Rapid Pedigree Construction from SNP Genotypes • geno2ped

Overview

geno2ped is an R package for rapid pedigree construction from SNP genotype data. It supports PLINK (bed/bim/fam) and VCF/VCF.GZ input formats and infers familial relationships from genomic similarity.

Installation

Install the development version from GitHub:

# install.packages("devtools")
devtools::install_github("ampofoissabel-quangen/geno2ped")

Dependencies

Install required Bioconductor and CRAN packages:

# For PLINK files
install.packages("bigsnpr")

# For VCF files
install.packages("BiocManager")
BiocManager::install("VariantAnnotation")

Usage

Read Genotypes from PLINK files

library(geno2ped)

data <- read_genotypes(
  bed_prefix = "path/to/mydata",        # points to mydata.bed/.bim/.fam
  sample_metadata = "path/to/meta.csv"  # optional: columns ID, Sex, BirthYear
)

# Access results
data$G    # genotype matrix (samples x SNPs), values 0/1/2
data$ids  # sample IDs
data$map  # SNP map (chr, pos, snp)
data$meta # sample metadata

Read Genotypes from VCF

data <- read_genotypes(
  vcf = "path/to/mydata.vcf.gz",
  sample_metadata = "path/to/meta.csv"
)

Sample Metadata Format

Your metadata CSV should look like this:

ID	Sex	BirthYear
Sample001	M	1985
Sample002	F	1990

Input File Requirements

Format	Required Files
PLINK	`.bed`, `.bim`, `.fam`
VCF	`.vcf` or `.vcf.gz`

Output

read_genotypes() returns a list with: - $G — numeric matrix of genotypes (0/1/2), rows = samples, cols = SNPs - $ids — character vector of sample IDs - $map — data.frame with columns chr, pos, snp - $meta — data.frame of sample metadata (or NULL if not provided)

geno2ped