Package {SNPkit}

SNPkit: S4 tools for reading and organizing genetic data

Description

Utilities for reading, cleaning, summarizing, and modeling SNP genotype data.

Author(s)

Vinícius Junqueira junqueiravinicius@hotmail.com Roberto Higa roberto.higa@embrapa.br Fernando Flores Cardoso fernando.cardoso@embrapa.br Marcos Jun Iti Yokoo marcos.yokoo@embrapa.br

See Also

Useful links:

• https://viniciusjunqueira.github.io/SNPkit/

• https://github.com/viniciusjunqueira/SNPkit

• Report bugs at https://github.com/viniciusjunqueira/SNPkit/issues

FImputeExport Class

Description

A class to handle export preparation for FImpute.

Slots

geno

A SnpMatrix or NULL containing genotype data.

map

A data.frame containing marker information.

path

Output file path.

name

Project or file name.

Build FImputeRunner object

Description

A convenience function to construct a 'FImputeRunner' object from a 'SNPDataLong' object.

Usage

FImputeRunner(object, path, exec_path = "FImpute3", name = "data")

Arguments

Value

An object of class 'FImputeRunner'.

FImputeRunner Class

Description

A class to manage FImpute execution and results.

Slots

export

An FImputeExport object.

par_file

Path to parameter file.

exec_path

Path to FImpute executable.

results

A data.frame containing results or summary information.

SNPFileConfig Class

Description

A class for configuring SNP file import options.

Slots

path

Path to the SNP file.

fields

A list specifying column mappings or field configurations.

codes

Character vector for genotype or allele codes.

threshold

Numeric value for filtering or quality control.

sep

Character specifying the field separator.

skip

Number of lines to skip at the top of the file.

SNPImportList Class

Description

A class for managing a list of SNP file import configurations.

Slots

configs

A list of SNPFileConfig objects.

Subset an SNPDataLong object

Description

Subsets an SNPDataLong object by rows (individuals) or columns (SNPs). You can specify which individuals or SNP markers to keep or remove.

Usage

Subset(object, index, margin = 1, keep = TRUE)

## S4 method for signature 'SNPDataLong'
Subset(object, index, margin = 1, keep = TRUE)

Arguments

Value

A new SNPDataLong object, subsetted accordingly.

Convert a genotype matrix or data.frame to snpStats::SnpMatrix

Description

This function converts a genotype matrix coded as 0/1/2/NA or AA/AB/BB to a snpStats::SnpMatrix object. It includes checks for coding validity, missing values, and duplicate sample or SNP IDs, and preserves row and column names from the input.

Usage

as_snpmatrix(
  geno,
  coding = c("012", "AAABBB"),
  missing_codes = c("NA", "-9", ".", ""),
  check_ids = TRUE
)

Arguments

Details

The function accepts both matrix and data.frame inputs. For data.frame objects, all columns are coerced to a common type using as.matrix(), which preserves rownames and colnames.

The returned SnpMatrix object stores each genotype as a single byte, which is memory-efficient compared to integer storage. However, large datasets still require substantial RAM. For very large genotype sets, consider using on-disk formats such as SNPRelate (GDS) or bigsnpr.

Value

A snpStats::SnpMatrix with the same dimnames as geno.

Examples

# Numeric 0/1/2 with NAs
set.seed(1)
geno <- matrix(sample(c(0L,1L,2L,NA), 20, replace=TRUE), nrow=5)
rownames(geno) <- paste0("ind", 1:5)
colnames(geno) <- paste0("snp", 1:4)
SM <- as_snpmatrix(geno)

# Character AA/AB/BB
geno_c <- matrix(sample(c("AA","AB","BB","."), 20, replace=TRUE,
                        prob=c(.35,.3,.3,.05)), nrow=5)
rownames(geno_c) <- rownames(geno)
colnames(geno_c) <- colnames(geno)
SMc <- as_snpmatrix(geno_c, coding="AAABBB", missing_codes=".")

