(predOf=obsOf) %>%
mutate(Model=ifelse(Model=="DirDom",
ifelse(predOf=="VarBV","DirDomBV","DirDomAD"),
Model))
obsVars %>% count(Model,predOf)# A tibble: 4 x 3
Model predOf n
<chr> <chr> <int>
1 A VarBV 40374
2 AD VarTGV 40374
3 DirDomAD VarTGV 40374
4 DirDomBV VarBV 40374obsVarBLUPs<-readRDS(here::here("output/crossRealizations","realizedCrossVars_BLUPs.rds"))
obsVSpredVars<-bind_rows(left_join(predvars,obsVars) %>% mutate(ValidationData="GBLUPs"),
left_join(predvars,obsVarBLUPs) %>% mutate(ValidationData="iidBLUPs"))
obsVSpredVars %>% count(Model,ValidationData,predOf)# A tibble: 8 x 4
Model ValidationData predOf n
<chr> <chr> <chr> <int>
1 A GBLUPs VarBV 99648
2 A iidBLUPs VarBV 99648
3 AD GBLUPs VarTGV 99648
4 AD iidBLUPs VarTGV 99648
5 DirDomAD GBLUPs VarTGV 99648
6 DirDomAD iidBLUPs VarTGV 99648
7 DirDomBV GBLUPs VarBV 99648
8 DirDomBV iidBLUPs VarBV 99648For ValidationData==“GBLUPs”, weight by the observed “FamSize”.
# add Family Sizes, for weighted correlations
obsVSpredVars %<>%
left_join(readRDS(file=here::here("output/crossRealizations","realizedCrossMetrics.rds")) %>%
distinct(Repeat,Fold,sireID,damID,FamSize) %>% ungroup())
obsVSpredVars %>% head# A tibble: 6 x 13
Repeat Fold Model sireID damID Trait1 Trait2 VarMethod predVar predOf
<chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <dbl> <chr>
1 Repea… Fold1 A IITA-… IITA… biofo… biofo… PMV 55.4 VarBV
2 Repea… Fold1 A IITA-… IITA… biofo… biofo… VPM 5.86 VarBV
3 Repea… Fold1 A IITA-… IITA… DM DM PMV 4.35 VarBV
4 Repea… Fold1 A IITA-… IITA… DM DM VPM 0.355 VarBV
5 Repea… Fold1 A IITA-… IITA… DM logFY… PMV -0.0762 VarBV
6 Repea… Fold1 A IITA-… IITA… DM logFY… VPM -0.00459 VarBV
# … with 3 more variables: obsVar <dbl>, ValidationData <chr>, FamSize <dbl>For ValidationData==“iidBLUPs”, weight by the number of observed non-missing BLUPs per family per trait.
parentfolds<-readRDS(file = here::here("data","parentwise_crossVal_folds.rds")) %>%
rename(Repeat=id,Fold=id2) %>%
select(Repeat,Fold,testparents)
ped<-readRDS(here::here("data","ped_awc.rds")) %>%
nest(FamilyMembers=FullSampleName)
parentfolds %<>%
mutate(CrossesToPredict=map(testparents,~filter(ped,sireID %in% . | damID %in% .))) %>%
select(-testparents)
indices<-readRDS(file=here::here("data","selection_index_weights_4traits.rds"))
blups<-readRDS(here::here("data","blups_forawcdata.rds")) %>%
select(Trait,blups) %>%
unnest(blups) %>%
select(Trait,germplasmName,BLUP) %>%
spread(Trait,BLUP) %>%
select(germplasmName,all_of(c("DM","logFYLD","MCMDS","TCHART"))) # precaution to ensure consistent column order
crossblups<-parentfolds %>%
unnest(CrossesToPredict) %>%
distinct(sireID,damID,FamilyMembers) %>%
unnest(FamilyMembers) %>%
rename(germplasmName=FullSampleName) %>%
left_join(blups) %>%
select(sireID,damID,germplasmName,all_of(indices$Trait)) %>%
nest(famblups=c(germplasmName,all_of(indices$Trait))) %>%
mutate(stdSI=map(famblups,~as.data.frame(.) %>%
column_to_rownames(var = "germplasmName") %>%
as.matrix(.)%*%indices$stdSI),
biofortSI=map(famblups,~as.data.frame(.) %>%
column_to_rownames(var = "germplasmName") %>%
as.matrix(.)%*%indices$biofortSI))
nObs<-bind_rows(crossblups %>%
select(-famblups) %>%
mutate(stdSI=map_dbl(stdSI,~length(which(!is.na(.)))),
