acvill/nodeseqs

Shell script for extracting sequences from high-degree nodes in metagenomic assembly graphs

nodeSeqs

nodeSeqs.sh is a command-line script for extracting sequences from high-degree nodes within assembly graphs, using a GFA file as input. In the language of GFA, this script looks for segments that have a large number of links. The creation of this script was motivated by my question on Bioinformatics Stack Exchange, and my approach to address this problem is modeled on an answer by Maximilian Press.

For a use case, see my blog post: Lost at the crossroads: genes at the nodes of short-read assembly graphs

Inputs

1. GFA file (-g, REQUIRED)
2. minimum number of links for a segment to be considered (-d, REQUIRED)
3. nucleotides of context to export (-c, default = 100)
4. k-mer size used during de Bruijn graph assembly (-k, default = 55)
5. minimum coverage for a segment to be considered (-m, default = 10)

For the full list of options, call the usage statement with ./nodeSeqs.sh -h

Usage: ./nodeSeqs.sh -g <file> -d <integer> [-o,-c,-k,-m,-t]
  -h    print this usage and exit
  -g    (REQUIRED) GFA file
  -d    (REQUIRED) minimum number of links (degree) needed to pull a segment (node)
  -o    output path (default: pwd)
  -c    nucleotide sequence context to extract from segment ends (default: 100)
  -k    k-mer size used in assembly (default for metaSPAdes: 55)
  -m    minimum coverage filter for segments with degree >= d (default: 10)
  -t    keep all temporary files (exclude -t to delete temp folder)

Output

This script returns a single .fasta file with headers formatted as follows:

>segment[segment ID]_degree=[number of links]_[orientation of overlap]_cov=[calculated coverage]

Here are the first few lines of the example output segments.fasta:

>segment376898_degree=8_+_cov=69.91
TGATATAATCCCCTTAGAGTAGACACATGAAATAATAAAATGGTGTAACTCTAAAGGGGGTTATTTTATGTCAAAGAAGAAATCACTTACAAGTGAAGAA
>segment1437344_degree=7_-_cov=102.50
GCCGATGTGGCTCAATTGGCAGAGCAGCTGATTTGTAATCAGCAGGTTATCGGTTCGAGTCCGATCATCGGCTT
>segment8608999_degree=6_-_cov=22.66
TTGACACACTCCCACGATTAAAATCGTGGGATTCTACTTCAACGAGGCCGCTGGCTG

Dependencies

This script should work on Unix systems with Core Utilities and GNU bc.

Example

Download nodeSeqs.sh and assembly_graph_with_scaffolds.gfa.gz and place them in the same directory. Decompress the GFA file and call nodeSeqs.sh, specifying a minimum degree of 5 links.

cd /workdir/user/nodeSeqs
wget https://github.com/acvill/nodeSeqs/raw/main/assembly_graph_with_scaffolds.gfa.gz
wget https://raw.githubusercontent.com/acvill/nodeSeqs/main/nodeSeqs.sh
gunzip assembly_graph_with_scaffolds.gfa.gz
./nodeSeqs.sh -g assembly_graph_with_scaffolds.gfa -d 5

Progress will print to the command line. Once finished, segments.fasta is written to the working directory. Use checksums to compare your output to the example output.

Details

Link lines (beginning with L) are extracted from the GFA file, and segment occurrences are counted from both the From and To fields. To ensure that the final sequences capture the overlapping (linked) portions of segments, links are processed as segment-orientation pairs. Segment IDs with at least the specified minimum degree (-d) are written to temp_XXXXX/segments.txt, and the complementary segment lines (beginning with S) are pulled. k-mer counts (KC tag) are used to compute coverage for each segment using the approximation from Gonnella & Kurtz 2016; rewritten:

where K_S is the k-mer count for a segment, L_S is the length of a segment, and k is the k-mer size used in de Bruijn graph construction (-k). For (meta)SPAdes, graphs are built iteratively using an increasing k-mer size, so k is equal to the largest k-mer size used. Sequences are extracted corresponding to segments whose coverage is at least the specified minimum coverage (-m). Sequences longer than the specified context (-c) are truncated to the context length with respect to the link orientation. Including the -t flag will automatically remove the temporary folder when the program is finished, which holds intermediate files and log.txt.