SCARAB recruits and bins metagenomic contigs using sequence composition, read abundance, and optional trusted genomes such as SAGs. It produces extended population-genomes (xPGs) and supports optional guided reassembly in the same environment.
Full user guide · Test · Issues and feature requests
Install a released package with Mamba:
mamba create -n scarab --strict-channel-priority -c conda-forge -c bioconda -c hallamlab scarab=1.0.0
mamba activate scarab
scarab recruit --helpSCARAB 1.0.0 is available from Anaconda and Quay. The user guide covers Docker, Apptainer, source installation and the test dataset.
The tiny recruitment-and-reassembly demo is bundled in examples/test/demo/ (about 1.7 MiB). No separate data download is needed. To obtain the example matching the release:
git clone https://github.com/hallamlab/SCARAB.git
cd SCARAB/examples/test/demo
scarab recruit -m k12.gold_assembly.fasta -l read_list.txt -s SAG -o SCARAB_out -t 4
scarab reassemble \
-i SCARAB_out/algo_defaults/Default/xpgs/ecoli-COLI-K12.3578.intersect.xPG.fasta \
-l read_list.txt -o reassembly -t 4 --memory 4
python ../../../scripts/validate_test.py --dataset . --output SCARAB_out --reassembly reassembly
For input preparation, parameters, outputs, guided reassembly, and resource settings, see the user guide.
Report issues or request features. Documentation source is maintained in docs/.