biofortSI=map_dbl(biofortSI,~length(which(!is.na(.))))) %>%
pivot_longer(cols = c(stdSI,biofortSI), names_to = "Trait1", values_to = "Nobs",values_drop_na = TRUE) %>%
mutate(Trait2=Trait1),
crossblups %>%
select(sireID,damID,famblups) %>%
mutate(famblups=map(famblups,function(famblups){
NobsMat<-psych::pairwiseCount(famblups %>% select(-germplasmName),diagonal=TRUE)
NobsMat[lower.tri(NobsMat, diag = F)]<-NA
NobsMat %<>%
as.data.frame %>%
rownames_to_column(var = "Trait1") %>%
pivot_longer(cols = -Trait1, names_to = "Trait2", values_to = "Nobs",values_drop_na = TRUE)
return(NobsMat) })) %>%
unnest(famblups))
rm(parentfolds,ped,indices,blups,crossblups)
# add N obs, for weighted correlations
obsVSpredVars %<>%
left_join(nObs) %>%
mutate(CorrWeight=ifelse(ValidationData=="GBLUPs",FamSize,Nobs))# Usefulness
realizedcrossmetrics<-readRDS(file=here::here("output/crossRealizations","realizedCrossMetrics.rds"))
## Join the predicted means and variances
## Only for Sel Indices
#predvars %>% count(Model,predOf)
#predmeans %>% count(Model,predOf)
predUsefulness<-left_join(predvars %>% # Variances
filter(Trait1 %in% c("stdSI","biofortSI"),
Trait1==Trait2) %>%
rename(Trait=Trait1) %>%
select(-Trait2) %>%
mutate(predOf=gsub("Var","",predOf)),
predmeans %>% # Means
filter(Trait %in% c("stdSI","biofortSI")) %>%
mutate(predOf=gsub("Mean","",predOf))) %>%
mutate(predSD=sqrt(predVar)) %>%
select(-predVar)## Add the realized selection intensities
## Create a variable "Stage" for which there are several applying to "Usefulness" for TGV
predBVs<-predUsefulness %>%
filter(predOf=="BV") %>%
left_join(realizedcrossmetrics %>%
select(Repeat,Fold,Model,sireID,damID,FamSize,realIntensityParent) %>%
rename(realIntensity=realIntensityParent) %>%
mutate(Stage="Parent",
Model=ifelse(Model=="ClassicAD","A","DirDomBV")))
predTGVs<-predUsefulness %>%
filter(predOf=="TGV") %>%
left_join(realizedcrossmetrics %>%
select(Repeat,Fold,Model,sireID,damID,FamSize,contains("realIntensity"),-realIntensityParent) %>%
pivot_longer(cols = contains("realIntensity"),
names_to = "Stage",
values_to = "realIntensity",
names_prefix = "realIntensity") %>%
mutate(Model=ifelse(Model=="ClassicAD","AD","DirDomAD")))
predUsefulness<-bind_rows(predBVs,
predTGVs) %>%
# include a "stage" (=="ConstIntensity")
# where intensity for predicted UC is set to 2.67
bind_rows(predUsefulness %>%
left_join(realizedcrossmetrics %>%
distinct(Repeat,Fold,sireID,damID,FamSize)) %>%
mutate(Stage="ConstIntensity",
realIntensity=2.67))
## Compute predicted UCs
predUsefulness %<>%
dplyr::mutate(predUC=predMean+(realIntensity*predSD))
predUsefulness %>% count(Model,predOf,Stage)# A tibble: 14 x 4
Model predOf Stage n
<chr> <chr> <chr> <int>
1 A BV ConstIntensity 16608
2 A BV Parent 16608
3 AD TGV AYT 16608
4 AD TGV CET 16608
5 AD TGV ConstIntensity 16608
6 AD TGV PYT 16608
7 AD TGV UYT 16608
8 DirDomAD TGV AYT 16608
9 DirDomAD TGV CET 16608
10 DirDomAD TGV ConstIntensity 16608
11 DirDomAD TGV PYT 16608
12 DirDomAD TGV UYT 16608
13 DirDomBV BV ConstIntensity 16608
14 DirDomBV BV Parent 16608predUsefulness %>% filter(!is.na(predUC)) %>% count(Model,predOf,Stage)# A tibble: 14 x 4
Model predOf Stage n
<chr> <chr> <chr> <int>
1 A BV ConstIntensity 16608
2 A BV Parent 1732
