Example

In this example, we show how to use eFindSitePPI to predict interface residues of a given protein. Our target is 249 aa long glutathione isopentenyl phosphate kinase chain B (IPK, PDB-ID: 3ll9B), whose amino acid sequence is shown below.


3ll9B.fasta>3ll9B 249
GSHMIILKLGGSVITRKDSEEPAIDRDNLERIASEIGNASPSSLMIVHGAGSFGHPFAGEYRIGSEIENEEDLRRRRFGF
ALTQNWVKKLNSHVCDALLAEGIPAVSMQPSAFIRAHAGRISHADISLIRSYLEEGMVPVVYGDVVLDSDRRLKFSVISG
DQLINHFSLRLMPERVILGTDVDGVYTRNPKKHPDARLLDVIGSLDDLESLDGTLNTDVTGGMVGKIRELLLLAEKGVES
EIINAAVPGNIERALLGEEVRGTRITGKH

First, using the target sequence, you need to run threading or a similar procedure to collect evolutionary related templates. Here, we will use eThread, but in principle, you can use other techniques as well; see notes on template libraries and mapping files.

The crystal structure of IPK has been solved and can be obtained from PDB. Note that the original PDB file contains two chains: A and B. Since we will be focusing on chain B, you need to do some editing of the structure file to make sure that the input file for eFindSitePPI contains chain B only.

In addition, you will need  a sequence profile file. The construction of sequence profiles is discussed in the previous section.

For eFindSitePPI, three input files are required:

Lets put these files in a directory called example/. Now, export environmental variables :

[example]$ export PPI_FRTMALIGN=/usr/local/frtmalign/frtmalign

[example]$ export PPI_NACCESS=/usr/local/naccess2.1.1/naccess

[example]$ export PPI_LIBSVM=/usr/local/libsvm-3.11/svm-predict

[example]$ export PPI_LIBRARY=/usr/local/libraries/efindsiteppi-lib-oct2013

[example]$ export PPI_MODELS=/usr/local/efindsiteppi-1.0/mod/

[example]$ export PPI_ETHREADMAP=/usr/local/libraries/efindsiteppi-lib-oct2013/ethread-lib-jan2012.map

and run efindsiteppi:

[example]$ /usr/local/efindsiteppi-1.0/bin/efindsiteppi -s 3ll9B.pdb -t 3ll9B.ethread -e 3ll9B.prf -o 3ll9B-efindsiteppi

You should get 2 output files:

Images below visualize the results. The IPK molecule forms a homodimer composed of two structurally similar subunits related to each other by a 2-fold symmetry. Figure on the left presents the PPI interface predicted for a IPK monomer by eFindSitePPI; true positives, true negatives, false positives, and false negatives are colored in green, gray, red, and cyan, respectively. Moreover, as shown in figure on the right, eFindSitePPI correctly predicted the location of specific interactions, including a salt bridge between K87-chain B and D94-chain A as well as a hydrogen bond between D94-chain B and N83-chain A (click on images to zoom).

3ll9B 3ll9B
Interface residues predicted for 3ll9B Specific interactions predicted for 3ll9B

Format and content of the output files is described in the following sections.

© Michal Brylinski
This website is hosted at the CCT