onehot2seq 0.0.2

decode a one-hot numpy array to biological sequences

Description

onehot2seq is a command-line tool decoding a one-hot numpy array to DNA/RNA/protein sequences.

To encode sequences to a one-hot numpy array, use seq2onehot.
https://github.com/akikuno/seq2onehot

Installation

You can install onehot2seq using pip:

pip install onehot2seq

Usage

onehot2seq [options] -t/--type <dna/rna/protein> -i/--input <in.npy> -o/--output <out.txt/fasta>

Options

-a/--ambiguous: include ambiguous characters
-f/--format <txt/fasta>: output as a FASTA format (default: txt)

The ambigous characters are:

• XBZJ for amino acid
• NVHDBMRWSYK for DNA and RNA

The detail of ambiguous characters is described here:
https://meme-suite.org/meme/doc/alphabets.html

The header IDs of FASTA format are sequential numbers (e.g. >seq1, >seq2)

Examples

# Output DNA sequences
onehot2seq -t dna -i example/dna.npy -o dna.txt

# Output DNA sequences as a FASTA format
onehot2seq -t dna -f fasta -i example/dna.npy -o dna.fasta

# RNA sequences
onehot2seq -t rna -i example/rna.npy -o rna.txt

# Protein sequences
onehot2seq -t protein -i example/protein.npy -o protein.txt

One-hot array

The input file must contain 3d one-hot array of RxNxL (Read x Nucreotide/Amino acid x Letter)

• The order of nucreotide is ACGT (+ NVHDBMRWSYK) for DNA, ACGU (+ NVHDBMRWSYK) for RNA
• The order of amino acid is ACDEFGHIKLMNPQRSTVWY (+ XBZJ)