Skip to main content

Posts

Showing posts from 2010

Xmas holiday: fun videos

For those who are looking for some entertainment during the Xmas holidays, here are the best videos I've seen this year:

Hilarious! "My BlackBerry is not working"


"India-Pakistan border closes"


"Soccer vs. football"


"Juggling with a bike"


"How to say Elevun.."


"Ikea doggy style"


"Duet on violin"


"Barça is champion of Spanish league"


And one of the all-time favorites:

"Arabs skating"


And that's all folks!

Reading with your tongue

Just finished an impressive book: The Mind's Eye by Oliver Sacks, in which he tells about his own loss of sight in one eye, and which kind of problems this brings about. Even more surprising, and curious is what he tells about a patient who reads with his tongue.

This man lost his reading ability, but curiously enough not his ability to write. So, he could actually write, but not read back what he had written. He could however redraw the patterns (actually the letters of words) in the air, and by the movement of his tongue, he could then decipher which letter he was writing. So in fact, he was reading with his tongue. This sounds mysterious, but indeed when he one day bit his tongue and did not have sensitivity in his tongue for a few days, he was unable to "read". Only when the tongue went back to normal did he retain his ability to "read" with his tongue.

There are also other stories, his own about the loss of 3D-vision after some 70 years, but also about a wo…

P450 Compound I observed

One of the most elusive compounds in bio(inorganic)chemistry is posed by Compound I, a highly reactive possible intermediate in the catalytic cycle of cytochrome P450 enzymes. These enzymes are involved in the metabolism of approximately 75% of known pharmaceuticals. Rittle and Green have now characterized compound I for one particular cytochrome P450 enzyme, CYP119, as reported in Science, and introduced by Stephen Sligar.

Hidden code in proteins

There are 64 combinations of DNA-base triplets that code for the 20 amino acids, and STOP codes. As such there exists redundancy in which more than one triplet codes for the same amino acid. So far it was unclear why this could be, and unknown what function it might serve.

In a recent study, it was shown that different triplets actually result in fast or slow transcription, and thus giving mRNA more flexibility. Kashina and co-workers reported in the study that it is possible to change the arginylation of β- and γ-actin by playing around with these triplets. Since slower transcription leads to exposure of a hidden lysine, it results in ubiquitination (degradation of the protein). Thus, this study reveals for the first time the hidden code in protein coding, as described also in this highlight.

Scientist Impact Factors (SIF)

Nowadays the impact of a scientist's work is as important as the work itself. And of course, there are many many different ways of measuring this impact: (i) number of citations, (ii) h-index, (iii) publish or perish, etc. Of course, for all of these self-citations are an important issue to be taken care of. There are many pro's and con's for any measure, e.g. the h-index has been suggested to be replaced by a g-index. Both the h- and g-index measure the combination of productivity and impact, but the values are calculated differently. A h-index of 20 means that the author has at least 20 papers that are cited at least 20 times. The g-index corresponds to the (unique) largest number such that the top g articles received (together) at least g2 citations.

Another much more simple measure has so far not been discussed, and which is related to the impact factor for journals. For a given year, say 2008, one looks at the total number of citations for papers in the journal from th…

Sílvia Osuna leaving for the States

Today dr. Sílvia Osuna has moved to the USA, for a two-year period as Marie Curie IOF Fellow at the group of prof. Houk (UCLA).


She will be badly missed! But luckily she will return to Girona for her third year of the Marie Curie fellowship.

Lost correspondence of Francis Crick

Francis Crick is best known for his 1953 paper in Nature together with James Watson about the structure of DNA, which they put together with data from Rosalind Franklin and Ray Gosling.
In fact, this was their second attempt after a fatally flawed proposed structure where the phosphates were on the inside. Many books have been written about this period in the UK, because there were many things happening: misappropriation of scientific data, infighting, competition between different research labs, personal antipathies. And of course, the awarding of the Nobel prize to Crick, Watson and Wilkins in 1962: this was the only objective of honest Jim Watson. (unfortunately, the one person who made it all possible, Rosalind Franklin, and without whose data the others would not have been able to propose the DNA structure, by that time had passed away and was therefore not eligible for the Nobel prize).

The best and most objective book about this period is undoubtedly the biography of Rosalind Fra…

Premi Extraordinari de Doctorat for Mireia Güell

Yesterday the tribunal has decided to award the Premi Extraordinari de Doctorat en Química 2009 to dr. Mireia Güell for her thesis entitled: "Theoretical Studies of Systems of Biochemical Interest Containing Fe and Cu Transition Metals".

For her thesis she had already received the grade excel· lent cum laude.


The official award description can be found here.

Intra- and intermolecular dispersion

As reported recently on this blog, dispersion energy corrections play an important role in computational chemistry, in particular with density functional theory.