Check SNP call rate

Description

Identifies SNPs with call rates below a minimum threshold.

Usage

check.call.rate(summary, min.call.rate)

Arguments

Value

Character vector with SNP names below threshold. Returns 'NULL' if none.

Examples

df <- data.frame(Call.rate = c(0.85, 0.95), row.names = c("SNP1", "SNP2"))
check.call.rate(df, 0.9)

Check Identity-By-State (IBS) for a genotype pair

Description

Checks IBS status for two genotypes.

Usage

check.ibs(gen)

Arguments

Value

Integer: 2 if identical non-heterozygotes, 0 if opposite homozygotes, -1 otherwise.

Examples

check.ibs(c(1, 1))
check.ibs(c(1, 3))

Check identical samples based on distance

Description

Identifies sample pairs considered identical based on genotype distances.

Usage

check.identical.samples(genotypes, threshold = 0)

Arguments

Value

Data frame of identical sample pairs.

Examples

mat <- matrix(sample(0:2, 20, TRUE), nrow = 5)
rownames(mat) <- paste0("S", 1:5)
check.identical.samples(mat, 0.5)

Check identical samples by block

Description

Identifies sample pairs that stay identical (within threshold) across every SNP block, scanning the markers in blocks of blcsize. Each block only re-checks the samples still in a confirmed pair, and pairs that separate in any block are dropped, so the result is the intersection of the per-block identical pairs.

Usage

check.identical.samples.by.block(genotypes, blcsize, threshold = 0)

Arguments

Value

A data.frame of identical sample pairs (columns Sample1, Sample2, Distance); Distance is taken from the first block. Empty data.frame if none.

Examples

set.seed(1)
mat <- matrix(sample(1:3, 40, TRUE), nrow = 4)
rownames(mat) <- paste0("S", 1:4)
check.identical.samples.by.block(mat, blcsize = 5, threshold = 0)

Check Mendelian inconsistencies

Description

Identifies Mendelian inconsistencies between father-child pairs.

Usage

check.mendelian.inconsistencies(genotypes, father, child)

Arguments

Value

Data frame summarizing inconsistencies per pair.

Examples

set.seed(1)
genotypes <- matrix(sample(1:3, 30, TRUE), nrow = 3,
                    dimnames = list(c("F1", "C1", "C2"), NULL))
check.mendelian.inconsistencies(genotypes,
                                father = "F1",
                                child  = c("C1", "C2"))

Check Mendelian inconsistencies for a pair

Description

Calculates number of inconsistencies and total comparable SNPs for a parent-child pair.

Usage

check.mendelian.inconsistencies.pair(g1, g2)

Arguments

Value

Numeric vector: [# inconsistencies, # comparable SNPs].

Examples

g1 <- c(1, 1, 3, 3, 2)
g2 <- c(3, 1, 1, 3, 2)
check.mendelian.inconsistencies.pair(g1, g2)

Check Sample Call Rate

Description

Identifies samples with call rate below a given threshold.

Usage

check.sample.call.rate(sample.summary, min.call.rate)

Arguments

Value

A character vector with the names of samples to remove.

Check sample heterozygosity

Description

Identifies samples with heterozygosity values deviating beyond a specified threshold.

Usage

check.sample.heterozygosity(sample.summary, max.dev)

Arguments

Value

Character vector with sample names considered outliers. Returns 'NULL' if none.

Examples

ss <- data.frame(Heterozygosity = c(0.2, 0.5, 0.7))
rownames(ss) <- c("Ind1", "Ind2", "Ind3")
check.sample.heterozygosity(ss, 1)

Check SNP by chromosome

Description

Filters SNP names belonging to specified chromosomes.

Usage

check.snp.chromo(snpmap, chromosomes)

Arguments

Value

Character vector with SNP names.

Examples

snpmap <- data.frame(Chromosome = c(1, 1, 2), Name = c("SNP1", "SNP2", "SNP3"))
check.snp.chromo(snpmap, 1)

Check SNP Hardy-Weinberg equilibrium deviation

Description

Identifies SNPs deviating from HWE beyond a z-score threshold.

