anvi-script-gen-reads

Generate synthetic sequencing reads (Illumina, PacBio HiFi, ONT) from reference FASTA files with optional SNV injection at controlled multi-allele frequencies.

🔙 To the main page of anvi’o programs and artifacts.

Authors

Florian Trigodet

EmailTwitterGithub

Can consume

fasta

Can provide

paired-end-fastq single-end-fastq

Usage

This program generates synthetic sequencing reads from a reference fasta file. It supports Illumina paired-end and single-end short reads, PacBio HiFi, and Oxford Nanopore long reads, with optional SNV injection at controlled multi-allele frequencies.

The key feature is multi-allele SNV injection: you can specify exactly which positions should be variable and what the base frequencies should be at each position. This is critical for benchmarking tools that rely on SNV patterns where real biological variability involves 3-4 different bases at a position rather than just 2.

Quick start with presets

The easiest way to use this program is with a preset. Presets set sensible defaults for read length, insert size, error rate, and quality scores:

This generates OUTPUT_PREFIX-R1.fastq and OUTPUT_PREFIX-R2.fastq with 150 bp paired-end reads at 50X coverage.

Available presets:

Short reads (Illumina):

Long reads (PacBio):

Long reads (Oxford Nanopore):

ONT presets use a lognormal length distribution, which produces the right-skewed shape typical of nanopore data (mode lower than mean, with a long tail of very long reads). PacBio presets use a normal distribution, which better reflects the tighter length control of SMRT sequencing.

You can override any preset parameter individually. For example, to use the Illumina paired-end preset but with 100X coverage and 250 bp reads:

SNV injection with a mutations file

To inject SNVs at specific positions with controlled allele frequencies, provide a TAB-delimited mutations file:

The mutations file must have the following columns:

contig_name     position    freq_A    freq_T    freq_C    freq_G
contig_1        1000        0.25      0.25      0.25      0.25
contig_1        2000        0.0       0.4       0.3       0.3
contig_2        500         0.5       0.5       0.0       0.0

Positions are 0-indexed and frequencies must sum to 1.0 for each row. Each frequency represents the probability that a read covering that position will carry that base. For example, a position with freq_A=0.25, freq_T=0.25, freq_C=0.25, freq_G=0.25 will have all four bases represented equally across reads – the kind of multi-allele variability you see in real DGR variable regions.

Random SNV injection

If you don’t need precise control over SNV positions, you can have the program randomly place SNVs at a given density:

This places SNVs at approximately 1% of positions (1 per 100 bp), each with 3 different alleles at random frequencies. --num-alleles can be 2, 3, or 4.

Custom read type (no preset)

You can specify all parameters manually instead of using a preset:

Available read types are paired-end, single-end, long-fixed, and long-distributed.

Reproducibility

All runs are deterministic by default (seed = 42). To get a different random realization, change the seed:

Typical workflow

Generate reads, then run the standard anvi’o pipeline:

Edit this file to update this information.

Additional Resources

Are you aware of resources that may help users better understand the utility of this program? Please feel free to edit this file on GitHub. If you are not sure how to do that, find the __resources__ tag in this file to see an example.