lenarother / moderna Goto Github PK

View Code? Open in Web Editor NEW

10.0 8.0 2.0 13.03 MB

A program for comparative RNA modeling.

License: GNU General Public License v3.0

Python 99.87% Shell 0.02% HTML 0.10%

moderna's Introduction

ModeRNA - A program for comparative RNA modeling

Version 1.7.1

Homepage: iimcb.genesilico.pl/moderna

Technical Support: [email protected]

Build Status

Installation Instructions

1. Quick guide

python setup.py install

2. Requirements

ModeRNA runs on any modern Windows or Linux PC. It requires the Biopython library.

3. Installation

3.1 Installing ModeRNA on Linux

To install ModeRNA on Linux, you need to:

Download the source distribution [Moderna Version 1.7.1 (source)].
Unzip the archive. A catalog with the main program moderna.py is created.
Make sure Python 2.6 or a higher version is installed. (on Ubuntu Linux, use sudo apt-get install python.
Make sure Numpy is installed (sudo apt-get install python-numpy).
Make sure BioPython is installed (sudo apt-get install python-biopython).
(tested with BioPython 1.53-1.58)
Make sure the moderna/moderna.py file is executable:
chmod a+x moderna/moderna.py
Add the path to the moderna directory to your PYTHONPATH variable,e.g.:
export PYTHONPATH=$PYTHONPATH:/home/lena/moderna/
run:
python setup.py install
After this, you can from any location write:

python

from moderna.moderna import *

3.2 Installing ModeRNA on Windows

To install ModeRNA on Windows, you need to:

Download the source distribution [Moderna Version 1.7.1 (source)].
Unzip the archive. A catalog with the main program moderna.py is created.
Make sure that these libraries are installed:
(all available from iimcb.genesilico.pl/moderna)
- Python 2.6 or a higher
- Numpy
- BioPython (tested with BioPython 1.53 - 1.58)
Run from the terminal:

C:/Python26/python.exe setup.py install
After this, you can write in the Python shell:

from moderna.moderna import *

4. Web Interface and Docker Container

There is a very simple web interface based on Flask delivered with ModeRNA:

cd server/
python server.py

Also, if you are using Docker, you can deploy ModeRNA as a local sevice:

docker pull krother/moderna
docker run -t -i -p 5000:5000 krother/moderna
cd moderna/server
python server.py

Which should run the server inside the Docker environment.

5. Legal Disclaimer

ModeRNA is released under the GPL license, a copy of which is included in the distribution (See LICENSE_GPL.TXT for details). For the files in the PDB/ directory, the Biopython License applies as well. See PDB/LICENSE_BIOPYTHON.TXT for details).

This software is provided "as-is". There are no expressed or implied warranties of any kind, including, but not limited to, the warranties of merchantability and fitness for a given application. In no event shall the authors be liable for any direct, indirect, incidental, special, exemplary or consequential damages (including, but not limited to, loss of use, data or profits, or business interruption) however caused and on any theory of liability, whether in contract, strict liability or tort (including negligence or otherwise) arising in any way out of the use of this software, even if advised of the possibility of such damage.

The authors take no responsibility for damage caused by this program or its components.

6. Contributors

Magdalena Rother - implementation
Pawel Piatkowski - implementation
Kristian Rother - architecture and unit tests
Tomasz Puton - model validation and testing
Janusz Bujnicki - concept and supervision

7. Acknowledgements

Credit goes to our lab colleagues Pawel Skiba, Piotr Byzia, Irina Tuszynska, Joanna Kasprzak, Jurek Orlowski, Pawel Lukasz, Tomasz Osinski, Marcin Domagalski, Anna Czerwoniec, Stanislaw Dunin-Horkavic, Marcin Skorupski, and Marcin Feder for their comments and constructive criticism during development.

The PDB parser ued by Moderna uses BioPython with kind support by Thomas Hamelryck. The unit test framework was brought near to us by Sandra Smit, Rob Knight, and Gavin Huttley. We also would like to thank Neocles Leontis, Fabrice Jossinet, Francois Major, and Eric Westhof who provided helpful advice on various occasions.

Special thanks go to the group of Russ Altman, who provided us with their modeling example to test ModeRNA.

8. References

Components of ModeRNA are based upon the following pieces of scientific literature:

[1] Czerwoniec A, Dunin-Horkawicz S, Purta E, Kaminska KH, Kasprzak JM, Bujnicki JM, Grosjean H, Rother K. MODOMICS: a database of RNA modification pathways. 2008 update. Nucleic Acids Res. 2008 Oct 14.

