GENLIB: an R package for the analysis of genealogical data

Abstract

BackgroundFounder populations have an important role in the study of genetic diseases. Access to detailed genealogical records is often one of their advantages. These genealogical data provide unique information for researchers in evolutionary and population genetics, demography and genetic epidemiology. However, analyzing large genealogical datasets requires specialized methods and software. The GENLIB software was developed to study the large genealogies of the French Canadian population of Quebec, Canada. These genealogies are accessible through the BALSAC database, which contains over 3 million records covering the whole province of Quebec over four centuries. Using this resource, extended pedigrees of up to 17 generations can be constructed from a sample of present-day individuals.ResultsWe have extended and implemented GENLIB as a package in the R environment for statistical computing and graphics, thus allowing optimal flexibility for users. The GENLIB package includes basic functions to manage genealogical data allowing, for example, extraction of a part of a genealogy or selection of specific individuals. There are also many functions providing information to describe the size and complexity of genealogies as well as functions to compute standard measures such as kinship, inbreeding and genetic contribution. GENLIB also includes functions for gene-dropping simulations.The goal of this paper is to present the full functionalities of GENLIB. We used a sample of 140 individuals from the province of Quebec (Canada) to demonstrate GENLIB’s functions. Ascending genealogies for these individuals were reconstructed using BALSAC, yielding a large pedigree of 41,523 individuals. Using GENLIB’s functions, we provide a detailed description of these genealogical data in terms of completeness, genetic contribution of founders, relatedness, inbreeding and the overall complexity of the genealogical tree. We also present gene-dropping simulations based on the whole genealogy to investigate identical-by-descent sharing of alleles and chromosomal segments of different lengths and estimate probabilities of identical-by-descent sharing.ConclusionsThe R package GENLIB provides a user friendly and flexible environment to analyze extensive genealogical data, allowing an efficient and easy integration of different types of data, analytical methods and additional developments and making this tool ideal for genealogical analysis.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-015-0581-5) contains supplementary material, which is available to authorized users.

Analysis of 100 high-coverage genomes from a pedigreed captive baboon colony

“…Typically, the sex, birth date, and identity of at least one parent were known for each individual. We used GENLIB (v1.0.6) (Gauvin et al 2015) to calculate inbreeding coefficients of all individuals. GENLIB requires that all individuals must be labeled as either male or female, so in cases where the sex of an individual was missing, but the individual was identified as a mother or as a father elsewhere in the pedigree, we filled in the missing sex.…”

Deep genealogical analysis of a large cohort of participants in the CARTaGENE project (Quebec, Canada)

“…It is worth nothing that results obtained by Roy-Gagnon et al (2011) and Gauvin et al (2014Gauvin et al ( , 2015, which included genomic material along with genealogical data, showed a good agreement between the two types of data. This concordance could be relevant for researchers who plan to use data from the CARTaGENE project, since the present genealogies were constructed with the same resources available at BALSAC.…”

“…e Roy-Gagnon et al (2011). The genealogical sample used in the Roy-Gagnon study was also used by Gauvin et al (2014Gauvin et al ( , 2015. f Not available.…”

“…In both cases, the mean value of the coefficients and the proportion of participants (or pair of participants in the case of kinship) with a coefficient value larger than 0 were calculated. Statistical tests included in the GENLIB software (BALSAC, 2016;Gauvin et al, 2015) were used to compare kinship and inbreeding coefficients between the four regions. A permutation test was used to compare mean kinship coefficients between each pair of regions at each generation level, as described in Lavoie et al (2005) (p-values were obtained by performing 5000 permutations).…”

Inferring Transmission Histories of Rare Alleles in Population-Scale Genealogies

“…Both forward and backward simulations can be performed efficiently in very large genealogies. 11,12 However, neither can be easily conditioned on observed data: forward simulations (allele dropping) are unlikely to produce the observed distribution of carriers, while unbiased backward simulations (allele climbing) are unlikely to produce plausible coalescence histories for rare variants, as we show in the Material and Methods section below.…”