README.md

In maia-sh/ctregistries: Data Package and R Functions for Clinical Trial Registries and other MEDLINE Databanks

ctregistries

In Development

The goal of ctregistries is to facilitate the detection and analysis of clinical trial registration numbers. ctregistries is primarily a data package of regular expressions (regexes) and provides some R functions for implementing the regexes.

Regular expressions were developed for trial registration numbers (TRN) from World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) Primary Registries (https://www.who.int/ictrp/network/primary/en/) and MEDLINE Databank Sources (https://www.nlm.nih.gov/bsd/medline_databank_source.html).

Additional, non-trial databanks indexed by MEDLINE (e.g., figshare) are also included, without regexes.

Installation

You can install the development version of ctregistries from GitHub:

# install.packages("devtools")
devtools::install_github("maia-sh/ctregistries")

Dataset

ctregistries provides the registries dataframe with regular expressions for each registry. The registries dataframe is a subset of the larger databanks dataframe which additionally includes non-trial databanks indexed by MEDLINE (e.g., figshare) without regexes. registries is created by filtering databanks for databank_type == "registry.

library(ctregistries)

head(registries) %>% knitr::kable()

| registry | databank_full_name | medline_start_date | databank_type | trn_regex | medline_si | who_ictrp_primary_registry | registry_website | | :----------------- | :------------------------------------------------------- | -------------------: | :------------- | :----------------- | :---------- | :---------------------------- | :----------------------------------------------------------- | | ANZCTR | Australian New Zealand Clinical Trials Registry | 2014 | registry | (?i)(ACTRN|ANZCTR) | TRUE | TRUE | https://www.anzctr.org.au/ | | ChiCTR | Chinese Clinical Trials Registry | 2014 | registry | (?i)ChiCTR(|) | TRUE | TRUE | http://www.chictr.org.cn/ | | CRiS | Clinical Research Information Service, Republic of Korea | 2014 | registry | (?i)KCT | TRUE | TRUE | http://cris.nih.go.kr/cris/en/use_guide/cris_introduce.jsp | | ClinicalTrials.gov | ClinicalTrials.gov Database (NIH/NLM) | 2005 | registry | (?i)NCT | TRUE | FALSE | https://clinicaltrials.gov/ | | CTRI | Clinical Trials Registry - India | 2014 | registry | (?i)CTRI/// | TRUE | TRUE | http://ctri.nic.in/ | | DRKS | German Clinical Trials Register | 2014 | registry | (?i)DRKS | TRUE | TRUE | http://www.germanctr.de/ |

databanks$databank
#>  [1] "ANZCTR"             "ChiCTR"             "CRiS"              
#>  [4] "ClinicalTrials.gov" "CTRI"               "DRKS"              
#>  [7] "EudraCT"            "IRCT"               "ISRCTN"            
#> [10] "JapicCTI"           "JMACCT"             "JPRN"              
#> [13] "jRCT"               "LBCTR"              "NTR"               
#> [16] "PACTR"              "ReBec"              "REPEC"             
#> [19] "RPCEC"              "SLCTR"              "TCTR"              
#> [22] "UMIN-CTR"           "BioProject"         "dbGaP"             
#> [25] "dbSNP"              "dbVar"              "Dryad"             
#> [28] "figshare"           "GDB"                "GENBANK"           
#> [31] "GEO"                "OMIM"               "PDB"               
#> [34] "PIR"                "PubChem-BioAssay"   "PubChem-Compound"  
#> [37] "PubChem-Substance"  "RefSeq"             "SRA"               
#> [40] "SWISSPROT"          "UniMES"             "UniParc"           
#> [43] "UniProtKB"          "UniRef"

Functions

ctregistries provides some functions implementing the registries dataset to detect trial registration numbers and registries in both vectors and dataframes.

library(ctregistries)

# Check whether there is a TRN
has_trn(c("NCT00312962", "hello", "euctr2020-001808-42", NA))
#> [1]  TRUE FALSE  TRUE    NA

# Extract the TRNs
which_trn("NCT00312962 and euctr2020-001808-42")
#> [1] "NCT00312962"    "2020-001808-42"
which_trns(c("NCT00312962", "hello", "euctr2020-001808-42", NA))
#> [1] "NCT00312962"    NA               "2020-001808-42" NA

# Identify the registry
which_registry("NCT00312962 and euctr2020-001808-42")
#> [1] "ClinicalTrials.gov" "EudraCT"
which_registries(c("NCT00312962", "hello", "euctr2020-001808-42", NA))
#> [1] "ClinicalTrials.gov" NA                   "EudraCT"           
#> [4] NA

# Add the trn and registry to a dataframe
mutate_trn_registry(sample_trn_df, text)
#> # A tibble: 7 x 5
#>      id text                               registry_guess   registry    trn     
#>   <dbl> <chr>                              <chr>            <chr>       <chr>   
#> 1     1 NCT00312962                        clinicaltrials.… ClinicalTr… NCT0031…
#> 2     2 hello                              <NA>             <NA>        <NA>    
#> 3     3 <NA>                               ChiCTR           <NA>        <NA>    
#> 4     4 euctr2020-001808-42                EudraCT          EudraCT     2020-00…
#> 5     5 German Clinical Trial Registry Id… DRKS             DRKS        DRKS000…
#> 6     6 ClinicalTrials.gov number, NCT002… ISRCTN           ClinicalTr… NCT0026…
#> 7     6 ClinicalTrials.gov number, NCT002… ISRCTN           ISRCTN      ISRCTN7…

To Do

• [ ] Add checks within functions
  • [ ] Check input variable type
  • [ ] Check input (column) name conflicts (e.g., databank, trn_regex)
• [ ] Add tests: use_testthat()
• [ ] Consider removing check_registry since just comparing strings…
• [ ] Change license (CCBY or AGPL, unclear since partially data package)
  • https://r-pkgs.org/lsicense.html
  • https://thinkr-open.github.io/licensing-r/rlicense.html
  • https://kbroman.org/pkg_primer/pages/licenses.html
• [ ] Get functions to work with multiple and no trns (character(0) and NA) in one element
  • rapply(\~ifelse(. == 1,NA,.), how = “replace”)
  • purrr::modify_if()
  • any(lengths(y) > 1) # flatten if max 1 trn per element…but issue with character(0)

maia-sh/ctregistries documentation built on July 20, 2024, 8:08 p.m.