High energy X-rays damage matter, mainly through the photoelectric effect. A photon is absorbed by an atom and a photo electron is emitted from the inner (K) shell of the atom. The remaining hollow ion will relax through Auger decay, in which an electron from a higher level falls down and a further electron is emitted with an energy around 250 eV for atoms in the first and second rows of the periodic table. The basic atomic physics for these processes is well understood and cross-sections for photon scattering have been tabulated. Furthermore, Auger line-widths have been measured that give K-hole life times, e.g. for carbon of 11 fs. This means that it takes on average 11 fs before the Auger decay. For lighter atoms it takes longer than for heavier atoms, (mainly) due to smaller charge on the nuclues.
During the photoionization process the photoelectron may interact with electrons in the valence shells. For elements of biological significance (C, N, O, S) this may lead to so called shake-on and shake-off effects in which a further electron of low energy (10-100 eV) is emitted. Semi-empirical quantum calculations were used to estimate that such shake-off ionization occurs in 10 - 30% of the photoelectric events. These processes will be neglected here.
The advent of X-ray free electron lasers (FEL) will within a few years
enable a whole range of new experiments in physics, chemistry and
biology. In the latter category we envision performing structural
studies on large biomolecules, biomolecular aggregates or
nanocrystals. Molecular dynamics simulations of a protein molecule in
a FEL beam are encouraging as to the feasibility of such experiments.
The molecular dynamics algorithm was modified for the explosion simulations. At each time step we perform the following additional steps.
In the following assignment you will need to work without strict guidelines. Note that all GROMACS programs have a flag -h that gives helpful information about what the program does, and what the options are. Start by making a special directory for this assignment. Then download a protein structure from the protein databank. Try to find a protein with 300-500 residues, without cofactors and with a globular shape. Let's assume the protein structre is downloaded in a file called 1abc.pdb (note that the .pdb extension to the file name is obligatory). Now we have to generate a GROMACS topology using the pdb2gmx program (pdb2gmx -f 1abc). Select 0 as choice of force field. If you encounter problems at this stage due to cofactors (unknown molecules/residues) then fetch a different protein. If all is well you now have a file conf.gro containing the coordinates, and a file topol.top containing the molecular topology.
The simulations we want to do here are an extension to what was
treated by Neutze
Now copy the file em.mdp from the speptide subdirectory and modify it to have the line:
pbc = noCopy the modified file to each of the explosion directories, and create an input file for an energy minimization (grompp -v -f em -c confXXX -o em). Then run the minimization: mdrun -v -s em -c after_em.pdb. Check the resulting structure file with rasmol again. If the energy has converged sufficiently we can now use these structures for explosion simulations.
A Perl script has been created to facilitate running the explosion simulations for different pulse widths and intensities. This script you can copy from this link to your working directory. If you have the correct files (after_em.pdb, topol.top) you can just execute the script: nohup ./explode.perl >& explode.log &. Now a series of 18 simulations with six different pulse widths (2, 5, 10, 20, 50, 100 fs) and three intensities (1012, 3x1012 and 1013) will be performed, and the results are located in subdirectories in a subdirectory called SIMS.
From these simulations you will first analyze the results as printed directly in e.g. ionize.xvg. You should see something like this:
For some of the simulations (in particular the longer ones) it is instructive to watch the trajectory (traj.trr) using ngmx. Then for all simulations do the following analysis:
0.0 1.5 0.3 1.5 0.6 1.4 0.9 1.35 1.2 1.33 1.5 1.32then you make a text file containing that information, and load it in to xmgrace to view it (you can also print it). Your report should contain an analysis of all these properties as a function of water shell radius, along with your other observations and conclusions.
To make your analysis easier, a small Perl script has been created. You can copy it (by shift-clicking) here. Don't forget to make it executable after downloading (chmod +x analyse.perl). Go to the directory above all the imax* directories, and execute it. It will do all the four analyses above in each directory. Please embrace and extend the script.
As a final assignment (if there is time), you can select one of the trajectories, convert it to xtc using trjconv -f traj.trr -o traj.xtc and then visualize it using the program dino. This program can make glossy pictures, and what's more, it can make mpeg animations. Try reading the dino documentation in order to find out how to convert your simulation of choice into an mpeg animation. The nicest ones will be used on our website...