Skip to contents

Run quality control on a studbook or pedigree

Source: R/qcStudbook.R
qcStudbook.Rd

Main pedigree curation function that performs basic quality control on pedigree information

Usage

qcStudbook(
  sb,
  minSireAge = NULL,
  minDamAge = NULL,
  minParentAge = lifecycle::deprecated(),
  reportChanges = FALSE,
  reportErrors = FALSE
)

Arguments

sb

A dataframe containing a table of pedigree and demographic information.

The function recognizes the following columns (optional columns will be used if present, but are not required):

  • id — Character vector with Unique identifier for all individuals

  • sire — Character vector with unique identifier for the father of the current id

  • dam — Character vector with unique identifier for the mother of the current id

  • sex — Factor (levels: "M", "F", "U") Sex specifier for an individual

  • birth — Date or NA (optional) with the individual's birth date

  • departure — Date or NA (optional) an individual was sold or shipped from the colony

  • death — date or NA (optional) Date of death, if applicable

  • status — Factor (levels: ALIVE, DECEASED, SHIPPED, UNKNOWN) (optional) Status of an individual (see convertStatusCodes)

  • origin — Character or NA (optional) Facility an individual originated from, if other than ONPRC

  • ancestry — Character or NA (optional) Geographic population to which the individual belongs

  • spf — Character or NA (optional) Specific pathogen-free status of an individual

  • vasxOvx — Character or NA (optional) Indicator of the vasectomy/ovariectomy status of an animal; NA if animal is intact, assume all other values indicate surgical alteration

  • condition — Character or NA (optional) Indicator of the restricted status of an animal. "Nonrestricted" animals are generally assumed to be naive.

  • placeholder — Logical or NA (optional) TRUE marks a made-up stand-in for an unknown parent, FALSE a real animal. TRUE, true, True, T and 1 count as TRUE; FALSE, false, False, F and 0 as FALSE; a blank as unmarked. Any other value is an error (listed in errorLst$invalidPlaceholderRows with reportErrors == TRUE, otherwise the function throws an error naming the rows).

minSireAge

numeric minimum age in years for a male to have sired an offspring. NULL (default) looks up the floor for each sire's species via getSpeciesMinBreedingAge (falling back to 2 years when the species is missing or unknown); a supplied value overrides that floor.

minDamAge

numeric minimum age in years for a female to have borne an offspring. NULL (default) looks up the floor for each dam's species via getSpeciesMinBreedingAge (falling back to 2 years when the species is missing or unknown); a supplied value overrides that floor.

minParentAge

[Deprecated] Deprecated scalar minimum parent age. Supplying it sets both minSireAge and minDamAge; use those sex-specific parameters instead.

reportChanges

logical value that if TRUE, the errorLst contains the list of changes made to the column names. Default is FALSE.

reportErrors

logical value if TRUE will scan the entire file and report back changes made to input and errors in a list of list where each sublist is a type of change or error found. Changes will include column names, case of categorical values (male, female, unknown), etc. Errors will include missing columns, invalid date rows, male dams, female sires, an ID that is both a sire and a dam, duplicated IDs, IDs with invalid characters, invalid placeholder values, and records with one or more parents below minimum age of parents.

The following changes are made to the column names.

  • Column names are converted to all lower case

  • Spaces, periods (".") and underscores ("_") within column names are removed

  • egoid and ego are converted to id

  • sireid is converted to sire

  • damid is converted to dam

  • birthdate and deathdate are converted to birth and death

  • recordstatus, fromcenter and geographicorigin are converted to recordStatus, fromCenter and geographicOrigin

If the dataframe (sb) does not contain the five required columns (id, sire, dam, sex, and birth) and reportErrors == FALSE, the function throws an error by calling stop(). With reportErrors == TRUE the missing names are returned in errorLst$missingColumns.

Animal IDs (id, sire, dam) must not contain a period ("."); other characters are accepted. Periods cause problems across software environments (R column-name and formula parsing, file-name extensions, programming-language namespaces, and regular expressions), so any id, sire, or dam value containing a period is treated as an error. With reportErrors == TRUE the offending values are returned in errorLst$invalidIdChars; otherwise the function throws an error. All automatically generated IDs (see addUIds) honor this rule.

