PIPSA

Documentation of PIPSA, webPIPSA and multipipsa

PIPSA (Protein Interaction Property Similarity Analysis) was initially developed as a standalone commandline tool. For a documentation of the stand-alone tool as well as additional information about the methods used in PIPSA please visit our tool usage page or our theory page.

multipipsa

The newly added python wrapper (available as multipipsa code) is documented within the python package README and INSTALL files.

The python wrapper also allows analysis as described in
Tong R, Wade RC, Bruce NJ.
Comparative electrostatic analysis of adenylyl cyclase for isoform dependent regulation properties.
Proteins. 2016 Dec;84(12):1844-1858.
doi: 10.1002/prot.25167.

The python wrapper is also used in Jupyter Notebooks developed for the Human Brain Project. The respective notebookes can be found in the ebrains colab pages section 2.1 to 2.4 (you need to request an ebrains account).

Run commandline version in Google Colab

You can run the commandline version of PIPSA in a google colab. Please login to your google account, visit the colab PIPSA in google colab, save a copy of the notebook in your own google drive. First run the example case and then modify this to your needs.

webPIPSA

A walkthrough demonstration can be found here. A PIPSA analysis allows the comparision of proteins with respect to their electrostatic properties. Currently two different entry points exist for starting an analysis using webPIPSA (the webserver implementation of PIPSA).

Starting with a set of PDB entry codes or PDB protein structure files
Starting with a SWISSPROT accession code

webPIPSA user guide for an analysis starting with PDB structures

Prerequisites

The standard PIPSA analysis requires protein structures as input data. These structures need to be of a similar fold and may optionally be superimposed before upload.

A variant of the PIPSA workflow can be launched using a SWISSPROT accession code as input. The related SWISSPROT entry needs to be an enzyme and annotated with an EC number. Since for this workflow, a protein structural modelling is performed using MODELLER, a license key for MODELLER is required. This workflow may be used to assist kinetic parameter estimation by qPIPSA and is also integrated in SYCAMORE (SYstems biology's Computational Analysis and MOdeling Research Environment). Additional information on the qPIPSA workflow can be found in the respective user guide.

Performing a standard PIPSA analysis with webPIPSA

For the PIPSA analysis a set of proteins will be compared based on their electrostatic properties.

A minimum of two, better more protein structures in PDB format should be uploaded by entering a PDB entry code or by using the Java applet from the start page.
webPIPSA currently only processes the protein chain in the PDB structure provided. All ligands, water etc will be removed from the file provided (every line in your PDB entry marked by HETATM). Later versions will allow the upload of PQR files (PDB files with charge field). This will enable users to run webPIPSA analysis of proteins including ligands.
Java must be enabled and it is advisable to enable JavaScript in your browser. The following browsers have been tested for compatibility with the web server: Mozilla Firefox (Linux/Windows), Google Chrome. Use the applet to browse to the directory where your structures are stored. You may select individual files or a directory. In the latter case, all files ending with .pdb will be uploaded.

How do I launch an analysis?

Protein structures can be uploaded using the applet, or by entering PDB id codes from RCSB.
In case you want to restrict your analysis to a chain in your PDB entry, you need to edit your PDB entry file. Open the file in a text based editor, search for lines like:
```
...
ATOM      4  O   ASP A  47       8.935   1.466  32.973  1.00 19.12           O  
ATOM      5  CB  ASP A  47      12.150   1.716  32.915  1.00 17.85           C  
...
```
The character after the amino acid code (ASP in the case above) states the chain identifier. Remove all lines starting with "ATOM" that are not from the protein chain you want to analyse.
Provide an email address to receive notification when the analysis is complete.
After the upload of the protein structures is complete, a form showing all structures will appear. You may add further structures or delete others. If your structures are not superimposed, select a method for superpositioning your structures. Normally, the first protein structure is taken and all others are superposed to it. In the optimized version all proteins are used as starting points for the superpositioning. It will then align all structures against all others. The set of structures with the least mean RMSD is used.
To use the 'Region selection' option, the structures must already be superimposed. The analysis is restricted to the selected region. The selected region may eg. be an active site, a binding domain etc. Region selections can be done by specifying a sphere with user defined coordinates of the center and radius of the sphere. Once you have completed one domain's coordinates you may press TAB or click on a field to enter additional domains (only available if JavaScript is enabled).

