normal there isn't any ff-parameters for chitosan polymer

  • boshra
  • boshra's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
7 months 3 weeks ago #9679 by boshra
Hello every body
I want to simulate chitosan polymer with a Martini coarse-grained force field.
The monomer of chitosan polymer has a structure like glucose, but the carbon with number 2 is linked to the amine group instead of hydroxyl group. This compound does not exist in the force field parameters part of the Martini homepage. To get the force field of this polymer, can I refer to a similar compound?
What do you recommend?

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

More
7 months 3 weeks ago #9681 by MatsMD
Hi,I don't think there are any M3 parameters for this polysaccharide. If you want a coarse grained model, you would have to parametrize it yourself from an atomistic reference. You can find suggestions for bead assignments here: pubs.acs.org/doi/full/10.1021/acs.jctc.2c00757

Follow the process as explained here: cgmartini.nl/index.php/martini-3-tutoria...a-new-small-molecule . I can give additional pointers if you want me to, just ask.

Cheers,

Mats

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

  • boshra
  • boshra's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
7 months 3 weeks ago #9682 by boshra
Dear MatsMD,
Thank you very much for your reply.
First of all, I apologize for my long message.
Because the monomer of the chitosan has the structure like glucose and only in carbon number 2, it has an amine group instead of a hydroxyl group.
I assigned the beads base on pubs.acs.org/doi/10.1021/ct900313w. therefore each monomer was mapped to three beads, and I selected a tetramer for full atoms simulation, therefore I had12 beads, 11 bonds. 12 angles, and 9 dihedrals. Based on cgmartini.nl/index.php/martini-3-tutoria...a-new-small-molecule , the result of AA-simulation and above article were used for first assignment of bond lengths, angles, dihedrals and force constants respectively.I first start some iterations to determine bond lengths and force constants of 11 bonds. After 22 iterations, the distribution functions of cg and aa simulation fit well. But in the case of angles, the work became very difficult, and the distribution function an angle that was matched well in one stage was not matched in the next stage, and also, like the part of the bonds, between the force constant and the width of the distribution function of angles, there was no relationship, In such a way that the width of a distribution did not be reduced by increasing the force constants.
even after 38 iterations for angles, some bonds distribution function that previously matched were no longer logical. I am really helpless.

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

More
7 months 3 weeks ago #9683 by MatsMD
Hi,the paper you mentioned includes the carbohydrate model for Martini 2, are you interested in M2 sugars or M3 sugars (as Martini 3 is significantly different, see the paper I linked before)? Not sure what you're investigating but note that the M2 sugar model is very prone to aggregation.

A beta1,2 bond should be relatively rigid, I am sure you can get it stable in Martini3 if you follow the potential scheme suggested in the M3 carbohydrate paper. I am a bit busy but if need be, I can try sending you some quick suggestions on the mapping and choice of potentials tomorrow.

Cheers,

Mats

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

  • boshra
  • boshra's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
7 months 3 weeks ago #9684 by boshra
Dear MatsMD,
I am extremely grateful for your kindness and your time.
Best regards,
Boshra Moradi

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

  • boshra
  • boshra's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
7 months 3 weeks ago #9685 by boshra
Dear MatsMD,
H read the article you mentioned. In choice of beads, the first rule says that you must "maximizes the number of diols assigned to a single bead." In chitosan the carbon with number 2 is linked to the amine group instead of hydroxyl group. therefore, should I put (OH-CH2-CH2-NH2) group in one bead?
Best regards,
Boshra Moradi

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

More
7 months 3 weeks ago - 7 months 3 weeks ago #9686 by MatsMD
Hi Boshra, good to hear, everything you need will be in that manuscript. Anyway, I derived the mapping based on that paper for the most basic chitosan unit, a trimer: i.imgur.com/KR2SVR2.png

Basically, this resembles lactose very much: mad.ibcp.fr/molecule/LAC?version=820730973224162929 . Same glycosidic linkage, just one amine group vs alcohol. I followed the bead types and bead names of lactose, maybe take a peek in that .itp

For a trimer, you would need the 8 angles and 3 dihedrals discussed in the figure. One dihedral type crosses two monosaccharides, the other bridges three. This should keep the polysaccharide stable. A similar scheme has worked for cellulose and dextran (see: github.com/marrink-lab/polyply_1.0 ).

You can intuitively see how a tetramer, pentamer etc develop from these basic potentials. In essence you can map a pentamer/hexamer, and then use polyply or some other tool to generate arbitrary DoP-lengths from those basic potentials you mapped.

Only thing I am not certain about is the bead types for the groups with amines (light blue and pink). It depends a bit on what you are trying to study how accurate the bead assignment needs to be, without partitioning data I would keep them as the suggested bead types. The pink group is roughly comparable to an amide but less polar so I reduced one polarity level, the primary amine (light blue) is less polar than an alcohol group so I reduced it one level too.

For the mapping, you can do it manually or use something like this: github.com/fgrunewald/fast_forward .

If you have more detailed questions, please shoot me an email: mats.punt (at) helsinki.fi

Cheers,

Mats
Last edit: 7 months 3 weeks ago by MatsMD. Reason: mistake in link

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

  • boshra
  • boshra's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
7 months 3 weeks ago #9687 by boshra
Dear MatsMD,
Thank you very much, I will definitely use your advice. Thank you for your time.

Best regards,
Boshra Moradi

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

Time to create page: 0.091 seconds