SWISS-MODEL WORKSPACE TUTORIAL

   
[Tutorial Index]

 

If I have no idea about possible templates for my target, and I want to identify possible template structures.

The degree of difficulty in identifying a suitable template for a target sequence can range from "trivial" for well-characterized protein families to "impossible" for proteins with an unknown fold. The SWISS-MODEL Workspace provides access to a set of increasingly complex and computationally demanding methods to search for templates within the SWISS-MODEL Template library.

SwissModel Template Library (ExPDB)

The template structure database used by SWISS-MODEL (SMTL or ExPDB library) is derived from the Protein Data Bank (Westbrook et al.). In order to allow sequence-based template searches, each PDB entry is split into individual chains. The separated template chains are annotated with information about experimental method, resolution (if applicable), ANOLEA mean force potential (Melo et al.), Gromos96 energy (van Gunsteren et al.) and PQS (Henrick et al.) quaternary state assignment to allow for rapid retrieval of the relevant structural information during template selection. Theoretical models, structures only consisting of C alpha atoms and irregularly formatted database entries are removed. Templates sharing 100% sequence identity are grouped into a SMTL100 library using the program CD-HIT (Li et al.). Clusters of sequences having 90%, 70% and 50% sequence identity are derived from the RCSB non-redundant PDB lists.

[Tools] [ SwissModel Template Library ]

You may query if a certain PDB entry is part of SMTL. In this example, we search for chains of PDB entry "1HIV". SMTL provides information about the experimental methods used for structure determination, resolution (if applicable), and links to the original PDB entry as well as protein structure classification by SCOP and CATH.

Caveat: A significant part of proteins are multimeric in their biologically active state. Single chains, or raw PDB entries often do not represent the biologically correct assembly. The PQS Protein Quaternary Structure Server (Henrick et al.) allows for searching of the list of likely quaternary structures generated at the EBI. As in our example, HIV-I protease is known to be active as a dimer. Multimeric proteins can be modelled in SWISS-MODEL Workspace using the Project Mode.

 

The target sequence can be used to query the SMTL for suitable template structures using "Template identification" in the Tools menu:

[Tools] [ Template Identification]

A condensed graphical view of the modeling task is provided containing the target sequence, the template matches sorted and colored according to the associated E-value. Clickable bars indicate the matched regions and guide the user to the underlying original program output.

[Display Alignment in DeepView]

Target-template alignments from the search tools (BLAST or SAM) can be visualized in DeepView to correct misplaced insertions and deletions in the structural context of the template, and to manually adjust misaligned regions. The modified project can then be saved to disk and submitted as "project mode" to the workspace for model building by the SWISS-MODEL pipeline.


http://swissmodel.expasy.org/workspace/   [ Contact ]