Parse OMIM Entry Names

Parses OMIM entry names — often listed in all-uppercase with an "inverted filing" convention (primary disease term first, then comma-separated qualifiers) and an optional semicolon-separated abbreviation — into case- and order-normalized names and separate abbreviations. Mixed-case input (e.g. from genemap2.txt) is also accepted.

Usage

parse_omim_name(
  x,
  col = NULL,
  eponyms = disease_eponyms,
  patterns = disease_cap_patterns
)

Arguments

x

A character vector of OMIM entry name strings (e.g. "SPASTIC PARAPLEGIA 14, AUTOSOMAL RECESSIVE; SPG14"), or a data frame with such a column.

col

[data.frame only] The name of the column in x that contains OMIM entry name strings, passed as a string (e.g. "entry").

eponyms

A named character vector for proper noun (eponym) capitalization, where names are lowercase words and values are their capitalized replacements (e.g. c("waardenburg" = "Waardenburg")). Applied as whole-word, case-insensitive substitutions after lowercasing. Defaults to disease_eponyms. Pass NULL to disable, or supply your own vector to override.

patterns

A named character vector of phrase-level regex substitutions applied to the full lowercased name after eponyms, longest-first. Useful for context-sensitive capitalization (e.g. c("short syndrome" = "SHORT syndrome")). Defaults to disease_cap_patterns. Pass NULL to disable.

Value

• Character vector input: a tibble with columns:

  • name: normalized name after rearrangement and capitalization fixes.

  • abbreviation: the semicolon-separated abbreviation, or NA.

• Data frame input: the input data frame with name and abbreviation appended (or replaced if already present).

Details

Reverses OMIM\'s inverted filing convention by reclassifying comma-separated qualifier tokens (inheritance terms, type codes, onset adjectives, etc.) and reassembling them in natural-language order. Rearrangement is only applied when at least one qualifier matches a recognized forcing pattern; otherwise tokens are kept in original order and lowercased. See the algorithm article for full details on qualifier classification, capitalization rules, and known limitations.

See also

disease_eponyms for the curated eponym replacement vector; disease_cap_patterns for the curated phrase pattern replacement vector; the algorithm article for full details on how parsing and rearrangement work.

Examples

parse_omim_name(c(
    "SCHIZOPHRENIA 12",
    "DYSTONIA 12; DYT12",
    "SPASTIC PARAPLEGIA 14, AUTOSOMAL RECESSIVE; SPG14",
    "OSTEOGENESIS IMPERFECTA, TYPE XI; OI11",
    "SCOLIOSIS, ISOLATED, SUSCEPTIBILITY TO, 1; IS1",
    "EPILEPSY, PROGRESSIVE MYOCLONIC, 4, WITH OR WITHOUT RENAL FAILURE; EPM4"
))
#> # A tibble: 6 × 2
#>   name                                                           abbreviation
#>   <chr>                                                          <chr>       
#> 1 schizophrenia 12                                               NA          
#> 2 dystonia 12                                                    DYT12       
#> 3 autosomal recessive spastic paraplegia 14                      SPG14       
#> 4 osteogenesis imperfecta type XI                                OI11        
#> 5 susceptibility to isolated scoliosis 1                         IS1         
#> 6 progressive myoclonic epilepsy 4 with or without renal failure EPM4        

# Data frame input
df <- data.frame(entry = c("HURIEZ SYNDROME; HRZ", "SCHWANNOMATOSIS, VESTIBULAR; SWNV"))
parse_omim_name(df, col = "entry")
#>                               entry                       name abbreviation
#> 1              HURIEZ SYNDROME; HRZ            Huriez syndrome          HRZ
#> 2 SCHWANNOMATOSIS, VESTIBULAR; SWNV vestibular schwannomatosis         SWNV

# Proper noun correction via custom eponyms (overrides disease_eponyms)
parse_omim_name(
    "PERIPHERAL DEMYELINATING NEUROPATHY, CENTRAL DYSMYELINATION, WAARDENBURG SYNDROME, AND HIRSCHSPRUNG DISEASE; PCWH",
    eponyms = c("waardenburg" = "Waardenburg", "hirschsprung" = "Hirschsprung")
)
#> # A tibble: 1 × 2
#>   name                                                              abbreviation
#>   <chr>                                                             <chr>       
#> 1 peripheral demyelinating neuropathy, central dysmyelination, Waa… PCWH        

# Context-sensitive capitalization via phrase patterns (overrides eponyms)
parse_omim_name(
    "SHORT SYNDROME; SHORTSYN",
    patterns = c("short syndrome" = "SHORT syndrome")
)
#> # A tibble: 1 × 2
#>   name           abbreviation
#>   <chr>          <chr>       
#> 1 SHORT syndrome SHORTSYN