[2] Yang H, Jossinet F, Leontis N, Chen L, Westbrook J, Berman H, Westhof E. Tools for the automatic identification and classification of RNA base pairs. Nucleic Acids Res. 2003 Jul 1;31(13):3450-60.

[3] Gendron P, Lemieux S, Major F. Quantitative analysis of nucleic acid three-dimensional structures. J Mol Biol. 2001 May 18;308(5):919-36.

[4] Michalsky E, Goede A, Preissner R. Loops In Proteins (LIP)–a comprehensive loop database for homology modelling. Protein Eng. 2003 Dec;16(12):979-85.

[5] Cock PJ, Antao T, Chang JT, Chapman BA, Cox CJ, Dalke A, Friedberg I, Hamelryck T, Kauff F, Wilczynski B, de Hoon MJ. Biopython: freely available Python tools for computational molecular biology and bioinformatics. Bioinformatics. 2009 Jun 1;25(11):1422-3. Epub 2009 Mar 20.

moderna's People

Contributors

Stargazers

Watchers

Forkers

mmagnus chemikeris

moderna's Issues

Review the modify-and-copy-operations

What should we do about the methods that e.g. add modifications, and optionally copy + renumber residues. If we split the two parts (adding / copying) into two separate operations, the code would get a lot easier to maintain. Opinions?

Is the B-factor feature used?

ModeRNA annotates modified, removed etc residues in the B-factor column. This feature is essentially untested and undocumented. If it is not used, I suggest removing it.

(if it turns out later that something similar is needed, it might be easier to make a separate module that annotates the finished model based on the alignment).

Move standalone scripts to tools/

Create a toplevel directory for tools, like the fragmentDB checker or the scripts generating nucleotides in data/. Also the minimize script could go there.

Outdated email addresses and headers in code files

There are many code files with Lenas genesilico address and old version numbers in the header. Update or remove completely? Scaffold offers a mechanism to deal with version numbers.

Refactor ModeRNA alignment

The class ModeRNAAlignment does not need to decorate alignment. In my opinion it does not really belong to sequence/ (branch sequence), rather to the automatic T+A=M process. Tests for the class should be divided among itself and RNAAlignment, and a better name + location found.

get rid of files produced in tests

Add tearDown t tests that produce .pdb or .ent structures so they are removed afterwards.

include LIR fragments

Find a way to include LIR _fragment.lib and rnaDB05 without adding it to the repo.

Implement AnchoringSite

class that can perform a single superposition all by itself.

Merge secstruc library by Tomasz Puton

Tomek indicated it might be a good idea to merge his secstruc code with ModeRNA. It provides a lot of functions to work with secondary structures including I/O. The library is in very good shape and has automatic tests. See https://bitbucket.org/tomasz_puton/secstruc

Merge README files

Slow and fast tests.

Facilitate development by having a suite of fast passing tests. Switch to pytest.

Moderna_server encountered a problem -> moderna/PdbConverter.py

Problem occured in 144077733467_frag

The problem is:
Traceback (most recent call last):
File "/home/webadmin/django_dbs/moderna_server/moderna_launcher.py", line 549, in main_production
launch.run_moderna()
File "/home/webadmin/django_dbs/moderna_server/moderna_launcher.py", line 247, in run_moderna
self.convert_format()
File "/home/webadmin/django_dbs/moderna_server/moderna_launcher.py", line 542, in convert_format
conv.convert_structure()
File "/usr/local/lib/python2.6/dist-packages/moderna/PdbConverter.py", line 250, in convert_structure
converted_resi = self.convert_resi_list(resi_list)
File "/usr/local/lib/python2.6/dist-packages/moderna/PdbConverter.py", line 204, in convert_resi_list
new_resi_list.append(self.convert_resi(resi))
File "/usr/local/lib/python2.6/dist-packages/moderna/PdbConverter.py", line 192, in convert_resi
atom_list = self.get_new_at_list(resi)
File "/usr/local/lib/python2.6/dist-packages/moderna/PdbConverter.py", line 181, in get_new_at_list
else: new_at_list.append(self.change_at_name(at, at.id))
File "/usr/local/lib/python2.6/dist-packages/moderna/PdbConverter.py, line 77, in change_at_name
element = new_name.strip()[0]
IndexError: string index out of range

python viertual env don't copy the Data folder

When I installed Moderna using pip in a python virtualenv, and I tried to import the package, I got a 'File not found' error. The problem was that the Data folder was not copied to the site-packages/moderna. The solution was easy: copy/past manually the Data folder to site-packages/moderna. But anyway, it was an error that should not have happend.

