Skip to main content

Scoring protein-protein interface using RWGK and SVM

Project description


Support Vector Machine on Graph Kernel for Ranking Protein-Protein Docking Models

Build Status Documentation Status Coverage Status Codacy Badge DOI

alt text

1. Installation

Minimal information to install the module

  • Check if command mpiexec is available or not in your console. If not, download and install openmpi or mpich.
  • Install iScore using pip install iScore

Possible problems:

  • If pip install iScore gives problems on installing mpi4py, try to first install mpi4py using conda install mpi4py and then pip install iScore.

2. Documentaion

The documentaion of the pacakge can be found at:

3. Quick Examples

iScore offers simple solutions to classify protein-protein interfaces using a support vector machine approach on graph kernels. The simplest way to use iScore is through dedicated binaries that hide the complexity of the approach and allows access to the code with simple command line interfaces. The two binaries are iscore.train and iscore.predict (iscore.train.mpi and iscore.predict.mpi for parallel running) that respectively train a model using a trainging set and use this model to rank the docking models of a protein-protein complex.

Requirements for preparing data:

  • Use the following file structure

    |    |__ pdb/
    |    |__ pssm/
    |    |__ class.lst
          |__ class.lst (optional)

    The pdb folder contains the PDB files of docking models, and pssm contains the PSSM files. The class.lst is a list of class ID and PDB file name for each docking model, like 0 7CEI_10w.

         Check the package subfolders `example/train` and `example/test` to see how to prepare your files.
  • PDB files and PSSM files must have consistent sequences. PSSMGen can be used to get consistent PSSM and PDB files. It is already installed along with iScore. Check README to see how to use it.

Example 1. Use our trained model

You can directly use our trained model to score your docking conformations. The model we provide is trained on docking benchmark version 4 (BM4) data, in total 234 different structures were used (117 positive and 117 negative). More details see this paper. You can find the model in the package subfolder model/training_set.tar.gz.

To use this model go into your test subfolder and type:

# Without MPI

# With MPI
mpiexec -n ${NPROC} iScore.predict.mpi

The code will automatcially detect the path of the model.

This binary will output the binary class and decision value of the conformations in the test set in a text file iScorePredict.txt.

  For the predicted iScore values, the lower value, the better quality of the conformation.

Example 2. Train your own model

To train the model simply go to your train subfolder and type:

# Without MPI

# With MPI
mpiexec -n ${NPROC} iScore.train.mpi

This binary will generate a archive file called by default training_set.tar.gz that contains all the information needed to predict binary classes of a test set using the trained model.

To use this model go into your test subfolder and type:

# Without MPI
iScore.predict --archive ../train/training_set.tar.gz

# With MPI
mpiexec -n ${NPROC} iScore.predict.mpi --archive ../train/training_set.tar.gz

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for iScore, version 0.3.4
Filename, size File type Python version Upload date Hashes
Filename, size iScore-0.3.4-py3-none-any.whl (628.1 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size iScore-0.3.4.tar.gz (620.1 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page