Usage

check.snp.hwe(snp.summary, max.dev)

Arguments

Value

Character vector with SNP names deviating from HWE. Returns 'NULL' if none.

Examples

df <- data.frame(z.HWE = c(2, 5), row.names = c("SNP1", "SNP2"))
check.snp.hwe(df, 3)

Check SNPs for Hardy-Weinberg equilibrium deviation using chi-square p-values

Description

This function identifies SNP markers whose Hardy-Weinberg equilibrium (HWE) chi-square p-values indicate significant deviation beyond a specified threshold. It uses the p-values computed by get.hwe.chi2 on the input summary data frame.

Usage

check.snp.hwe.chi2(snp.summary, max.dev)

Arguments

Details

Any SNP with missing p-value (NA) is treated as not failing (returned as FALSE).

Value

A character vector of SNP identifiers (rownames) that fail the HWE test (p-value < max.dev). If no SNPs fail, an empty vector is returned.

See Also

get.hwe.chi2

Examples

snp.summary <- data.frame(
  Calls = c(100, 100),
  P.AA  = c(0.25, 0.7),
  P.AB  = c(0.50, 0.05),
  P.BB  = c(0.25, 0.25),
  row.names = c("SNP1", "SNP2")
)
check.snp.hwe.chi2(snp.summary, max.dev = 0.05)

Check SNP minor allele frequency

Description

Identifies SNPs with minor allele frequency below a minimum threshold.

Usage

check.snp.maf(snp.summary, min.maf)

Arguments

Value

Character vector with SNP names below threshold. Returns 'NULL' if none.

Examples

df <- data.frame(MAF = c(0.01, 0.2), row.names = c("SNP1", "SNP2"))
check.snp.maf(df, 0.05)

Check SNP missing genotype frequencies

Description

Identifies SNPs with genotype frequencies below a minimum threshold.

Usage

check.snp.mgf(snp.summary, min.mgf)

Arguments

Value

Character vector with SNP names below threshold. Returns 'NULL' if none.

Examples

df <- data.frame(P.AA = c(0.01, 0.5), P.AB = c(0.02, 0.4), P.BB = c(0.01, 0.1))
rownames(df) <- c("SNP1", "SNP2")
check.snp.mgf(df, 0.05)

Check SNP monomorphic status

Description

Identifies SNPs considered monomorphic.

Usage

check.snp.monomorf(snp.summary)

Arguments

Value

Character vector with monomorphic SNP names. Returns 'NULL' if none.

Examples

df <- data.frame(P.AA = c(1, 0.5), P.AB = c(0, 0.5), P.BB = c(0, 0))
rownames(df) <- c("SNP1", "SNP2")
check.snp.monomorf(df)

Check SNP no position

Description

Identifies SNPs without a usable genomic position, i.e. whose position is missing ('NA'), blank, non-numeric, or zero. The 'Position' column may be numeric or character ('getGeno()' reads maps as character), so it is coerced to numeric first.

Usage

check.snp.no.position(snpmap)

Arguments

Value

Character vector with SNP names without position. Returns 'NULL' if none.

Examples

df <- data.frame(Position = c(0, 100, NA), Name = c("SNP1", "SNP2", "SNP3"))
check.snp.no.position(df)  # SNP1 (zero) and SNP3 (missing)

Check SNPs with same position

Description

Identifies SNPs that share the same position on the same chromosome.

Usage

check.snp.same.position(snpmap)

Arguments

Value

List of SNP groups sharing positions.

Examples

df <- data.frame(Chromosome = c(1, 1, 2),
                 Position = c(100, 100, 200),
                 Name = c("SNP1", "SNP2", "SNP3"))
check.snp.same.position(df)

Combine multiple SNPDataLong objects

Description

This function merges a list of SNPDataLong objects, typically representing different SNP panels or datasets, into a single unified SNPDataLong object. It ensures that all genotype matrices have the same set of SNPs (filling missing SNPs with NA), and merges the marker map information while removing duplicate SNP entries.