We have applied our SSB-D functional to a number of systems where weak interactions are important. In particular, we have studied the six isomers of water-hexamer systems, dimers of noble gases, twelve isomers of C12H12 hydrocarbons, Grimme’s difficult systems (branching energy of octane, dimerization of anthracene) and the stacking of adenine dimers. For all of these systems, accurate reference data at the CCSD(T) level are available, either from the literature or calculated by ourselves. We also studied these interactions with other popular current density functionals (B3LYP, B3LYP-D, BLYP, BLYP-D, PBE, PBE-D, PBEsol, PBEsol-D, OPBE, M05-2X, B3PW91, B3PW91-D, B2PLYP, B2PLYP-D, B97-D), but found that indeed the inclusion of Grimme's dispersion correction improves the results. However, it does not correct all shortcomings …

Popularity poll density functionals 2010 (DFT2010)

We have recently hosted the IXth Girona Seminar, which had Density Functional Theory as one of the topics. One of the things that came up was the popularity of current density functionals, and how these could be used to generate a "consensus" DFT method by taking a linear combination of a number of popular DFT methods.

This "consensus" method will probably be more accurate than the GFSRPM method that was also revealed during the conference, and therefore we have setup a popularity poll for density functional methods for the year 2010.

The number of density functionals is limited by choice to 20, with 5-10 places reserved for additional suggestions; in future editions this may be extended. Please add your choice for the best (and worst) density functional at:

http://bit.ly/cQPX2w

The poll is open until Oct. 1 after which a ranking will be made.

Importance of dispersion energy for DFT methods

Last week at the IXth Girona Seminar, there were many presentations by renowned scientists (e.g. Becke, Grimme) that showed the importance of dispersion energy for the results of density functional methods.

Many different approaches are being pursued, based either on the exchange hole (Becke) or by including an empirical atompair-based correction (Grimme). We have also shown our results for the SSB-D functional, which was shown to work excellently for both intra- and intermolecular dispersion interactions.

This is now also being acknowledged more and more in other areas of research, such as bioinorganic chemistry where Siegbahn and co-workers showed the importance of van der Waals effects in transition metal complexes. It would be interesting to see the application of the SSB-D functional for these systems as well.

Spin-state-corrected GTOs

One of the fundamental basic tools in quantum chemistry is formed by the basis set, where Gaussian-type orbitals (GTOs) are fast (because of analytical integrals) but not so good, and Slater-type orbitals (STOs) are excellent but have not all analytical integrals. The latter is not a problem for density functional theory (DFT), because one needs to a numerical integration for the exchange-correlation potential anyway, so one might as well use STOs (as in done in the Amsterdam Density Functional program).

The results obtained depend critically on the quality and size of the basis set, where more is better. This is however not true for spin-state energies as shown by us recently in J. Phys. Chem. A2008, 112, 6384-6391. There we showed that STOs work fine, already with small basis sets, i.e. they converge fast. Large GTOs give the same result, but the convergence is much slower. And basis sets containing effective core potentials (ECPs) give systematically different (wrong!!) answers, so …

IXth Girona Seminar

In just a couple of weeks (July 5-8) there will be the IXth Girona Seminar, which will deal with in the 2010 version with "Electron Density, Density Matrices and Density Functional Theory". Especially the last part (DFT) will be interesting with many internationally renowned speakers (Evert Jan Baerends, Axel Becke, Stefan Grimme, Gus Scuseria, Weitao Yang to name just a few).

IQC at Science Park

Here some photos taken at the IQC a few weeks ago.
The post-doc/staff room:
And the PhD students room:

Minimum polarizability principle of spin states

In a recently published paper in Journal of Computational Methods in Sciences and Engineering, we have been investigating if the MPP (minimum polarizability principle) holds for spin states. I.e. if the most stable spin state has the lowest polarizability value. Intuitively, one might expect such a relationship to hold, since the more strongly bound are electrons, the more difficult it is to let them move (i.e. polarize); also, the more strongly bound are electrons, the more stable should be the molecule.

However, this turns out to correct only in certain cases. Several examples are presented where the low spin state is more stable, but has a higher polarizability than the high spin state; and reversely.

Therefore, the MPP principle does not hold for spin states.

Fullerenes at the Temps de Flors

In the past couple of years we have been studying different aspects of fullerenes (e.g. Chem. Eur. J.2010, 16, 3207-3214, Chem. Eur. J.2009, 15, 13111-13123, J. Am. Chem. Soc.2009, 131, 129-139, J. Am. Chem. Soc.2008, 130, 6206-6214), and today at the first day of the annual flower festival in Girona, I've seen it has not gone unnoticed:


Especially in this second one, you can see three different fullerenes we have studied: C60 (blue), D3h-C78 (pink) and C2-C78 (yellow).


Also notice the snail in the next picture:

Good performance of SSB-D for NMR

We have been testing our newly developed SSB-D functional on NMR shieldings of simple and more complex molecules, and it has withstood the scrutiny.

