Package {pedFamilias}

Type:	Package
Title:	Import and Export 'Familias' Files
Version:	0.3.0
Description:	Tools for exchanging pedigree data between the 'pedsuite' packages and the 'Familias' software for forensic kinship computations (Egeland et al. (2000) <doi:10.1016/s0379-0738(00)00147-x>). These functions were split out from the 'forrel' package to streamline maintenance and provide a lightweight alternative for packages otherwise independent of 'forrel'.
License:	GPL (≥ 3)
URL:	https://github.com/magnusdv/pedFamilias
BugReports:	https://github.com/magnusdv/pedFamilias/issues
Depends:	pedtools (≥ 2.11.1), R (≥ 4.2)
Imports:	pedmut
Suggests:	testthat
Encoding:	UTF-8
Language:	en-GB
Config/roxygen2/version:	8.1.1
NeedsCompilation:	no
Packaged:	2026-11-11 21:35:01 UTC; magnu
Author:	Magnus Dehli Vigeland [aut, cre]
Maintainer:	Magnus Dehli Vigeland <m.d.vigeland@medisin.uio.no>
Repository:	CRAN
Date/Publication:	2026-11-12 10:40:18 UTC

pedFamilias: Import and Export 'Familias' Files

Description

Tools for exchanging pedigree data between the 'pedsuite' packages and the 'Familias' software for forensic kinship computations (Egeland et al. (2000) doi:10.1016/s0379-0738(00)00147-x). These functions were split out from the 'forrel' package to streamline maintenance and provide a lightweight alternative for packages otherwise independent of 'forrel'.

Author(s)

Maintainer: Magnus Dehli Vigeland m.d.vigeland@medisin.uio.no (ORCID)

Authors:

• Magnus Dehli Vigeland m.d.vigeland@medisin.uio.no (ORCID)

See Also

Useful links:

• https://github.com/magnusdv/pedFamilias

• Report bugs at https://github.com/magnusdv/pedFamilias/issues

Convert Familias R objects to ped

Description

Convert pedigrees and marker data from the Familias R package into the ped format used by the pedsuite.

Usage

Familias2ped(
  familiasped,
  datamatrix,
  loci,
  locusAttributes = readFamiliasLoci(loci),
  matchLoci = FALSE,
  prefixAdded = "added_"
)

readFamiliasLoci(loci)

Arguments

Details

The definition of a pedigree in Familias is more liberal than that implemented in the pedsuite, which requires that each ped object is a connected pedigree, and that each member has either 0 or 2 parents. The conversion function Familias2ped takes care of all potential differences. Specifically, it converts each FamiliasPedigree object into a list of connected ped components, and adds missing parents when needed.

Value

A ped object, or a list of such.

References

Familias is freely available from https://familias.name.

See Also

readFam().

Examples

famPed = list(id = c('mother', 'daughter', 'AF'),
              findex = c(0, 3, 0),
              mindex = c(0, 1, 0),
              sex = c('female', 'female', 'male'))
class(famPed) = "FamiliasPedigree"

datamatrix = data.frame(
  M1.1 = c(NA, 8, NA),
  M1.2 = c(NA, 9.3, NA),
  row.names = famPed$id)

famLoc = list(locusname = "M1",
              alleles = c("8" = 0.2, "9" = 0.5, "9.3" = 0.3))
class(famLoc) = "FamiliasLocus"

Familias2ped(famPed, datamatrix, loci = famLoc, matchLoci = TRUE)

Read Familias .fam files

Description

Parses the content of a .fam file exported from Familias, and converts it into suitable ped objects. This function does not depend on the Familias R package.

Usage

readFam(
  famfile,
  useDVI = NA,
  Xchrom = FALSE,
  prefixAdded = "added_",
  fallbackModel = c("equal", "proportional"),
  simplify1 = TRUE,
  deduplicate = TRUE,
  includeParams = FALSE,
  convert = TRUE,
  verbose = TRUE
)

Arguments

Value

The output of readFam() depends on the contents of the input file, and the argument includeParams. This is FALSE by default, giving the following possible outcomes:

• If the input file only contains a database, the output is a list of information (name, alleles, frequencies, mutation model) about each locus. This list can be used as locusAttributes in e.g. pedtools::setMarkers().

• If the input file describes pedigree data, the output is a list of ped objects. If there is only one pedigree, and simplify1 = TRUE, the output is a ped object.

• If useDVI = TRUE, or useDVI = NA and the file contains DVI data, then the ⁠Reference Families⁠ section of the file is parsed and converted to ped objects. Each family generally describes multiple pedigrees, so the output gets another layer in this case.

If includeParams = TRUE, the output is a list with elements main (the main output, as described above) and params, a list with some or all of the following entries:

• version: The version of Familias

• dvi: A logical indicating if a DVI section was read

• dbName: The name of the database

• dbSize: A named numeric vector containing the DatabaseSize reported for each marker

• dropoutConsider: A named logical vector indicating for each person if dropouts should be considered

• dropoutValue: A named numeric vector containing the dropout value for each marker

• maf: A named numeric vector containing the "Minor Allele Frequency" given for each marker

• theta: The Theta/Kinship/Fst value given for the marker database

References

Egeland et al. (2000). Beyond traditional paternity and identification cases. Selecting the most probable pedigree. Forensic Sci Int 110(1): 47-59.