Usage

combineSNPData(lista)

Arguments

Value

A single SNPDataLong object containing the combined genotype matrix, merged map, and a concatenated path string.

Examples

make_obj <- function(samples, snps) {
  m <- methods::new("SnpMatrix",
                    matrix(as.raw(1:3),
                           nrow = length(samples),
                           ncol = length(snps),
                           dimnames = list(samples, snps)))
  methods::new("SNPDataLong",
               geno = m,
               map  = data.frame(Name = snps,
                                 Chromosome = 1,
                                 Position = seq_along(snps)),
               path = tempfile(),
               xref_path = "chip1")
}
obj1 <- make_obj(c("S1", "S2"), c("SNP1", "SNP2"))
obj2 <- make_obj(c("S3", "S4"), c("SNP2", "SNP3"))
combined <- combineSNPData(list(obj1, obj2))

Do genome relationship matrix PCA (deprecated)

Description

Deprecated. Performs PCA using the genome relationship matrix (GRM) on a raw SnpMatrix. Use runPCA instead, which operates on a SNPDataLong object, standardises SNPs, and returns scores directly comparable to runAnticlusteringPCA.

Usage

doPCA(genotypes)

Arguments

Value

List containing 'pcs' (principal components) and 'eigen' (eigenvalues).

See Also

runPCA

Examples

set.seed(1)
mat <- matrix(sample(as.raw(1:3), 200, TRUE), nrow = 10, ncol = 20)
geno <- methods::new("SnpMatrix", mat)
rownames(geno) <- paste0("S", 1:10)
colnames(geno) <- paste0("SNP", 1:20)
res <- suppressWarnings(doPCA(geno))  # doPCA is deprecated; use runPCA()
str(res)

Exploratory plots for SNP and sample summary

Description

Generates exploratory plots: MAF histograms, HWE plots, heterozygosity scatter, MDS, and dendrogram.

Usage

exploratory.plots(
  snp.summary,
  snps.plot,
  sample.summary,
  samples.plot,
  distm,
  glabels,
  mds.plot,
  hierq.plot
)

Arguments

Value

NULL (plots are saved as JPEG files)

Examples

tmp <- tempfile(fileext = ".jpg")
snp.summary <- data.frame(
  MAF   = runif(20),
  z.HWE = rnorm(20),
  Calls = rep(100, 20),
  P.AA  = runif(20, 0, 0.5),
  P.AB  = runif(20, 0, 0.5),
  P.BB  = runif(20, 0, 0.5)
)
sample.summary <- data.frame(
  Call.rate      = runif(5, 0.9, 1),
  Heterozygosity = runif(5, 0.2, 0.4),
  row.names = paste0("S", 1:5)
)
distm <- stats::dist(matrix(rnorm(25), nrow = 5))
exploratory.plots(snp.summary,
                  snps.plot      = tempfile(fileext = ".jpg"),
                  sample.summary = sample.summary,
                  samples.plot   = tempfile(fileext = ".jpg"),
                  distm          = distm,
                  glabels        = paste0("S", 1:5),
                  mds.plot       = tempfile(fileext = ".jpg"),
                  hierq.plot     = tempfile(fileext = ".jpg"))

Convert geno slot from SNPDataLong to a data.frame

Description

Converts the genotype matrix (geno slot) of a SNPDataLong object to a data.frame, with optional centering and scaling per SNP (column).

Usage

genoToDF(object, center = FALSE, scale = FALSE)

Arguments

Value

A data.frame with individuals as rows and SNPs as columns (numeric 0/1/2, or centered/scaled values).

