RANDOMSEQ: Python command‒line random sequence generator

Abstract

Randomization is a crucial aspect of any statistical tests as experimental control or to generate null hypotheses.1‒3 As such, at the core of many simulation experiments, such as Monte Carlo simulations, is a random number or sequence generator. For example, Monte Carlo simulations were used to study mutagenesis and probability of obtaining single cells from serial dilutions.4,5 Random nucleotide and amino acid sequences had been used in many studies; thus, demonstrating the importance of random sequence generators in sequence analyses. For example, 500 thousand randomly generated DNA sequences of 50 nucleotides each were used to examine the relationship between DNA sequences and gene expressions,6 random peptide sequences had been used to study randomly arising secondary structures,7 and natural peptides had been shown to have more long‒ disordered regions than randomly generated peptide sequences.8 Several random sequence generators had been developed over the years. Many random nucleotide or amino acid generators provided minimal options; such as, fixed length and fixed GC content (http:// www.faculty.ucr.edu/~mmaduro/random.htm), random selection from a given sequence (http://www.dave‒reed.com/Nifty/randSeq. html), and random peptide generation with defined amino acid composition to output into FASTA format (https://web.expasy.org/ randseq/). MacStAn aims to generate random nucleotide sequences for a pre‒defined GC content and dinucleotide composition, allowing for maximal base repetitions and user‒defined constant regions. Being a desktop application developed for classic Mac OS, MacStAn can be considered obsolete. RANDNA is a Windows desktop application developed in Borland Delphi for the generation of fixed length random nucleotide or amino acid sequences from a given frequency. Random ORF is a web tool to generate a random nucleotide sequence with a single open reading frame. NullSeq is a command‒line random sequence generator implemented in Python, which is able to generate random nucleotide or amino acid sequences from a given frequency or source sequence. Being a command‒line tool that writes the results into a FASTA file, NullSeq can be easily incorporated into analysis pipelines/tools.9‒12