A couple of months ago, maybe a little bit more, I got the news that the project I submitted to the National Council for Science and Technology (CONACyT) was approved! Now we only have to wait for the money to actually show up and that might take a while – a long while! Nevertheless this is always very good news and we are very excited about it because this means more money for research, specifically on the electronic molecular pathways of photosynthesis.
When I submitted the project I wrote a little post about the funding scheme which seemed, if not unfair, at least flawed, and I still believe in what I wrote. To be honest I thought it wouldn’t be funded but it turns out it was but I still think the reviewing process could be better.
There is a lot of research to do – too little time to do it.
Dr. Joaquin Barroso's Blog 
Dear Dr.Jacquin,
First of all congrats for the fund..We are expecting lots of work in the future from your side, but, don’t forget your blog student…
I want to ask about the best way to carry out an explicit/implicit solvation for the first solvation shell for a nucleolside..I got some ideas but i’m not pretty sure about the best.. Because each author of each manuscript claims that his model is the best and this way we are totally lost…
My first option is to run a classical MD with AMBER then run a QM/MM also with AMBER using something like the DFTB method for the nuclesode…I’m hesitating here because of the problem is that DFTB is not truly first principle but rather a semiempirical method…..
The second option is to run MD with AMBER and then use the new AMBER link interface…I’m hesitating here because of the charge problem…Usually AMBER modify the charges for the residues by the “Antechamber” program and this way i feel i will not be consistent as i will need to start the simulation with some charge and end with another one…
Third option is start from MD structure than rerun using ADMP at the QM/MM level…in this case, can i include something like PCM to take care of the implicit and the explicit interaction is already taken care of using the explicit water molecules..???
Could you please comment
Marawan
Thanks a lot Marawan!
I will try to be helpful for as long as I can.
About your MD concerns let me tell you this is no easy task, it depends a lot on what you are studying, what you want to achieve and what resources you have to do it. I think its best if you read AMBER’s tutorials at their website. I also recommend “Understanding Molecular Simulation” by Daan Frenkel, a great book that everyone in this business should read.
As for my opinion, I would go with option 1. Technically speaking all DFT methods are semiempirical in nature (they include parametrization coefficients). Just be careful with your cutoff values, select them in a way the RMS remains as constant as possible throughout the simulation.
Option 2 sounds good only if you fix the charges from the beginning with antechamber (view AMBER’s tutorials). Option 3 definitely is the most expensive and if you are only modeling a single nucleoside then probably you can get it done but I don’t think you’ll achieve so much more accuracy. In option 3 I don’t see the need to mix PCM with the QMMM level, there are too many borders: molecule-PCMcavity /cavity-continuum/continuum-explicit solvent. I’ve never seen it done but if you have a reference about it please share it! I’m pretty sure this calculation would blow up rather quickly.
I hope this helps, Marawan! Have a nice day!
Dear Sir,
Greetings. Congratulations for the fund you have got.
Hope most of the researcher like us will be benefited more in future.
thank you sir.
Thank you very much, Dr. Sasirekha. I wish you loads of success with your research.
Thanks for commenting and reading. Have a nice day!