Examples

set.seed(1)
raw_mat <- matrix(as.raw(sample(1:3, 100, TRUE)), nrow = 10, ncol = 10)
rownames(raw_mat) <- paste0("S", 1:10)
colnames(raw_mat) <- paste0("SNP", 1:10)
geno <- methods::new("SnpMatrix", raw_mat)
obj <- methods::new("SNPDataLong",
                    geno = geno,
                    map  = data.frame(Name = colnames(geno),
                                      Chromosome = 1,
                                      Position = 1:10),
                    path = tempfile(),
                    xref_path = "chip1")
df <- genoToDF(obj, center = TRUE, scale = TRUE)
head(df[, 1:5])

Get correlation (fc method)

Description

Calculates genotype correlation using a fast check (fc) method.

Usage

get.correl.fc(g1, g2)

Arguments

Value

Numeric value of correlation.

Examples

g1 <- sample(0:2, 10, TRUE)
g2 <- sample(0:2, 10, TRUE)
get.correl.fc(g1, g2)

Get gender based on heterozygosity

Description

Infers gender using heterozygosity thresholds.

Usage

get.gender(sample.summary, threshM, threshF)

Arguments

Value

Data frame with columns 'heterozygosity' and 'sex'.

Examples

df <- data.frame(Heterozygosity = c(0.1, 0.3, 0.6))
rownames(df) <- c("A", "B", "C")
get.gender(df, 0.2, 0.5)

Get HWE chi-square p-values

Description

Calculates Hardy-Weinberg equilibrium chi-square p-values for SNPs.

Usage

get.hwe.chi2(snp.summary)

Arguments

Value

Numeric vector with p-values.

Examples

df <- data.frame(Calls = c(100, 100), P.AA = c(0.6, 0.4), P.AB = c(0.3, 0.4), P.BB = c(0.1, 0.2))
get.hwe.chi2(df)

Flexible and efficient genotype file reading using fread

Description

Imports SNP genotype data from Illumina FinalReport files using data.table::fread and builds the SnpMatrix directly from the long-format calls. This is reliable even for very large files (millions of lines, hundreds of samples), where snpStats::read.snps.long may fail to read all samples. Empty or unreadable confidence values are treated as no calls. The original file on disk is never modified.

Usage

getGeno(...)

## S4 method for signature 'ANY'
getGeno(
  path,
  fields = list(sample = 2, snp = 1, allele1 = 7, allele2 = 8, confidence = 9),
  codes = c("A", "B"),
  threshold = 0.15,
  sep = "\t",
  skip = 0,
  verbose = TRUE,
  every = NULL
)

Arguments

Value

An SNPDataLong object

IBS pair statistics

Description

Calculates IBS mean and standard deviation between two samples.

Usage

ibs.pair(g1, g2)

Arguments

Value

Numeric vector: [mean IBS, standard deviation].

Examples

g1 <- sample(0:2, 10, TRUE)
g2 <- sample(0:2, 10, TRUE)
ibs.pair(g1, g2)

Import and combine multiple genotype configurations

Description

Imports genotype data from multiple configurations defined in an SNPImportList object and combines them into a unified SNPDataLong object.

Usage

importAllGenos(object)

## S4 method for signature 'SNPImportList'
importAllGenos(object)

Arguments

Value

A combined SNPDataLong object.

Import imputed FImpute results from disk

Description

Reads existing imputed results from a given path and returns an object of class SNPDataLong.

Usage

importFImputeResults(path, method = "R")

Arguments

Value

An object of class SNPDataLong containing the imputed genotypes and SNP map.

Import multiple genotype datasets from a list of configurations

Description

Reads and imports multiple genotype datasets specified in a list of configurations. Each configuration must include the path to the genotype data and information on field mapping. Optionally, you can also specify codes, quality threshold, separator, lines to skip, and a subset of IDs to retain. The function automatically fills the 'xref_path' slot per individual and combines maps into a single data.frame, adding a 'SourcePath' column indicating their origin and removing duplicated SNP rows (by Name). Prints progress messages indicating the current path being loaded (with counter).

