normal Nanoparticles cluster size analysis

  • Imamul
  • Imamul's Avatar Topic Author
  • Offline
  • Junior Boarder
More
5 years 8 months ago #7734 by Imamul
Nanoparticles cluster size analysis was created by Imamul
Hi all,
I got a problem while trying to use gmx clustsize. I have two coarse grained lipid monolayers separated by water slab. I placed 16 CG gold nanoparticles AuNPs) (3nm diameter, 887 beads in each) in the top of both layers and the NPs are separated about 2nm from each other. The NPs get aggregating after 300ns simulation and I run the system for 3 microsecond. I want to calculate how many NPs are making cluster together as well as many cluster are in each monolayer over time. What will be the better cutoff value? I used the following command " gmx clustsize -f pro.xtc -s pro.tpr -mcn maxclust.ndx -n index.ndx -cut 1.5 "
I selected "AuNP" for clustering.
The results don't agree with the actual number of clusters I got visualizing the system.
Thanks in advance.
Regards
Imamul

Please Log in or Create an account to join the conversation.

More
5 years 8 months ago #7735 by Pim
Replied by Pim on topic Nanoparticles cluster size analysis
More a gromacs question, but I can try to help: do you get a number that's too high or too low?

I would use a lower cut-off (0.6 or so), making sure the particles are really one bead away from each other.

If your number is too high, maybe you should use the -mol option, in the case your nanoparticles are defined as molecules at least.

Note that the maxclust.ndx will only ever contain the atomic indexes of the molecules in the largest cluster only, not in the other clusters. However, nclust.xvg should give you the correct number of clusters.

PS That's a lot of beads for a 3nm diameter particle but I guess you took that from somewhere. I guess it's a 1:1 mapping?

Please Log in or Create an account to join the conversation.

  • Imamul
  • Imamul's Avatar Topic Author
  • Offline
  • Junior Boarder
More
5 years 8 months ago #7736 by Imamul
Replied by Imamul on topic Nanoparticles cluster size analysis
Hi Pim,
Thanks for the reply.
For 1.5 nm cutoff, the number of clusters fluctuates 3 to 4 over time, I tried with different cutoff values and found that for cutoff 0.55 nm gives 12-16 clusters, cutoff value less than 0.3 nm gives extremely high value. The result should be exactly 16 clusters at the beginning of the simulations and will reduce the number of clusters after 300 ns (after 300 ns NPs start clustering).

I also tried mol option, but this option actually consider all molecules in the trajectory (mol option ignores index file). So there is no option to select which molecule I want to cluster.
Your guess is right, for mapping of 3nm AuNP I used 1:1 mapping and for lipid 4:1 mapping.
Regards
Imamul

Please Log in or Create an account to join the conversation.

More
5 years 7 months ago #7758 by Pim
Replied by Pim on topic Nanoparticles cluster size analysis
Yeah, with -mol maybe you have to first write out your trajectory with only the Au particles in there (gmx trjconv with index file) and then do the gmx cluster on this. You have to also create a new tpr file for this (gmx convert-tpr) with only the Au particles in it.

If your particles are really far enough apart at the start, it should work, but gmx cluster is not the most robust tool of the gromacs suite, so you have to fall back on other options such as MDAnalysis.

(Sorry for slowness, took a vacation from work)

Please Log in or Create an account to join the conversation.

  • Imamul
  • Imamul's Avatar Topic Author
  • Offline
  • Junior Boarder
More
5 years 7 months ago #7759 by Imamul
Replied by Imamul on topic Nanoparticles cluster size analysis
Hi Pim,
Thanks for your reply. I resolved the problem exacyly in the similar way you described in your last message.
Your advice are really helpful.
Regards
Imamul

Please Log in or Create an account to join the conversation.

Time to create page: 0.086 seconds