The swiss army knife for genome assembly.

gfastats is a single fast and exhaustive tool for summary statistics and simultaneous *fa* (fasta, fastq, gfa [.gz]) genome assembly file manipulation. gfastats also allows seamless fasta<>fastq<>gfa[.gz] conversion. It has been tested in genomes even >100Gbp.

Typical fast* metrics include:

• scaffold, contig and gap size
• number of scaffolds, contigs and gaps
• total length of scaffolds, contigs and gaps
• scaffold, contig, gap N50 and statistics (full N*/NG* statistics with the --nstar-report flag)
• area under the curve (AuN/AuNG) values for scaffolds, contigs and gaps
• average scaffold, contig, gap size
• largest scaffold, contig and gap
• base composition and GC content
• soft-masked base counts (lower case bases)

Typical gfa metrics include (see also note below when evaluating gfa):

• Number of nodes and edges
• Average degree
• Number of connected components, and length of the largets connected component
• Number of dead ends
• Number of disconnected components, and their total length
• Number of bubbles

Metrics for each scaffold/contig can be generated with the --seq-report flag.

Bed coordinates and sizes of scaffolds, contigs and gaps can be outputted with the options --out-coord and --out-size. By default, --out-coord produces a full representation of the assembly in agp format.

Additionally, input can be filtered using scaffold lists or bed coordinate files with the options --include-bed and --exclude-bed.

Importantly, the filtered input can be outputted in any *fa* (fasta, fastq, gfa [.gz]) format.

Either download one of the releases or git clone https://github.com/vgl-hub/gfastats.git --recursive and make -j in gfastats folder.

gfastats input.[fasta|fastq|gfa][.gz] [expected genome size] [header[:start-end]]

To check out all options and flags use gfastats -h.

Note: if you have a GFA without paths defined (e.g. as output from hifiasm) you will need to add the --discover-paths options in order to generate statistics for contigs and scaffolds. This is an attempt to clearly distinguish contigs from segments.

You can test some typical usage with the files in the testFiles folder, e.g.:

gfastats testFiles/random1.fasta -o gfa // converts fasta to gfa
gfastats testFiles/random2.gfa2.gfa -o fa // converts gfa to fasta

gfastats allows extensive assembly manipulation at the sequence level. Manipulation is achieved using a set of instructions provided as an ordered list in a file to the option -k / --swiss-army-knife:

gfastats testFiles/random1.fasta -k testFiles/random1.instructions.sak -o gfa // reads fasta applies a set of instructions and outputs gfa

The instructions are sequentially processed to generate the final output. Examples of instructions are:

JOIN contig1+ contig2+ 50 [gap1] [scaffold1] [this is a new scaffold] // introduces a new gap of 50 bp between scaffold1 and scaffold2 with optional id gap1, effectively joining the two sequences into a new sequences named scaffold1 with an optional comment
SPLIT contig1+ contig2+ // splits the scaffold containing contig1 and contig2, effectively removing the existing gap between them

The instructions directly provide the list of edits that were introduced. The instructions could be from an automated tool or from manual annotation.

A prime example of manipulations using input from an automated tool is overlaying AGP coordinates on top of the graph to generate new scaffolds, which can be achieved with:

gfastats input.fasta|input.gfa -a input.agp -o output.fasta|output.gfa

See the instruction wiki for a full list of instructions.

Please refer to gfastats paper for a complete description. Briefly, gfastats reads and stores any fasta<>fastq<>gfa[.gz] in gfa format. gfastats then builds a bidirected graph representation of the assembly using adjacency lists, where each node is a segment, and each edge is a gap (see figure below). The original sequence can be directly manipulated from the graph. Finally, walking the graph allows to generate different kinds of outputs, including manipulated assemblies and feature coordinates.

If you use gfastats in your work, please cite:

Gfastats: conversion, evaluation and manipulation of genome sequences using assembly graphs

Giulio Formenti, Linelle Abueg, Angelo Brajuka, Nadolina Brajuka, Cristo Gallardo, Alice Giani, Olivier Fedrigo, Erich D. Jarvis

doi: https://doi.org/10.1093/bioinformatics/btac460