Fast-SG

Fast-SG is an alignment-free algorithm for ultrafast scaffolding graph construction from short or long reads.

Table of contents

• Compilation Instructions
  • Get Fast-SG code
  • Get KMC (k-mer counter)
    • Mac users
    • Linux users
  • Compile Fast-SG
  • Test Fast-SG (small test)
  • Libraries used
• Usage instructions
  • Mandatory arguments
  • Running Fast-SG (example):
    • Using a single k-mer
    • Using a range of k-mers
    • Help and additional options
  • Hybrid assembly of NA12878
• Licence
• Citation

Compilation Instructions

Get Fast-SG code

git clone https://github.com/adigenova/fastsg

Get KMC (k-mer counter)

Mac users

Obtain pre-compiled binaries from:

wget  https://github.com/refresh-bio/KMC/releases/download/v3.0.0/KMC3.mac.tar.gz 

or compile yourself following the instructions provided in:

https://github.com/refresh-bio/KMC

Linux users

Obtain pre-compiled binaries from:

wget https://github.com/refresh-bio/KMC/releases/download/v3.0.0/KMC3.linux.tar.gz

or compile yourself following the instructions provided in:

https://github.com/refresh-bio/KMC

After getting KMC3, put the binaries inside Fast-SG directory, specifically in:

$Fast-SG/KMC/bin/kmc

$Fast-SG/KMC/bin/kmc_dump

Compile Fast-SG

make all

c++ compiler; compilation was tested with g++ version 5.3 (Linux) and clang version 4.2 (Mac OSX).

Test Fast-SG (small test)

make test

Libraries used

Currently Fast-SG uses the following C++ opensource libraries:

1.- kseqcpp (https://github.com/ctSkennerton/kseqcpp.git checkout cfa50bcd17bbcb3225d431df4a2c1396f58a0993)

2.- ntHash (https://github.com/bcgsc/ntHash.git checkout ff326a8c9ccf6186f42c1f49950c1ebaadbd7f7a)

3.- BBHash (https://github.com/rizkg/BBHash.git checkout 99c905828a58fa119979df5c26bdbea93f0a7696)

4,- quasi_dictionary (https://github.com/pierrepeterlongo/quasi_dictionary.git checkout 9e8c64b150b129035f92d010a12085bd6c9490f0)

Usage instructions

FAST-SG.pl is the wrapper script used to run FastSG++.

Mandatory arguments

FAST-SG requires 4 mandatory arguments:

1.- The k-mer size (-k) restricted to the range [12-256]

2.- The set of contig sequences (-r) in FASTA format.

3.- The output prefix (-p)

4.- The read configuration file (-r) having the following format (space separated):

Short reads:

#type libID Path(fwd) Path(rev) SAM(1:single 2:paired)	

short lib1 example/ecoli.ill-sim.fwd.fq.gz example/ecoli.ill-sim.rev.fq.gz 1

Long reads:

#type libID Path(long read) Insert-sizes SAM(1:single 2:paired)	

long pac example/ECOLI-PACSEQ.subset.fasta.gz 1000,2000,3000,5000 1

long ont example/ECOLI-ONT-1D.subset.fasta.gz 1000,2000,3000,5000 1

Example of a read configuration file can be found in examples/ecoli-reads.txt

Running Fast-SG (example):

Using a single k-mer

./FAST-SG.pl  -k 15 -l example/ecoli-reads.txt -r example/ecoli-illumina.fa.gz -p test

Using a range of k-mers

./FAST-SG.pl  -k 15-25:5 -l example/ecoli-reads.txt -r example/ecoli-illumina.fa.gz -p test

Help and additional options

./FAST-SG.pl  --help

Hybrid assembly of NA12878

The wiki page provide a full example of the use of Fast-SG for the hybrid assembly of NA12878.

Licence

Fast-SG software distributed under MIT licence

Citation

Alex Di Genova, Gonzalo A Ruz, Marie-France Sagot, Alejandro Maass; Fast-SG: an alignment-free algorithm for hybrid assembly, GigaScience, Volume 7, Issue 5, 1 May 2018, giy048, link