Interaction Fingerprint: Strasbourg lecture

Just read a lecture about Interaction Fingerprints (IFP) for the students of Master Chemoinformatique (University of Strasbourg) as tiny part of the course taught by Professor Alexandre Varnek.

What is IFP? Interaction fingerprint represent a binary string that encoding a presence or an absence of interactions of a ligand with amino acids of a protein binding site. It is a convenient and very easy way to compare and analyze binding poses of the ligands. IFPs are widely used to analyze binding poses, retrieve the correct one and help to improve virtual screening results and overcome the shortcomings of the existing scoring functions.


Nonsense article

Several months ago I have asked the question in the BlueObelisk QA site about what to do with nonsense article?

I need you advice: I have saw an accepted article in the journal about molecular modeling – QSAR, homology modeling, no wet-lab experiments. And I definitely can say that the article is total nonsense – the binding site is absolutely incorrect and QSAR data and interpretation treated terribly!

So, the question: what to do? write a letter to the editor?

And general question to discussion – What to do if you think that the article is incorrect not in terms of science hypothesis, but in terms of general sense? Authors have used incorrect data, or interpret results absolutely incorrectly.

Update: Wrote a letter to editor-in-chief of the journal. Will keep you with updates.

But the editor doesn't answer my anything. So, time to open the article name and explain why I think it is nonsense, but first of all I do not have any personal dislike of the authors or whatever, I am talking about science.

So, here is the article in Chemical Biology & Drug Design: A Structure-based QSAR and Docking Study on Imidazo[1,5-a][1,2,4]-triazolo[1,5-d][1,4,]benzodiazepines as Selective GABAAα5 Inverse Agonists.

Here are my points:

1. GABAA receptor is heteromeric receptor. Authors don't care about this, and have treated it as homopentameric.

2. It's well known that benzodiazepine binding site is formed by alpha and gamma subunits, and definitely binds in extracellular region and definitely not in the transmembrane domain (a lot of point mutation studies confirmed this). Authors didn't discuss any reasons for selecting the binding site at all.

3. QSAR...the correlation coefficient of predicted versus real I think will tell a lot. Q2 of 0.7 is really good...but if authors will remove all data above 20 affinity log values I think they will receive even 0.9.

Comment: I quit don't agree we Noel O'Boyle: "Just concentrate on doing your own work well. Our job as scientists is not to police the chemical literature. Weak, or wrong, papers are their own reward." Yes, we should concentrate on the science as our main job, but we are the part of the community and somehow we are responsible for the noise that society produce.


Using python3.0 under standard ubuntu python2.6.6 installation

I have default python2.6.6 installation on my machine, but I want to use the new version of the python3.0 for some scripts. The script.py is simply printing the Python version used, but with shebang version specified.

import platform

So, the results are following.
1. Starting the script from command line python test.py will give us the default version

2. Starting the script from command line with specified the version python3.1 test.py will give us the 3.1 python version.

3. If we remove the shebang line, so the script will behave absolutely the same, but ./test.py will fail, when launching with shebang line ./test.py - it will give us the correct 3.1 version. Launching without shebang line this way python3.1 test.py also will give us the 3.0 version.


PhD position in folktale classification and clustering

Reading the RSS of EURAXESS site, I found a very interesting position: PhD position in folktale classification and clustering. Sounds very funny from first point of view, but in reality is really excitement to use text mining and Natural Language Processing to analyze the myths and folktales.

Master degree in computer science, natural language processing, artificial intelligence or a related discipline (with a specialization in natural language processing)
Knowledge of, and preferably practical experience with, corpus-based methods for text analysis, machine learning, document classification and/or clustering
Good programming skills

Make a PCA of the dragon!
Train neural network based on Cinderella!
Make the ant colony based on gnomes!


Much ado about Foldit

Recent ado about FoldIT success for prediction of the structure of monomeric retroviral protease (PDB ID: 3HVP) stack me into the question: We didn't solve the structures of proteases of the most common viruses yet? I can't believe my eyes...so I begin small investigation.
1. Let's take a look at the seqeunce similarity of the solved monomeric retroviral protease - 3HVP
2. Does it have homologs? Of course! Seq.Similarity list 395 structures in the PDB with sequence similarity 95%.
3. I have randomly took the 2HS1 structure for protein superposition. RMSD of the Calpha-atoms alignment is 1.6 Angstrom. 2HS1 Structure is published in 2006 (the most old between them is 1989).
4. So the question arised - why authors can't model the protein by homology modeling, molecular dynamics optimization, etc. There is a lot of thechniques exist to model structure based on bad NMR data. So, the gamers from Foldit solved the problem better than the authors do later the same year. Seems very strange for me...


An example of ROC curves plotting with ROCR

Decided to start github with ROC curve plotting example. There is not a one ROC curve but several - according to the number of comparisons (classifications), also legend with maximal and minimal ROC AUC are added to the plot. ROC curves and ROC AUC were calculated with ROCR package.

What else should be added to the plot for ease of understanding?


Best Ligand-Receptor picture I ever saw

This is the best picture decribing the ligand-receptor interaction I ever saw. Taken from the article Molecular motions in drug design: the coming age of the metadynamics method. (DOI: 10.1007/s10822-011-9415-3).


EuroQSAR 2010

I was not attending the EuroQSAR 2010 symposium but was reading thru the book of abstract and decided to make a tag cloud with TagCrowd. So, you are wellcome, but seriously nothing really specific - it's all about compounds and models.

created at TagCrowd.com


Coot under Ubuntu 10.10

Nice and easy installation of the Coot - default software for every x-ray scientist (250 Mb space required):
#adding repository
sudo add-apt-repository ppa:mok0/ppa
# updating rep. list
sudo apt-get update
# installing
sudo apt-get install coot


Perelman million

Perelman solved the Poincaré conjecture and was recently asked by the reporter (he refused to do it for a long time):
- Why did you refused the 1 million price for the solving the Poincaré conjecture?
- I know how to rule the Universe, why I should run for the 1 million?


SpringerImages - an excellent source of data visualization

The very nice example of data visualization came from SpringerImages - 2.8 million images from Springer articles. Have very nice search across subject and type of a image (video also exists).

Add caption

An update:
It seems that SpringerImages generate new trend - CellImageLibrary a new picture-hub from the National Institute of General Medical Sciences (NIGMS) is supported by Award Number RC2GM092708. 
Purpose: This Image Library is a repository for images and movies of cells from a variety of organisms. It demonstrates cellular architecture and functions with high quality images, videos, and animations. This comprehensive and easily accessible Library is designed as a public resource first and foremost for research, and secondarily as a tool for education. The long-term goal is the construction of a library of images that will serve as primary data for research.


Mol2 bug in ChemAxon software

Interesting when the guys from ChemAxon will fix really old bug with mol2 files handling? The oldest post on the forum date back to the 2008.


Plantago for tuberculosis treatment

When I was  child my grandma use to use the Plantago leaves to protect the bleeding of scratches and to protect me from infection as she said. And I was amazed when I saw an article about antituberculosis properties of Plantago compounds, exactly diosporin derivatives. Antimicrobial activity was shown not only for Mycobacterium smegmatis and M. tuberculosis but also for other species of the pathogenic and opportunistic bacteria. So, the old treatment made a step to the real world science.

DOI: 10.1111/j.1747-0285.2010.00987.x

DOI: 10.1002/med.20235