DRSASP FunPDBe Annotation Utilities

Maintainer: Stuart MacGowan (smacgowan@dundee.ac.uk)

This repository contains the utility scripts we use to add DRSASP annotations to FunPDBe. For our convenience, we bundled some of the DRSASP tools into this repository and it turns out that the common interface and output formatting we built for the FunPDBe project is a good way for our users to run these tools locally themselves.

DRSASP tools included in this repository:

• 14-3-3-Pred
• NoD
• JPred4 (API Client)

If you use this resource in your work then please cite the articles for the tool(s) you used and the DRSASP article: MacGowan et al. The Dundee Resource for Sequence Analysis and Structure Prediction Protein Science, 2019, https://doi.org/10.1002/pro.3783

How to use

Running Barton Group Predictors on a collection of Fasta sequences

Running NOD

$ python run_predictors.py nod data/funpdbe_examples_list.fasta

Running 14-3-3-Pred

$ python run_predictors.py 1433pred data/funpdbe_examples_list.fasta

Running Jpred

$ python run_predictors.py jpred data/funpdbe_examples_list.fasta

Getting Fasta sequences from a list of PDB ids

Extracting the sequences from all chains in the PDB files

$ python extract_sequences.py data/funpdbe_examples_list.txt data/funpdbe_examples_list.fasta 

Getting Fasta sequences from PDB/CHAIN or UniProt IDs

Extracting the sequence from all chains based on a PDB ID

$ python extract_sequences.py --pdb <pdb_id> 

Extracting the sequence from all chains based on a PDB ID and Chain ID

$ python extract_sequences.py --pdb <pdb_id> --chain <chain_id>

Extracting the sequence from all PDB/Chain IDs based on a UniProt ID

$ python extract_sequences.py --uniprot <pdb_id>

Scoring and Classifier Thresholds

NOD

• NOD - Artificial Neural Network (20-length sequence windows with a score cut-off >= 0.8)

14-3-3-Pred

• ANN - Artificial Neural Network (cut-off >= 0.55)
• PSSM - Position-Specific Scoring Matrix (cut-off >= 0.80)
• SVM - Support Vector Machine (cut-off >= 0.25)
• Consensus - Average of the scores provided by the three methods (cut-off >= 0.50)

Dependencies

Using Python 3.5+. Check requirements.txt for all Python dependencies.

Other Dependencies:

• NOD requires batchman from SNNS package to be installed. See http://www.compbio.dundee.ac.uk/www-nod/downloads.jsp for more details.
• 14-3-3-Pred requires PyML. See http://pyml.sourceforge.net/index.html for more on how to install it.