fastasc

Remove completely/partially invariant sites from nucleotide aligned-FASTA for Lewis's ascertainment bias correction ( +ASC ).

Brief description

This is a simple Crystal-program and written under Crystal 1.17.1.
This program takes aligned-FASTA as input and remove invariant sites for phylogenetic inference with Lewis's ASC methods.
IUPAC nucleotide code in Upper/lower case and '?' can be accepted.
Gap ('.' and '-'), 'N', 'n', and '?' are treated as unknown (all 4 nucleotides are equally likely).

• Acceptable characters

AaTtGgCcRrYySsWwKkMmBbDdHhVvNn-.?

• Example input

cat ./test_data/test.fasta | awk 'NR%2==0{print $0}'
##ATGCatgccA
##ATGCatgccA
##ATGCatgcc.
##AAAYwnacy?
##ATR-wygyyT

• Example output

fastasc -f ./test_data/test.fasta | awk 'NR%2==0{print $0}'
##TGgA
##TGgA
##TGg.
##AAa?
##TRgT

Treating Whole-Genome SNP data can consume RAM.
On my PC, processing an aligned-FASTA with 28 sequences with 37 Mb length took about 5.7 GB of RAM and 50 seconds in real time.

Installation

Go to fastasc directory and type:

mkdir ./bin && crystal build --release -o ./bin/ ./src/fastasc.cr

This will compile the source code to an executable binary file (./bin/fastasc).
Crystal language can be easily installed by following Crystal Docs.

Usage

Usage: fastasc -f INPUT.fa > OUTPUT.fa
        or
       cat INPUT.fa | fastasc -f - > OUTPUT.fa

Remove completely/partially invariant sites from nucleotide aligned fasta for +ASC subst model
    -v, --version                    Show version
    -h, --help                       Show this help message
    -f fasta, --fasta=fasta          Path to input aligned fasta. Set '-' for STDIN.

Input format

fastasc takes aligned-FASTA as input.
Outputs of mafft, vcf2phylip.py, and vcf2alnfasta can be accepted.