Introduction
The Python Modeling Interface (PMI) is a powerful set of tools designed to handle all [steps of the modeling protocol](@ref procedure) for typical modeling problems. It is designed to be used by writing a set of Python scripts.
\section rnapolii_intro Introduction
We will illustrate the use of PMI by determining the localization of two subunits of RNA Polymerase II, utilizing chemical cross-linking coupled with mass spectrometry, negative-stain electron microscopy (EM), and x-ray crystallography data. We will try to reconstruct the stalk of the complex, comprising of subunits Rpb4 and Rpb7, hypothesizing that we know already the structure of the remaining 10-subunit complex. The example can be easily generalized to any other set of subunits.
To work through the example on your own system, you will need the following packages installed in addition to [IMP itself](@ref installation):
-
numpy and scipy for matrix and linear algebra
-
scikit-learn for k-means clustering
-
matplotlib for plotting results
-
Chimera for visualization of results
(If you are using Anaconda Python,
you can get the Python packages above by simply running
conda install numpy scipy scikit-learn matplotlib.
On a Mac you can get them using the
pip tool, e.g. by running a command like
sudo easy_install pip, then install the packages with something like
sudo pip install scikit-learn; sudo pip install matplotlib. numpy and scipy are already installed on modern Macs. Something
similar may also work on a Linux box, although it's probably better to install
the packages using the distribution's package manager, such as yum or
apt-get.)
Then download the input files, either by cloning the GitHub repository or by downloading the zip file.
The rnapolii example contains three directories: analysis, data and
modeling.
\section rnapolii_background Background of RNA Polymerase II
RNA Pol II is a eukaryotic complex that catalyzes DNA transcription to synthesize mRNA strands. Eukaryotic RNA polymerase II contains 12 subunits, Rpb1 to Rpb12. The yeast RNA Pol II dissociates into a 10-subunit core and a Rpb4/Rpb7 heterodimer. Rpb4 and Rpb7 ar e conserved from yeast to humans, and form a stalk-like protrusion extending from the main body of the RNA Pol II complex.
\section rnapolii_imp Integrative Modeling using IMP
This example will use data from chemical cross linking, EM and x-ray crystallography to localize the two subunits of the RNA Polymerase II stalk (Rpb4, Rpb7) to a static core of the remaining ten subunits.
\section rnapolii_stages The four stages of Integrative Modeling
Structural modeling using IMP is divided into [four stages](@ref procedure).
Click the links below to see a breakdown of all the modeling steps.
- \subpage rnapolii_1 Collect biophysical data that can be used as structural restraints and constraints
- \subpage rnapolii_2 Define representations for the RNA Poly II structural model and define each data point as a scoring function.
- \subpage rnapolii_3 Run a sampling protocol to find good scoring conformations.
- \subpage rnapolii_4 and \subpage rnapolii_4_2 Analysis of the good scoring conformations. Clustering; uncertainty; precision; etc...
\section rnapolii_script Running the script
The first three modeling stages are all contained within one script, modeling.py. You can get started by simply changing into the rnapolii/modeling direc
tory and then running the script with Python:
\code{.sh} python modeling.py \endcode
It will take a very long time to complete the sampling; to get an idea of what's going on you can run it with only 50 output frames by adding the --test op
tion:
\code{.sh} python modeling.py --test \endcode
[On to stage 1...](@ref rnapolii_1)
*/