issue running package example script

[rcorty]

I ran the following script, copied pretty much exactly from the package example script

and was met with:

Error in `inner_join()`:
! Join columns in `y` must be present in the data.
✖ Problem with `vocabulary_id` and `code`.
Traceback:

1. createPhenotypes(ex$id.vocab.code.count, aggregate.fun = sum, 
 .     id.sex = ex$id.sex, vocabulary.map = phecodeX_map, rollup.map = phecodeX_rollup_map, 
 .     sex.restriction = phecodeX_sex)
2. mapCodesToPhecodes(id.vocab.code.index, make.distinct = map.codes.make.distinct, 
 .     vocabulary.map = vocabulary.map, rollup.map = rollup.map) %>% 
 .     transmute(id, code = phecode, index)
3. transmute(., id, code = phecode, index)
4. mapCodesToPhecodes(id.vocab.code.index, make.distinct = map.codes.make.distinct, 
 .     vocabulary.map = vocabulary.map, rollup.map = rollup.map)
5. withCallingHandlers(output <- inner_join(input, vocabulary.map, 
 .     by = c("vocabulary_id", "code")), warning = function(w) {
 .     if (grepl("coercing into character vector", w$message)) {
 .         invokeRestart("muffleWarning")
 .     }
 . })
6. inner_join(input, vocabulary.map, by = c("vocabulary_id", "code"))
7. inner_join.data.frame(input, vocabulary.map, by = c("vocabulary_id", 
 .     "code"))
8. join_mutate(x = x, y = y, by = by, type = "inner", suffix = suffix, 
 .     na_matches = na_matches, keep = keep, multiple = multiple, 
 .     unmatched = unmatched, relationship = relationship, user_env = caller_env())
9. join_cols(x_names = x_names, y_names = y_names, by = by, suffix = suffix, 
 .     keep = keep, error_call = error_call)
10. check_join_vars(by$y, y_names, by$condition, "y", error_call = error_call)
11. abort(bullets, call = error_call)
12. signal_abort(cnd, .file)

devtools::install_github("PheWAS/PheWAS")

library(PheWAS)

ex=generateExample()

phecodeX_labels = read.csv('https://github.com/PheWAS/PhecodeX/raw/main/phecodeX_R_labels.csv')
phecodeX_rollup_map = read.csv('https://github.com/PheWAS/PhecodeX/raw/main/phecodeX_R_rollup_map.csv')
phecodeX_map = read.csv('https://github.com/PheWAS/PhecodeX/raw/main/phecodeX_R_labels.csv')
phecodeX_sex = read.csv('https://github.com/PheWAS/PhecodeX/raw/main/phecodeX_R_sex.csv')

createPhenotypes(ex$id.vocab.code.count,
                 aggregate.fun = sum, 
                 id.sex = ex$id.sex,
                 vocabulary.map = phecodeX_map,
                 rollup.map = phecodeX_rollup_map,
                 sex.restriction = phecodeX_sex)

environment information is:

R version 4.2.2 Patched (2022-11-10 r83330)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 20.04.3 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3
LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/liblapack.so.3

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods  
[8] base     

other attached packages:
[1] PheWAS_0.99.6-1 ggplot2_3.4.4   tidyr_1.3.0     dplyr_1.1.3    

loaded via a namespace (and not attached):
 [1] jsonlite_1.8.7       splines_4.2.2        foreach_1.5.2       
 [4] metafor_4.4-0        ggrepel_0.9.4        numDeriv_2016.8-1.1 
 [7] pillar_1.9.0         backports_1.4.1      lattice_0.20-45     
[10] glue_1.6.2           uuid_1.1-0           digest_0.6.33       
[13] meta_6.5-0           minqa_1.2.6          colorspace_2.1-0    
[16] htmltools_0.5.7      Matrix_1.5-3         pkgconfig_2.0.3     
[19] broom_1.0.5          purrr_1.0.2          scales_1.2.1        
[22] metadat_1.2-0        lme4_1.1-35.1        tibble_3.2.1        
[25] mgcv_1.8-41          generics_0.1.3       DT_0.30             
[28] withr_2.5.2          repr_1.1.4           pan_1.9             
[31] nnet_7.3-18          formula.tools_1.7.1  cli_3.6.1           
[34] survival_3.4-0       magrittr_2.0.3       crayon_1.5.2        
[37] mitml_0.4-5          evaluate_0.23        mice_3.16.0         
[40] fansi_1.0.5          operator.tools_1.6.3 nlme_3.1-160        
[43] MASS_7.3-58.1        xml2_1.3.5           tools_4.2.2         
[46] lifecycle_1.0.4      munsell_0.5.0        glmnet_4.1-8        
[49] compiler_4.2.2       logistf_1.26.0       rlang_1.1.2         
[52] grid_4.2.2           nloptr_2.0.3         pbdZMQ_0.3-8        
[55] iterators_1.0.14     IRkernel_1.3.1       CompQuadForm_1.4.3  
[58] htmlwidgets_1.6.2    base64enc_0.1-3      boot_1.3-28.1       
[61] gtable_0.3.4         codetools_0.2-18     R6_2.5.1            
[64] zoo_1.8-12           fastmap_1.1.1        utf8_1.2.4          
[67] mathjaxr_1.6-0       jomo_2.7-6           shape_1.4.6         
[70] IRdisplay_1.1        Rcpp_1.0.11          vctrs_0.6.4         
[73] rpart_4.1.19         lmtest_0.9-40        tidyselect_1.2.0 

[emilyvansyoc]

I had the same problem. Any updates?

[rcorty]

I think people are just rolling their own phewas scripts these days

[RobertJCarroll]

Apologies for missing this originally- I don't keep an eye on this repo. Two notes:

1. @adlewismbb is working on updating the PheWAS package for CRAN + making it easier to use PhecodeX and other maps.
2. You are loading the wrong file in this line; it loads the labels into the map variable:
   phecodeX_map = read.csv('https://github.com/PheWAS/PhecodeX/raw/main/phecodeX_R_labels.csv')
   It should be:
   phecodeX_map = read.csv('https://github.com/PheWAS/PhecodeX/raw/main/phecodeX_R_map.csv')
   It works as anticipated when that change is made.

[chrisodhams]

Note that for the phecodeX_R_map_ICD_10_WHO.csv file, one needs to reorder and rename the columns for it to work, e.g.:

phecodex_mapicd10 <- read_csv("phecodeX_R_map_ICD_10_WHO.csv", col_names = TRUE) %>% select(code = ICD, vocabulary_id, phecode)

[RobertJCarroll]

Thanks for that note. Order doesn't matter in my tests, which reflects expected behavior, but the name change would be helpful. I've made a PR #5 and will ask about getting the outstanding ones merged.