read.big.fasta: Reading large FASTA alignments

In pievos101/PopGenome: An Efficient Swiss Army Knife for Population Genomic Analyses

Reading large FASTA alignments

Description

This function splits FASTA alignments that are too large to fit into the computer memory into chunks.

Usage

read.big.fasta(filename,populations=FALSE,outgroup=FALSE,window=2000,
               SNP.DATA=FALSE,include.unknown=FALSE,
               parallized=FALSE,FAST=FALSE,big.data=TRUE)

Arguments

filename	the basepath of the FASTA alignment
outgroup	vector of outgroup sequences
populations	list of populations
window	chunk size: number of columns/nucleotide sites
SNP.DATA	should be switched to TRUE if you use SNP data in alignment format
include.unknown	include unknown positions in the biallelic.matrix
parallized	Use parallel computations to speed up the reading - works only on UNIX systems!
FAST	Fast computation. see readData()
big.data	use the ff-package

Details

The algorithm reads the data for each individual and stores the information
on disk. The data can be analyzed as regions of the defined window size, or can
be concatenated in the PopGenome framework via the function concatenate.regions.
This function should only be used when the FASTA file does not fit into the RAM;
else, use the function readData.

Value

The function creates an object of class "GENOME"

———————————————————
The following slots will be filled in the "GENOME" object
———————————————————

	Slot	Description
1.	n.sites	total number of sites
2.	n.biallelic.sites	number of biallelic sites
3.	region.names	names of regions
4.	region.data	some detailed information about the data

Examples

# GENOME.class <- read.big.fasta("Alignment.fas", big.data=TRUE)
# GENOME.class
# [e-mail]
# CON <- concatenate.regions(GENOME.class)
# [e-mail]@biallelic.sites
# GENOME.class.slide <- sliding.window.transform(GENOME.class,100,100)
# GENOME.class <- neutrality.stats(GENOME.class,FAST=TRUE)
# show the result:
# get.sum.data(GENOME.class)
# [e-mail]

pievos101/PopGenome documentation built on Feb. 24, 2023, 7:11 a.m.