soedinglab/uniclust-pipeline

Uniclust Pipeline

Setup

Make sure to install the HH-Suite3 and MMseqs2 and adjust the paths in paths.sh. Also make sure that awk, tar, pigz, cstranslate_mpi, sed, md5deep, clustalo, kalign, timeout are all installed and available in PATH.

Usage

To build your own databases based on the uniclust pipeline you can use the following three scripts:

• run_main.sh: Run Main does the clustering, builds the uniclust30/50/90 and does the sequence enrichment of the uniboost10/20/30 databases.
• run_hhdatabase.sh: Builds the uniclust30_hhsuite database
• run_annotate.sh: Builds the annotation files

Make sure to run the scripts in this order.

LSF

The pipeline was custom build for out LSF cluster computing environment and can be submitted to the LSF with bsub < run_mpi.sh for example. Please adjust the LSF parameters at the beginning of the run_ scripts. The pipeline assumes a shared file system between the computing nodes.

Webserver & Downloads

We provide a webserver and the Uniclust based on the UniProtKB on https://uniclust.mmseqs.com/.