Usage

import_geno_list(config_list)

Arguments

Value

An object of class 'SNPDataLong' containing: - Combined genotype matrix ('geno'). - Combined map ('map') as a single data.frame with 'SourcePath' column and without duplicated rows. - Combined 'xref_path' vector (one entry per individual). - 'path' slot as a semicolon-separated string of all input dataset paths.

Convert pairs to sets

Description

Groups sample pairs into sets of related samples.

Usage

pairs2sets(pairs)

Arguments

Value

List of sets of samples.

Examples

pairs <- matrix(c("A", "B", "B", "C", "D", "E"), ncol = 2, byrow = TRUE)
pairs2sets(pairs)

Plot PCA groups from anticlustering result

Description

Plot PCA groups from anticlustering result

Usage

plotPCAgroups(pca_res, groups, pcs = c(1, 2), filename = NULL)

Arguments

Value

A ggplot object (also prints to screen).

Examples

set.seed(1)
pca_res <- stats::prcomp(matrix(rnorm(200), nrow = 20))
groups <- sample(1:2, 20, replace = TRUE)
plotPCAgroups(pca_res, groups)

Print method for SNPDataLong summary

Description

Displays the contents of a summary.SNPDataLong object on the console.

Usage

## S3 method for class 'summary.SNPDataLong'
print(x, ...)

Arguments

Value

The input x, returned invisibly.

Quality Control for SNPDataLong with optional criteria

Description

Applies flexible quality control filters on an object of class SNPDataLong. Supports call rate filtering, minor allele frequency (MAF), Hardy-Weinberg equilibrium (HWE), removal of monomorphic SNPs, exclusion of specific chromosomes, optionally removing SNPs without positions, and optionally removing SNPs at the same genomic position (keeping the one with highest MAF).

Usage

qcSNPs(x, ...)

## S4 method for signature 'SNPDataLong'
qcSNPs(
  x,
  min_snp_cr = NULL,
  min_maf = NULL,
  hwe = NULL,
  snp_position = NULL,
  no_position = NULL,
  snp_mono = FALSE,
  remove_chr = NULL,
  action = c("report", "filter", "both")
)

Arguments

Value

Depending on the action argument: - "report": list of SNPs removed by each filter and SNPs retained. - "filter": filtered SNPDataLong object. - "both": list containing the filtered object and detailed report.

Examples

set.seed(123)
raw_mat <- matrix(as.raw(sample(1:3, 100, TRUE)), nrow = 10, ncol = 10)
colnames(raw_mat) <- paste0("snp", 1:10)
rownames(raw_mat) <- paste0("ind", 1:10)
geno <- methods::new("SnpMatrix", raw_mat)
map <- data.frame(Name = colnames(geno), Chromosome = 1, Position = 1:10)
x <- methods::new("SNPDataLong",
                  geno = geno,
                  map  = map,
                  path = tempfile(),
                  xref_path = "chip1")

qcSNPs(x,
       min_snp_cr = 0.8,
       min_maf = 0.05,
       snp_mono = TRUE,
       no_position = TRUE,
       snp_position = TRUE,
       action = "filter")

Quality control on samples

Description

Applies quality control (QC) procedures to samples in a 'SNPDataLong' object, based on heterozygosity and call rate thresholds.

Usage

qcSamples(x, ...)

## S4 method for signature 'SNPDataLong'
qcSamples(
  x,
  heterozygosity = NULL,
  smp_cr = NULL,
  action = c("report", "filter", "both")
)

Arguments

Value

Depending on the 'action' argument: - '"report"': returns a list with removed and kept samples; - '"filter"': returns a new 'SNPDataLong' object with filtered genotypes; - '"both"': returns a list with: - 'filtered': the filtered 'SNPDataLong' object; - 'report': a list of removed and kept samples.

Formatted header message

Description

Prints a formatted message with a border for section titles in the console.

Usage

qc_header(title)

Arguments

Value

No return value. Used for side effects (message).