3 AD TGV AYT 3728
4 AD TGV CET 14892
5 AD TGV ConstIntensity 16608
6 AD TGV PYT 10328
7 AD TGV UYT 1412
8 DirDomAD TGV AYT 3728
9 DirDomAD TGV CET 14892
10 DirDomAD TGV ConstIntensity 16608
11 DirDomAD TGV PYT 10328
12 DirDomAD TGV UYT 1412
13 DirDomBV BV ConstIntensity 16608
14 DirDomBV BV Parent 1732## Format observed UCs
obsUCgca<-realizedcrossmetrics %>%
select(Repeat,Fold,Model,sireID,damID,contains("realizedUCparent")) %>%
pivot_longer(cols = contains("realizedUCparent"),
names_to = "Trait",
values_to = "obsUC",
names_prefix = "realizedUCparent_",
values_drop_na = T) %>%
mutate(predOf="BV",
Stage="Parent")
#obsUCgca %>% count(VarComp,Stage,Model,Trait)
obsUCtgv<-realizedcrossmetrics %>%
select(Repeat,Fold,Model,sireID,damID,contains("realizedUCat")) %>%
pivot_longer(cols = contains("realizedUCat"),
names_to = "Trait",
values_to = "obsUC",
names_prefix = "realizedUCat",
values_drop_na = T) %>%
separate(Trait,c("Stage","Trait"),"_") %>%
mutate(predOf="TGV")
#obsUCtgv %>% count(VarComp,Stage,Model)
obsUsefulness<-bind_rows(obsUCgca,obsUCtgv)
obsUsefulness %<>%
bind_rows(realizedcrossmetrics %>%
select(Repeat,Fold,Model,sireID,damID,contains("meanTop1pctGEBV")) %>%
pivot_longer(cols = contains("meanTop1pctGEBV"),
names_to = "Trait",
values_to = "obsUC",
names_prefix = "meanTop1pctGEBV_",
values_drop_na = T) %>%
mutate(predOf="BV",
Stage="ConstIntensity")) %>%
bind_rows(realizedcrossmetrics %>%
select(Repeat,Fold,Model,sireID,damID,contains("meanTop1pctGETGV")) %>%
pivot_longer(cols = contains("meanTop1pctGETGV"),
names_to = "Trait",
values_to = "obsUC",
names_prefix = "meanTop1pctGETGV_",
values_drop_na = T) %>%
mutate(predOf="TGV",
Stage="ConstIntensity"))
obsUsefulness %<>%
mutate(Model=ifelse(Model=="ClassicAD",
ifelse(predOf=="BV","A","AD"),
ifelse(predOf=="BV","DirDomBV","DirDomAD")))
# obsUsefulness
# predUsefulness %>% count(Model,predOf,Stage)
# obsUsefulness %>% count(Model,predOf,Stage)
# obsUsefulness %>% filter(!is.na(obsUC))
# predUsefulness %>% filter(!is.na(predUC)) %>% count(Model,predOf,Stage)
#predUsefulness %>% filter(is.na(FamSize)) %>% count(Model,predOf,VarMethod,Stage)obsVSpredUC<-left_join(predUsefulness,obsUsefulness) %>% ungroup()
obsVSpredUC %<>% drop_na(.) %>% ungroup()
obsVSpredUC %>% strtibble [130,616 × 15] (S3: tbl_df/tbl/data.frame)
$ Repeat : chr [1:130616] "Repeat1" "Repeat1" "Repeat1" "Repeat1" ...
$ Fold : chr [1:130616] "Fold1" "Fold1" "Fold1" "Fold1" ...
$ Model : chr [1:130616] "A" "A" "A" "A" ...
$ sireID : chr [1:130616] "IITA-TMS-IBA030075" "IITA-TMS-IBA030075" "IITA-TMS-IBA030075" "IITA-TMS-IBA030075" ...
$ damID : chr [1:130616] "IITA-TMS-IBA940006" "IITA-TMS-IBA940006" "IITA-TMS-IBA940006" "IITA-TMS-IBA940006" ...
$ Trait : chr [1:130616] "biofortSI" "biofortSI" "stdSI" "stdSI" ...
$ VarMethod : chr [1:130616] "PMV" "VPM" "PMV" "VPM" ...
$ predOf : chr [1:130616] "BV" "BV" "BV" "BV" ...
$ predMean : num [1:130616] 4.49 4.49 9.21 9.21 -1.03 ...
$ predSD : num [1:130616] 7.26 2.29 8.82 3.03 7.34 ...
$ FamSize : num [1:130616] 38 38 38 38 7 7 7 7 28 28 ...
$ realIntensity: num [1:130616] 2.32 2.32 2.32 2.32 1.58 ...
$ Stage : chr [1:130616] "Parent" "Parent" "Parent" "Parent" ...
$ predUC : num [1:130616] 21.33 9.81 29.65 16.23 10.57 ...
