search

OpenSourceMycetoma / Series-1-Fenarimols

Open Source Mycetoma's First Series of Molecules
10 stars 2 forks source link

Molecule size & effects on grain penetration #24

Closed rosalynclz closed 3 years ago

rosalynclz commented 4 years ago

Below is the data on molecular weight and polar surface area of 19 compounds tested in the in vivo larvae model. (Data on PSA were obtained from ChemSpider ACD predictions). image Graph of in vivo efficacy vs molecular weight: image Graph of in vivo efficacy vs PSA: image

Would @bendndi and @edwintse mind suggesting some methods/software to calculate the approximate size/diameter of the compounds above as I am interested in studying the effect of size of compounds on their in vivo efficacy and would like to plot a graph of those variables to determine if there is a size exclusion barrier cut-off points for the grains?

mattodd commented 4 years ago

Nice analysis. Looks like logD more important factor than MW or PSA. Are we overlooking anything?

MFernflower commented 4 years ago

Following with interest

mattodd commented 4 years ago

Just pinging @bendndi and @edwintse on this in case they have any chance to look at ways to calculate approximate size/diameter of the compounds above, or indeed anyone else with a preferred method? I know the answer depends on the question you ask, but something that allows a reasonably uncontroversial assessment of size?

bendndi commented 4 years ago

@rosalynclz @mattodd One thing that could be interesting to check would be Principal Moments of Inertia PMI plot, which is more linked to shape than size. See here You would have to plot by PMI and then colour by in vivo activity to check any link

A simpler alternative to this which could be analysed in same way as the MW / LogD is molecular shape index available in datawarrior(could also look at molecular flexibility too):

image

There may be some other descriptors available via RDKit but I'm not sure how "simple" these would be to implement.

rosalynclz commented 4 years ago

@bendndi Thank you for the reply. Would looking at total surface area (Data warrior) and molecular volume (Molinspiration) and calculating the TSA:MV ratio be a suitable size approximator?

bendndi commented 4 years ago

@rosalynclz I guess you could look at the molecular volume alone, and also the TSA:MV - could be interesting.

mattodd commented 3 years ago

I finally uploaded Rosalyn Chan's very nice thesis that includes all this work - sorry for the delay and thanks to all those generous souls who helped Rosalyn with her calculations and ideas. The thesis is here with a CC-BY licence. I'll close this issue and install a link to it on the wiki at https://github.com/OpenSourceMycetoma/Series-1-Fenarimols/wiki/Current-Targets-and-Activities so this discussion is not lost.