[Jmol-users] visualizing channels in proteins

[Theis, K. W. <kt...@we...>]

Hi,

I am trying to visualize the channel in aquaporin1, 1IH5. I used Eric Martz's PACUPP<https://proteopedia.org/w/PACUPP> scripts to get the residues lining the channel, and now wanted to use that selection to make a surface showing the channel. My best result so far was with a command such as this:

select 13 , 14 , 17 , 21 , 24 , 25 , 28 , 29 , 32 , 33 , 35 , 37 , 49 , 52 , 53 , 56 , 57 , 60 , 61 , 65 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 , 79 , 80 , 83 , 84 , 87 , 96 , 97 , 100 , 101 , 105 , 120 , 126 , 127 , 128 , 149 , 152 , 153 , 154 , 156 , 157 , 168 , 169 , 171 , 172 , 173 , 175 , 176 , 180 , 183 , 184 , 185 , 186 , 187 , 189 , 190 , 191 , 192 , 193 , 194 , 195

isosurface insideout molecular slab plane (151:A.C)(62:A.O)(81:A.CA) slab plane (50:A.N)(210:A.NE1)(126:A.NH1)

This shows the channel, slabbed at both entrances, with the outside of the channel (which is the inside of the protein) lit up. However, I get some artifact surface patches because some of the atoms I selected to define the channel are also on the outer surface of the protein.

Is there an established way to show channels in Jmol? I see two ways of getting rid of the outer surface patches. I could more carefully select the lining atoms (I just took the residue numbers from Eric's output, not atom by atom). Or I could "bathe" the protein in fake atoms to push out the outer surface so it would no longer show. Or a combination of both.

I'd be happy to learn a better way,
Karsten

--
Karsten Theis
Chemical & Physical Sciences Department
Westfield State College