Calculation and results

After you have clicked 'Start PIPSA run', the page refreshes at regular intervals and prints messages about the advancement of the analysis.
- As long as your browser is not closed, the page will continue to show the advancement of your analysis.
- After the analysis is finished, you will receive an email with some of the results, a link to the results page and a link to additional intermediate results calculated during the analysis (eg. the electrostatic grid files used for the comparison).
  These intermediate results include grid files containing the electrostatic properties of the protein in question. Please follow the link if you need assistance in visualizing the electrostatic properties of a protein
- The result page shows a colorized matrix view of the comparison of all electrostatic potential fields of all proteins. In addition a tree describing the similarity is shown. The matrix and the tree can be visualized using the Hodgkin or Carbo similarity index as well as the absolute difference of potential values.
If you encounter problems you may consult the demonstation page or send an email to: mcmsoft@h-its.org.

Used tools

This webserver implements a workflow utilizing several commandline tools (not all may be used in a single workflow). Here we acknowledge the use of the following scientific tools:

Tool	Usage in webPIPSA	Reference	Details (eg. input parameters etc).
BL2SEQ	Used within the structural alignment protocol.	Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990) Basic local alignment search tool. J Mol Biol, 215, 403-410.
WHATIF	Used for adding polar hydrogen.	Vriend, G. (1990) WHAT IF: A molecular modeling and drug design program. J. Mol. Graph., 8, 52-56.
UHBD	One option for calculating electrostatic potentials.	Madura, J.D., Briggs, J.M., Wade, R.C., Davis, M.E., Luty, B.A., Ilin, A., Antosiewicz, J., Gilson, M.K., Bagheri, B., Scott, L.R. et al. (1995) Electrostatics and diffusion of molecules in solution: Simulations with the University of Houston Brownian dynamics program. Comp. Phys. Comm., 1995.	UHBD template input file
PDB2PQR	Preparation of APBS runs.	Dolinsky TJ, Nielsen JE, McCammon JA, Baker NA. PDB2PQR: an automated pipeline for the setup, execution, and analysis of Poisson-Boltzmann electrostatics calculations. Nucleic Acids Research, 32, W665-W667 (2004).
APBS	One option for calculating electrostatic potentials.	Baker, N.A., Sept, D., Joseph, S., Holst, M.J. and McCammon, J.A. (2001) Electrostatics of nanosystems: application to microtubules and the ribosome. Proc. Natl. Acad. Sci. USA, 98, 10037-10041.	APBS template input file
TCOFFEE	Used for comparative modelling.	3DCoffee: Combining Protein Sequences and Structures within Multiple Sequence Alignments. O. O'Sullivan, K Suhre, C. Abergel, D.G. Higgins, C. Notredame. Journal of Molecular Biology,Vol 340, pp385-395,2004.
CLUSTALW	Used for comparative modelling.	Chenna, R. et al. (2003): Multiple sequence alignment with the Clustal series of programs. In: Nucleic Acid Research. Bd. 31, S. 3497-3500.
MODELLER	Used for generating protein structural models from protein sequences and structural templates.	Sali, A. and Blundell, T.L. (1993) Comparitive protein modelling by satisfaction of spatial restraints. J. Mol. Biol., 234, 779-815.
PIPSA	Comparison of electrostatic potentials.	Blomberg, N., Gabdoulline, R.R., Nilges, M. and Wade, R.C. (1999) Classification of protein sequences by homology modeling and quantitative analysis of electrostatic similarity. Proteins, 37, 379-387.
PHYLIP	Produces a tree view available in the result directory.	Felsenstein, J. 1989. PHYLIP - Phylogeny Inference Package (Version 3.2). Cladistics 5: 164-166.
R-PROJECT	Clustering of the output and drawing of heat map.	Ihaka, R. and Gentleman, R. (1996) R: A language for data analysis and graphics. Journal of Computational and Graphical Statistics, 5, 299-314.

webPIPSA Protein Interaction Property Similarity Analysis