Decide on interface for Renumerator

fix documentation in toplevel functions

I think the docstrings are messed up in a way Sphinx can't follow. Try:

One problem (easily solved) is that wrapping price_with_tax with currency changes its .name and .doc to that of currency, which is certainly not what we want. The functools modules contains a useful tool, wraps, which will restore these values to what we would expect them to be. It is used like so:

from functools import wraps

def currency(f):
@wraps(f)
def wrapper(_args, *_kwargs):
return '$' + str(f(_args, *_kwargs))

return wrapper

Ask users about renumbering feature.

Talk with Asia Kasprzak and Mateusz Dobrychlop:

Do they use the manual modeling mode?
Did they encounter any problems with renumbering so far?
Do they have any opinion on letters (as residue numbers) in the model?
Would it be OK for them if the numbering in the model changes after each insertion (the numbers move) while modeling manualy?
How they would expect program to behave in terms of numbering in the model.

add pip installation

add moderna to python official packages and enable pip install

add documentation

Make sure documentation compiles with Sphinx. Requires finding out in which repository the current documentation from genesilico/moderna is.

CI with Travis.

All tests pass. Travis gets green.

Add PYTHONPATH to documentation

Make sure users know how to do from moderna import * instead of from moderna.moderna import *

Fragmet library can be used.

Now ModernaFragment.lib and folder with pdb files are locally in the repository.
Reason: files are big. ModeRNA gets big.
Effects:

when somebody clone the repo ---> tests won't pass
when somebody downloads release ---> cannot insert fragments automatically, automatic modelling of alignment with gaps will result in incomplete model.

Proposed solutions:

make separate pip package FragmentLibrary.
include library in ModeRNA repo (this one).
store the files online.

Do you have any other solutions or favourite one from the above?

format readme

Format README.rst file so the automatically created documentation looks nicer - e.g. authors line.

Moderna server encountered a problem -> Bio/PDB/PDBParser.py

Problem occured in 144099280745_abc

The problem is:
Traceback (most recent call last):
File "/home/webadmin/django_dbs/moderna_server/moderna_launcher.py", line 549, in main_production
launch.run_moderna()
File "/home/webadmin/django_dbs/moderna_server/moderna_launcher.py", line 247, in run_moderna
self.convert_format()
File "/home/webadmin/django_dbs/moderna_server/moderna_launcher.py", line 540, in convert_format
st = PDBParser().get_structure('', self.template)
File "/usr/local/lib/python2.6/dist-packages/Bio/PDB/PDBParser.py", line 79, in get_structure
self._parse(file.readlines())
File "/usr/local/lib/python2.6/dist-packages/Bio/PDB/PDBParser.py", line 106, in _parse
self.trailer=self._parse_coordinates(coords_trailer)
File "/usr/local/lib/python2.6/dist-packages/Bio/PDB/PDBParser.py", line 164, in _parse_coordinates
resseq=int(line[22:26].split()[0]) # sequence identifier
ValueError: invalid literal for int() with base 10: 'A'

Design interface for automatic modeling from single template versus multi-template.

One of our goals for Pawels visit?

Decrease size of downloadable moderna package.

Some ideas in #56.
Ask users about opinion?

Data files not included in setup.py

The data files e.g. in moderna/data/ do not get copied when typing
python setup.py sdist

Improve command-line interface

Create a main, check whether the options are parsed in a good way, add color to command-line output. Try other stuff from EuroPython2014 talk "building awesome command-line applications".

setup.py test works.

Currently all tests in test_ModeRNA_all.py pass but there are fails in setup.py test. Perhaps path problem.

Move renumber_template from commands

Pawels addition is valuable - move it to a new place

Implement renumerator

Exchange ideas on the interface after extracting code from fixfrag.

Write recipes to text file, read them, model

All detailed steps for modeling in a single text file - is that a killer feature? Might at least help with debugging. Opinions?

fix_backbone via server

Add fix_backbone to the moderna server. (from AK)

Decide package structure with core developers

I think the package structure could be improved. The branches sequence, fragment_library, and isosteric point in a direction we could go further. Before taking any serious action, I'd like to talk to MR and PP at least. TP's opinion would be valuable as well.