search

NMRLipids / NMRlipidsIVPEandPG

NMRlipids IV project, PE and PG lipids
GNU General Public License v2.0
0 stars 7 forks source link

Area per molecules of PC:PE mixtures #7

Closed ohsOllila closed 3 years ago

ohsOllila commented 4 years ago

Patrick Fuchs wrote (http://nmrlipids.blogspot.com/2017/03/nmrlipids-iv-headgroup-glycerol.html?showComment=1524839635875#c4712801836450464031):

A few comments about my posts of today. I will focus on POPC/POPE 50:50 as suggested by Samuli. The values between the two force fields are very different, so we made an analysis of the area per lipid to try to understand what is going on:

Area results in nm^2, the error is <= 0.003 nm^2

  • pure POPC CHARMM36: 0.624 Berger : 0.649
  • POPC/POPE 50:50 CHARMM36 : POPC 0.609, POPE 0.557 Berger-hacked: POPC 0.637, POPE 0.632

One can see that CHARMM 36 predicts a drop in the area on going from pure POPC to POPC/POPE 50:50. This means that POPC pack tightly to POPE. In contrast, the values for Berger are not that changed. The POPE value predicted by CHARMM 36 (in the mixture POPC/POPE 50:50) is much smaller than that predicted by Berger.

This is interesting, but not exactly the core message of the manuscript. Maybe we should mention this very briefly? For example, we could just report the areas per lipid (without distinguishing PC and PE) and mention the difference between CHARMM36 and Berger. This as Todo point, currently in page 9.

ohsOllila commented 3 years ago

The focus of the manuscript is now updated and the results from the mixtures are moved to supplementary information. Therefore, I think that we should not go into this discussion in this manuscript (although this is an important point). Therefore, I close this issue.