.. Yet another CUDA/GPU quantum chemistry project under QUantum BIoinformatics Tools (QUBIT) hood:
CUDA development for charge transfer in Hydrogen Bond Networks (HBN)
Computational task: CUDA/GPU quantum dynamics development Time-dependent Schrodinger equation solver for proton transfer GPU parallel implementation via CUFFT library Fast Fourier transform routines
Computational task: CUDA/GPU propagation of quantum wavepackets in Hydrogen Bond Networks Non-linear Time-dependent Schrodinger equation solver via CUFFT library Fast Fourier transform routines
The PHEMTO Server is hosted by the Institute of Organic Chemistry, Biophysical Chemistry Lab at Bulgarian Academy of Sciences
The financial support comes from a collaborative project with Technical University - Sofia (TU-Sofia): grant D-002-126
Electric Dipole Moment computation of protein molecules implemented here is based on the electrostatic interaction methods and algorithms at our PHEPS server article published earlier in Nucleic Acids Research:
However PHEMTO is not just an upgrade of our server PHEPS. Inasmuch as biomolecules are a type of complex systems for which we nevertheless can calculate many properties from fundamental principles, Electric Dipole Moments occupy an unusual, if not unique, position in protein physics science. They may constitute an immensely valuable arena for investigating both the power and conceptual status of general principles of the way modern science relates atomic structure to function. To the best of our knowledge it is the first service for the community which allows pH-dependent electric dipole moment calculation - a crucial step in dissecting physical basis of protein interaction networks. In addition we implement in silico analysis of electrostatic mutants i.e. setting ionizable residues to neutral and exploring their effect on electrostatic interactions, pH dependence of molecular properties with special emphasis on electric dipole moments. Special attention is given to the ease of the service without sacrificing rigor of computational procedures.
We have designed two workflows:
One that puts emphasis on pH dependence (tutorial mode included):
A short introduction - fast electrostatic calculations of proteins
The server is supposed to be used mainly by protein scientists who need
pH-dependent fast mutant electrostatic analysis.It is easy and straightforward to use.
The algorithms implementing electrostatics modeling and computational algebra postprocessing are written in C/C++ , Perl and Haskell functional language by one of us (Alex Kant).
C++ codes algorithms that are computationally demanding (iterative Tanford-Kirkwood-Roxby style procedure as well as Poisson-Boltzmann finite-difference equation solver):
Solving Poisson-Boltzmann equation by feeding as input converged charge distribution.
Perl excels at efficient and elegant protein structure parsing and convenient data structure manipulation.
Haskell functional programming language is proficient at Advanced Computational Linear Algebra algorithms, such as SVD (Singular Value Decomposition) employed in pseudoinverse matrix CHELPG-like procedure. The latter is reminiscent of the widely applied methods in Quantum Mechanics of small molecules. However converged wavefunction or density matrix is used to comupte the electrostatic potential grid. Finally, the combination of efficiency and expressivity is based on GSL Haskell framework.
Why SVD of electrostatic potential grid? Why Haskell?
The web implementation itself is driven by CGI/PERL routines with Java employed to run molecular viewer for interactive visualization of dipole/electric moments relative to 3D protein structure.This Java applet is part of Jmol applet molecular viewer distribution.
Input
PHEMTO server expects as an input a coordinate file in PDB format - either user supplied or just as a PDB ID, following retrieval from our local PDB database. Protein structure files, containing HETATM records are given special attention - an option is present to account for ligand/cofactors/ions charge properties explicitly in the electrostatic interaction calculation. As an additional asset the user is given relevant information about the protein molecule and warned about certain inconsistencies in protein structure, that might impact adversely ensuing calculation e.g. interruption in residue numbering, which influences electrostatics through the appearance of terminal amino positive and carboxy negative charge sites with intrinsic pKs. The user is given the possibility to edit initial setup of ionogenic groups (attention to cystein residues in disulfide bonds and excluding covalently modified groups). This is accomplished by user-friendly panel selection of ionizable groups that are going to be accounted for in the consequent self-consistent electrostatic calculation, alleviating the efforts of the user to customize input protein structure. Direct edit of PDB file allows for a range of options aimed at the advanced user: adding missing terminal charges, fixed (non-titratable) integer or partial charges and titratable groups with user defined pKa intrinsic. We consider such rich electrostatic setup a distinction of our server PHEMTO. Reasonably acquainted users could address a number of subtle of issues e.g. effects of ligands, cofactors, inhibitors and ions. All other parameters used as input are predefined or automatically calculated. These steps complete initial setup. Calculation proceeds through aforementioned stages - evaluation of accessibilities and Born term pKBorn,i , perturbation of pKa by partial charges pKpar,I and finally the iterative procedure for self-consistent evaluation of titratable pKtit,i . For benchmark purpose PHEMTO server provides an option for electrostatic potential calculation by application of numerical Poisson-Boltzmann equation solver.
Output at the Intermediate stage (Basic Electrostatics) of Electric/Dipole Moment Workflow - formal PHEPS functionality
After initial setup completion the calculation proceeds through several steps - evaluation of accessibilities and Born term, perturbation of pKa by partial charges and finally the iterative procedure for self-consistent evaluation of pK values. The obtained results are organized in three sections - native, mutant, difference files. They all have similar structure:
Global Electrostatics - net charge, electrostatic term of free energy, electrostatic potential distribution
Local Electrostatics - proton population (degree of ionization of each i-th site), pK values, electrostatic energy of interaction of individual groups with entire multipole, electrostatic potential at user defined points
For each characteristic is elaborated deeper at a carefully dedicated section- just follow the respective web link. The contents of each page is comprised of the computational result itself, related derivatives (e.g. pI, pK values, etc) as well as a short description and examples for visualization of this type of data. All output data files are in standard plain text format. Visualization is straightforward with any 2-D plotting software and molecular graphics programs (JMol).
Output at Final Stage of Electric/Dipole Moment Workflow
Depending on chosen mode of electric/dipole moment computation (first moment of converged charge distribution, Singular Value Decomposition of mean-field potential matrix, Singular Value Decomposition of Poisson-Boltzmann potential matrix) the Server provides corresponding scalar and vector quantities. The vectors of electric/dipole mutated protein structure are visualized in molecular viewer applet. Scalar values of vector amplitudes and coordinate components (Cartesian X,Y,Z) of electric/dipole moment vectors are organized in tables with explicit pH dependence.
.. Yet another CUDA quantum chemistry project - QUBIT - QUantum BIoinformatics Tools:
Computational task: CUDA quantum dynamics development for Hydrogen Bond Networks Time-dependent Schrodinger equation solver for proton transfer GPU parallel implementation via CUFFT library Fast Fourier transform routines
Classical Electrostatics as Bioinformatics Tool (under development by Борис Атанасов - Boris Atanasov - boris@orgchm.bas.bg )
1. Evaluation of amino acid Residual Electron Transfer Coupling Indexes (RETCI) of Proteins.
2.Protein Electric Moment Vectors (not their Scalars) as key determinant in Macromolecular recognition
Old Bulgarians Discussion Table (A PreHistory sense talk) - SumWords.ppt
1) Lexical distribution of Sumerian-Bulgarian Names/Toponymes around the World
2) Archeodemography from time-correlated Y-DNA/mtDNA analysis (low-resolved Bulgarian HPG data)
3) Words KAR/HAR in world-distributed names and toponymes � a label of common ancient population
More details at: 