$ obsUC : num [1:130616] -0.886 -0.886 7.152 7.152 0.163 ...#obsVSpredUC %>% count(Model,predOf,Stage)saveRDS(obsVSpredMeans,here::here("output","obsVSpredMeans.rds"))
saveRDS(obsVSpredVars,here::here("output","obsVSpredVars.rds"))
saveRDS(obsVSpredUC,here::here("output","obsVSpredUC.rds"))rm(list=ls())
library(tidyverse); library(magrittr);
obsVSpredMeans<-readRDS(here::here("output","obsVSpredMeans.rds"))
obsVSpredVars<-readRDS(here::here("output","obsVSpredVars.rds"))
obsVSpredUC<-readRDS(here::here("output","obsVSpredUC.rds"))
# Means
obsVSpredMeans %<>%
drop_na(.) %>%
nest(predVSobs=c(sireID,damID,predMean,obsMean)) %>%
mutate(Accuracy=map_dbl(predVSobs,~cor(.$predMean,.$obsMean,use = 'complete.obs'))) %>%
select(-predVSobs)
# Variances
obsVSpredVars %<>%
drop_na(.) %>%
select(-FamSize,-Nobs) %>%
nest(predVSobs=c(sireID,damID,predVar,obsVar,CorrWeight)) %>%
mutate(AccuracyWtCor=map_dbl(predVSobs,~psych::cor.wt(.[,3:4],w = .$CorrWeight) %$% r[1,2]),
AccuracyCor=map_dbl(predVSobs,~cor(.$predVar,.$obsVar,use = 'complete.obs'))) %>%
select(-predVSobs)
# Usefulness
obsVSpredUC %<>%
select(-predMean,-predSD,-realIntensity) %>%
nest(predVSobs=c(sireID,damID,predUC,obsUC,FamSize)) %>%
mutate(AccuracyWtCor=map_dbl(predVSobs,~psych::cor.wt(.[,3:4],w = .$FamSize) %$% r[1,2]),
AccuracyCor=map_dbl(predVSobs,~cor(.$predUC,.$obsUC,use = 'complete.obs'))) %>%
select(-predVSobs)
obsVSpredUC %>% count(Model,predOf,Stage)# A tibble: 14 x 4
Model predOf Stage n
<chr> <chr> <chr> <int>
1 A BV ConstIntensity 100
2 A BV Parent 100
3 AD TGV AYT 100
4 AD TGV CET 100
5 AD TGV ConstIntensity 100
6 AD TGV PYT 100
7 AD TGV UYT 100
8 DirDomAD TGV AYT 100
9 DirDomAD TGV CET 100
10 DirDomAD TGV ConstIntensity 100
11 DirDomAD TGV PYT 100
12 DirDomAD TGV UYT 100
13 DirDomBV BV ConstIntensity 100
14 DirDomBV BV Parent 100saveRDS(obsVSpredMeans,here::here("output","accuraciesMeans.rds"))
saveRDS(obsVSpredVars,here::here("output","accuraciesVars.rds"))
saveRDS(obsVSpredUC,here::here("output","accuraciesUC.rds"))
sessionInfo()R version 4.0.2 (2020-06-22)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Catalina 10.15.7
Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRblas.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRlapack.dylib
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] predCrossVar_0.1.0 magrittr_2.0.1 forcats_0.5.1 stringr_1.4.0
[5] dplyr_1.0.3 purrr_0.3.4 readr_1.4.0 tidyr_1.1.2
[9] tibble_3.0.6 ggplot2_3.3.3 tidyverse_1.3.0
loaded via a namespace (and not attached):
[1] Rcpp_1.0.6 lattice_0.20-41 lubridate_1.7.9.2 here_1.0.1
[5] ps_1.5.0 assertthat_0.2.1 rprojroot_2.0.2 digest_0.6.27
[9] psych_2.0.12 utf8_1.1.4 R6_2.5.0 cellranger_1.1.0
[13] backports_1.2.1 reprex_1.0.0 evaluate_0.14 httr_1.4.2
[17] pillar_1.4.7 rlang_0.4.10 readxl_1.3.1 rstudioapi_0.13
[21] whisker_0.4 rmarkdown_2.6 munsell_0.5.0 broom_0.7.4
[25] compiler_4.0.2 httpuv_1.5.5 modelr_0.1.8 xfun_0.20
[29] pkgconfig_2.0.3 mnormt_2.0.2 tmvnsim_1.0-2 htmltools_0.5.1.1
[33] tidyselect_1.1.0 workflowr_1.6.2 fansi_0.4.2 crayon_1.3.4
[37] dbplyr_2.0.0 withr_2.4.1 later_1.1.0.1 grid_4.0.2
[41] nlme_3.1-151 jsonlite_1.7.2 gtable_0.3.0 lifecycle_0.2.0
[45] DBI_1.1.1 git2r_0.28.0 scales_1.1.1 cli_2.3.0
[49] stringi_1.5.3 fs_1.5.0 promises_1.1.1 xml2_1.3.2
[53] ellipsis_0.3.1 generics_0.1.0 vctrs_0.3.6 tools_4.0.2
[57] glue_1.4.2 hms_1.0.0 parallel_4.0.2 yaml_2.2.1
[61] colorspace_2.0-0 rvest_0.3.6 knitr_1.31 haven_2.3.1