Examples

qc_header("Quality Control on Samples")

Read imputed genotypes from FImpute output and return SNPDataLong object

Description

Reads imputed genotypes and SNP information from FImpute output, builds a SnpMatrix and a corresponding map, and returns an SNPDataLong object.

Usage

read.fimpute(file, method = c("R", "Rcpp"))

Arguments

Value

An object of class SNPDataLong with three slots: geno (a SnpMatrix with individuals as rows and SNPs as columns), map (a data.frame with columns Name, Chromosome, and Position), and path (the input directory).

Examples

## Not run: 
# Requires a directory containing FImpute output files
# (genotypes_imp.txt and snp_info.txt).
snp_long <- read.fimpute("output_fimpute", method = "R")

## End(Not run)

Run PCA and anticlustering on SNPDataLong

Description

Runs PCA on a SNPDataLong object (via runPCA) and then performs anticlustering on the selected principal components.

Usage

runAnticlusteringPCA(
  object,
  K = 2,
  n_pcs = 20,
  center = TRUE,
  scale = TRUE,
  anticlust_method = c("exchange", "fast")
)

Arguments

Value

A list with components:

groups

Integer vector with anticlustering group assignments.

pca

The PCA result object (a prcomp-like list), as returned by runPCA.

pcs

Numeric matrix of the PCs used for anticlustering.

Examples

res <- runAnticlusteringPCA(nelore_imputed, K = 2, n_pcs = 0.8)
table(res$groups)

Run FImpute from a FImputeRunner object

Description

This function runs the external FImpute software using a 'FImputeRunner' object, ensuring that all required input files are present and the results are imported.

Usage

runFImpute(object, verbose = TRUE)

## S4 method for signature 'FImputeRunner'
runFImpute(object, verbose = TRUE)

Arguments

Value

An updated 'FImputeRunner' object with the 'results' slot populated (SNPDataLong).

Examples

## Not run: 
# Requires the external FImpute3 binary in PATH.
path_fimpute <- file.path(tempdir(), "fimpute_run_example")
param_file <- file.path(path_fimpute, "fimpute.par")

export_obj <- methods::new("FImputeExport",
                           geno = geno_obj@geno,
                           map  = geno_obj@map,
                           path = path_fimpute)

runner <- methods::new("FImputeRunner",
                       export    = export_obj,
                       par_file  = param_file,
                       exec_path = "FImpute3")

res <- runFImpute(runner, verbose = TRUE)

## End(Not run)

Run PCA on a SNPDataLong object

Description

Computes principal components of the (optionally centered/scaled) genotype matrix, without any clustering. For wide data (more SNPs than individuals) the PCA is obtained from the n x n Gram matrix, so the large rotation matrix is never formed; when a fixed number of PCs is requested and RSpectra is installed, only the top PCs are computed with a matrix-free solver. This is the same PCA engine used by runAnticlusteringPCA, so the scores are directly comparable.

Usage

runPCA(object, n_pcs = NULL, center = TRUE, scale = TRUE)

Arguments

Value

A list with components:

pca

A prcomp-like object: sdev, x (scores) and totvar (total column variance). rotation is NULL for the wide-data paths.

pcs

Numeric matrix with the selected top principal components.

Examples

pr <- runPCA(nelore_imputed, n_pcs = 10)
head(pr$pcs[, 1:2])

Run ADMIXTURE analysis

Description

This function runs the ADMIXTURE program on a set of PLINK files (.bed/.bim/.fam) located in a specified directory, using a given file prefix. It supports both unsupervised and supervised analyses, optional cross-validation, and custom output file prefixes to avoid overwriting results.

Usage

run_admixture(
  path,
  prefix,
  admixture_path = "admixture",
  K,
  supervised = FALSE,
  pop_assignments = NULL,
  extra_args = NULL,
  out_prefix = NULL,
  cv = NULL
)

Arguments

Details

When supervised = TRUE, a '.pop' file is automatically created in the specified directory. Each line in this file corresponds to one individual, containing the population name or "-" for missing assignments.