See Also

writeFam().

Examples

# Using example file "paternity.fam" included in the package
fam = system.file("extdata", "paternity.fam", package = "pedFamilias")

# Read and plot
peds = readFam(fam)
plotPedList(peds, hatched = typedMembers, marker = 1)

# Store parameters
x = readFam(fam, includeParams = TRUE)
x$params

stopifnot(identical(x$main, peds))

# Return elements without full conversion to pedsuite
dat = readFam(fam, convert = FALSE)
dat$datamatrix

y = Familias2ped(dat$pedigrees, dat$datamatrix, dat$loci)
stopifnot(identical(y, peds))

Export data to .fam

Description

This function produces a .fam file readable by the Familias software (Egeland et al., 2000), containing all input pedigrees, their marker data and mutation models. The option openFam = TRUE calls openFamilias() to open a fresh Familias session with the produced file loaded.

Usage

writeFam(
  ...,
  famfile = "ped.fam",
  params = NULL,
  dbOnly = FALSE,
  openFam = FALSE,
  FamiliasPath = NULL,
  verbose = TRUE
)

openFamilias(famfile = NULL, FamiliasPath = NULL, verbose = TRUE)

Arguments

Details

Parameters

The params argument controls various Familias settings. Most entries below may be omitted, in which case suitable defaults are used. Single values are recycled over markers or individuals as appropriate. The params argument is broadly compatible with the params component produced by readFam() with includeParams = TRUE.

General parameters

• version: Familias version written to the file. Default: "3.4.1".

• dvi: Logical indicating whether to write a file for the Familias DVI module. Default: FALSE.

Database parameters

• dbName: Name of the marker database. Default: "unknown".

• dbSize: Database size for each marker. Default: 1000.

• maf: Minor allele frequency for each marker. Default: 0.

• theta: Theta/kinship/Fst correction. Default: 0.

• dropoutValue: Dropout probability for each marker. Default: 0.

• dropoutConsider: Logical vector indicating which pedigree members should be subject to dropout.

• dropout: Convenience alternative to dropoutConsider and dropoutValue; either a single dropout probability or a named vector identifying individuals with dropout. For DVI files dropout has a different meaning (see below).

DVI parameters

The following probabilities pertain only to Direct/Identity matching in the DVI module:

• dropout: Dropout probability. Default: 0.

• dropin: Dropin probability. Default: 0.

• typingError: Typing error probability. Default: 0.

DVI data

For DVI files, set params$dvi = TRUE and supply a single list with the following entries:

• pm: A list of singletons; the victim samples.

• am: A named list of reference families. Each must be a connected ped object, and have exactly one missing person.

• missing: A vector naming the missing persons.

To ensure that the input data has the correct format, we recommend writing DVI files through the wrapper dvir::writeFamDVI().

Value

The file name is returned invisibly.

References

Egeland et al. (2000). Beyond traditional paternity and identification cases. Selecting the most probable pedigree. Forensic Sci Int 110(1): 47-59.

See Also

readFam().

Examples

# Create pedigree with 2 markers
x = nuclearPed() |>
  addMarker(geno = c("2/2", "1/3", "2/3"), alleles = 1:3,
            afreq = c(.3,.3,.4), name = "M1")

# Write to .fam
tmp = writeFam(x, famfile = tempfile())

# Read back in
y = readFam(tmp)

stopifnot(identical(x, y))


### With stepwise mutation model
x2 = setMutmod(x, model = "stepwise",
               rate = list(male = 0.001, female = 0.002),
               range = 0.1, rate2 = 0.0001)

# Write and read
y2 = x2 |>
  writeFam(famfile = tempfile()) |>
  readFam()

stopifnot(identical(x2, y2))


### Read/write including detailed parameters
params = list(theta = 0.1, dbName = "myDB", dropout = c("3" = 0.01))
fam = writeFam(x2, famfile = tempfile(), params = params)

dat = readFam(fam, includeParams = TRUE)

# Pedigree is now in the `main` slot
stopifnot(identical(x2, dat$main))

# The `dropout` parameter is converted to (and is equivalent to):
dat$params$dropoutConsider
dat$params$dropoutValue


### Read/write frequency database

# Write database as fam file
dbfam = writeFam(x2, famfile = tempfile(), dbOnly = TRUE)

# Read back in: The result is a list of marker attributes
a = readFam(dbfam)

# Attach to a pedigree and write to a new file
z = singleton(1) |> setMarkers(locusAttributes = a)
dbfam2 = writeFam(z, famfile = tempfile(), dbOnly = TRUE)

stopifnot(identical(readLines(dbfam), readLines(dbfam2)))


### DVI file

pm = singleton("V1") |> addMarker(geno = "1/2", name = "M")

am = nuclearPed(children = "MP") |>
  addMarker(geno = c("1/1", "1/2", NA), name = "M")

x = list(pm = list(V1 = pm), am = list(F1 = am), missing = "MP")
writeFam(x, famfile = tempfile(fileext = ".fam"), params = list(dvi = TRUE))