Similar to other properties, the new functional functions better than either PBE or OPBE. The latter was recently (J. Comp. Chem. 2007, 28, 2431-2442; Chem. Phys. Lett.2006< 421, 383-388; J. Comput. Chem.2008, 29, 1798-1807; J. Phys. Chem. A2008, 112, 6794–6799) reported to already be one of the best DFT functionals.

In a recently published paper, we show that SSB-D improves upon OPBE.

CFOUR in parallel

Lately I have managed to get CFOUR working in parallel with the following options for the configuration script:

FC=ifort CC=gcc MPIFC=mpif90 ./configure --with-blas=blashc \
--enable-mpi=openmpi --with-mpirun="mpirun -np \$CFOUR_NUM_CORES" \
--with-exenodes="mpirun -np \$CFOUR_NUM_CORES"

The only thing that remains to be done is add the $CFOUR/bin directory to your PATH:

export PATH=$PATH:$CFOUR/bin

and set the CFOUR_NUM_CORES environment variable:

export CFOUR_NUM_CORES=`cat $PBS_NODEFILE|wc -l`

And this works, especially for larger systems.

NB. In the output it should say:

PARALLEL RUN STARTED ON 8 NODES.

and

parallel ececution on node 0 of 8

On 1 node, you can see the scale-up:

1 cpu
@CHECKOUT-I, Total execution time : 920.9800 seconds.
2 cpus
@CHECKOUT-I, Total execution time : 465.4900 seconds.
4 cpus
@CHECKOUT-I, Total execution time : 241.4300 seconds.
8 cpus
@CHECKOUT-I, Total execution time : 138.0800 seconds.

Mikael Johansson starts as Juan de la Cierva

Today dr. Mikael Johansson has started working in my group within the IQC. He will be working on spin-state reactivity and transition-metal complexes.


He had obtained a prestigious Juan de la Cierva post-doc position (3 years), which is being financed by the Spanish Ministry of Science and Innovation (MICINN).

Dr. Sílvia Osuna

Today Sílvia Osuna has successfully defended her PhD-thesis entitled "Theoretical studies of the exohedral reactivity of fullerene compounds", and obtained a grade excel·lent cum laude.


Wonderful news, and more than earned! Beautiful work done on the reactivity of fullerenes and nanotubes.

SSB-D functional in NWChem

Some months ago already the SSB-D functional was added to our local version of NWChem, and works for all things (energy, gradient, Hessian). It will be available in the next general release of NWChem. The input for our new functional is quite simple:

dft
xc ssb-d
end

plus of course the rest of the dft options (grid fine, tolerances tight, direct, mult)..

This new density functional is a new all-round DFT functional based on spin states and SN2 barriers, and works well for both these two AND weak interactions.

Occupations of irreps in NWChem

Today I have added a routine to our local version of NWChem that writes out the occupations of the irreps in case of symmetry, e.g. in the case of a quartet state for FeF3:

                DFT Occupations of the irreducible representations
                --------------------------------------------------

                        irrep         alfa         beta
                     --------     --------     --------
                     a1'               7.0          7.0
                     a1"               0.0          0.0
                     a2'               1.0          1.0
                     a2"               3.0          3.0
                     e'               13.0         12.0
                     e"                4.0          2.0

Of course, it is odd to have an uneven electrons in the e' irrep, which is why it is mandatory to be able to force these occupations on input. This will be one of the next steps to take.

My smart converter

My msc program (http://www.marcelswart.eu/msc) has been further improved and can now transform the vibrational frequencies from an ADF output into a PDB file that can be read by iMol and PyMol.

Moreover, it now writes the ZMAT part of the input for CFOUR, which is very critical with spaces etc. E.g. " H" is not the same as "H ". In the era of Science 2.0, this is more like Science 0.1 and should be adapted MTSAP (more than soon as possible).

Returning to MSC, there are now two versions available for download, one for Linux (64-bit) and one for Apple (Snow Leopard).

CFOUR

I have been testing the CFOUR (http://www.cfour.de) program for the past couple of weeks, and it seems to have potential. But at the same time it should be said that it is annoyingly critical with the input. Every space counts, i.e. it matters if you place an atom name as " H" (i.e. with a space before it) or "H" .. The same happens in the definition of the Z-matrix.
This makes the program not very user-friendly.

Also the installation is somehow not very user-friendly, e.g. it took me a while to get the program to work in the standard way of

PROGRAM < input > output

where I expect only the output to get back on my home disk. After some experimentation I found out that CFOUR works differently, and had to be modified with a runxcfour script.

BUT:
apart from those hiccups (that undoubtedly will be smoothed one day), the program is very very fast. E.g. I never imagined that I could do a NMR calculation with CCSD(T) on a water hexamer.
But we just did and that on just on…