If out_prefix is provided, the function renames the standard ADMIXTURE output files (e.g., '<prefix>.3.Q') to use this prefix (e.g., 'myrun.Q').

The function only works on Linux or macOS systems.

Value

No value returned. Runs ADMIXTURE as a side effect. Generates output files in the specified directory. Messages indicate progress and output file names.

Examples

## Not run: 
# Requires the external ADMIXTURE binary and PLINK files prepared beforehand.
work_dir <- file.path(tempdir(), "admixture_demo")
run_admixture(
  path = work_dir,
  prefix = "plink_data",
  admixture_path = "admixture",
  K = 3,
  out_prefix = "run1_k3"
)

pop_vec <- c("A", "A", "B", "B", "-", "-", "A", "B", "A", "-")
run_admixture(
  path = work_dir,
  prefix = "plink_data",
  admixture_path = "admixture",
  K = 3,
  supervised = TRUE,
  pop_assignments = pop_vec,
  cv = 10,
  out_prefix = "supervised_k3_cv10"
)

## End(Not run)

Save genotype and map files in FImpute format

Description

S4 method to export genotype (.gen), map (.map), and parameter (fimpute.par) files compatible with [FImpute](https://www.aps.uoguelph.ca/~msargol/fimpute/).

Usage

saveFImpute(object, ...)

## S4 method for signature 'FImputeExport'
saveFImpute(object)

## S4 method for signature 'SNPDataLong'
saveFImpute(object, path)

Arguments

Value

No return value, called for side effects. The function writes the files data.gen, data.map, and fimpute.par to the directory path and returns NULL invisibly.

Export genotypes and map using basic arguments

Description

Convenience function to export FImpute files directly from a 'SnpMatrix' and map 'data.frame'.

Usage

saveFImputeRaw(geno, map, path, xref = NULL)

Arguments

Value

No return value, called for side effects. The function writes three files (data.gen, data.map, and fimpute.par) to the directory specified by path and returns NULL invisibly.

Save SNPDataLong object to PLINK format

Description

Saves genotype and map data from an SNPDataLong object in PLINK format (.ped/.map and optionally binary files).

Usage

savePlink(
  object,
  path,
  name = "plink_data",
  run_plink = TRUE,
  chunk_size = 1000,
  extra_args = NULL
)

Arguments

Value

No return value, called for side effects. Files (.ped/.map, and .bed/.bim/.fam when run_plink = TRUE) are written under path.

Examples

set.seed(1)
raw_mat <- matrix(as.raw(sample(1:3, 100, TRUE)), nrow = 10, ncol = 10)
rownames(raw_mat) <- paste0("S", 1:10)
colnames(raw_mat) <- paste0("SNP", 1:10)
geno <- methods::new("SnpMatrix", raw_mat)
obj <- methods::new("SNPDataLong",
                    geno = geno,
                    map  = data.frame(Name = colnames(geno),
                                      Chromosome = 1,
                                      Position = 1:10),
                    path = tempfile(),
                    xref_path = "chip1")
savePlink(obj, path = tempdir(), name = "demo",
          run_plink = FALSE, chunk_size = 5)

Summary for SNPDataLong objects

Description

Provides a detailed summary of an SNPDataLong object, including sample and SNP counts, proportion of missing data, and SNP distribution by chromosome if mapping information is available.

Usage

## S4 method for signature 'SNPDataLong'
summary(object, ...)

Arguments

Value

An object of class summary.SNPDataLong, which is a list with the following elements:

n_individuals

Integer. Number of individuals (rows of geno).

n_snps

Integer. Number of SNPs (columns of geno).

n_missing

Integer. Total number of missing genotype calls.

prop_missing

Numeric. Proportion of missing genotype calls.

by_chromosome

Either a table of SNP counts per chromosome (when the map provides Name and Chromosome) or NULL.

missing_by_chromosome

Either a table of SNPs with at least one missing call per chromosome, or NULL.

The object also has a dedicated print method that displays the summary on the console.