If the id field has the string UNKNOWN (any case), the record is removed. A record whose sire and dam are both NA or UNKNOWN (any case) is kept as a founder; UNKNOWN in either field becomes NA. If only one of the fields sire or dam is missing, the missing parent is replaced with a unique identifier with the form Unnnn, where nnnn represents one of a series of sequential integers representing the number of missing sires and dams right justified in a pattern of 0000. See addUIds function.

The function addParents is used to add records for parents missing their own record in the pedigree.

The function convertSexCodes is used with ignoreHerm == TRUE to convert sex codes according to the following factors of standardized codes:

  • F – replacing "FEMALE" or "2"

  • M – replacing "MALE" or "1"

  • H – replacing "HERMAPHRODITE" or "4", if ignoreHerm == FALSE

  • U – replacing "HERMAPHRODITE" or "4", if ignoreHerm == TRUE

  • U – replacing "UNKNOWN" or "3"

  • U – replacing a missing, blank or unrecognized value

Case and any spaces around a code are ignored. qcStudbook always uses the default ignoreHerm == TRUE, so H is never assigned, although the returned factor keeps the level H.

The function correctParentSex is used to ensure no parent is both a sire and a dam. If this error is detected and reportErrors == FALSE, the function throws an error and halts the program (with reportErrors == TRUE the IDs are returned in errorLst$sireAndDam). It also recodes a female sire as male and a male dam as female; with reportErrors == FALSE this is silent, otherwise the IDs are reported in errorLst$femaleSires and errorLst$maleDams.

The function convertStatusCodes converts status indicators to the following factors of standardized codes. Case of the original status value is ignored.

  • "ALIVE" — replacing "alive", "A" and "1"

  • "DECEASED" — replacing "deceased", "DEAD", "D", "2"

  • "SHIPPED" — replacing "shipped", "sold", "sale", "s", "3"

  • "UNKNOWN" — replacing is.na(status)

  • "UNKNOWN" — replacing "unknown", "U", "4"

The function convertAncestry converts ancestry indicators using regular expressions such that the following conversions are made from character strings that match selected substrings to the following factors.

  • "INDIAN" — replacing "ind" and not "chin"

  • "CHINESE" — replacing "chin" and not "ind"

  • "HYBRID" — replacing "hyb" or "chin" and "ind"

  • "JAPANESE" — replacing "jap"

  • "UNKNOWN" — replacing NA

  • "OTHER" — replacing not matching any of the above

The function convertDate converts character representations of dates in the columns birth, death, departure, and exit to dates using the as.Date function.

The function setExit uses heuristics and the columns death and departure to set exit if it is not already defined.

The function calcAge uses the birth and the exit columns to define the age column. The numerical value is rounded to the nearest 0.1 of a year. If exit is not defined, the current system date (Sys.Date()) is used.

The function findGeneration is used to define the generation number for each animal in the pedigree.

The function removeDuplicates checks for any duplicated records and removes the duplicates. It also throws an error and stops the program if an ID appears in more than one record where one or more of the other columns have a difference.

The standard columns come first, followed by any unrecognized columns, which are retained. The rows are ordered by generation number and then ID.

Finally the columns id, sire, and dam are coerced to character.

The returned pedigree always has a logical placeholder column with no NA: TRUE for the stand-ins made here (see addUIds), the value the input gave for any other row that had one, and for the rest TRUE when the ID looks like a stand-in (see removeAutoGenIds). A center can mark a real animal whose ID looks like a stand-in (for example "U1234") as FALSE.

Value

With reportErrors == FALSE, a data.frame with standardized and quality controlled pedigree information. With reportErrors == TRUE, an nprcgenekeeprErr list of the changes and errors found, or NULL when there are no errors and no column changes to report.

Examples

examplePedigree <- nprcgenekeepr::examplePedigree
ped <- qcStudbook(examplePedigree,
  minSireAge = 2.0, minDamAge = 2.0, reportChanges = FALSE,
  reportErrors = FALSE
)
names(ped)
#>  [1] "id"           "sire"         "dam"          "sex"          "gen"         
#>  [6] "birth"        "exit"         "age"          "ancestry"     "origin"      
#> [11] "status"       "recordStatus" "fromCenter"   "placeholder"