9 replies [Last post]
LociOiling's picture
User offline. Last seen 1 hour 29 min ago. Offline
Joined: 12/27/2012
Groups: Beta Folders

Following a suggestion from Drum, I decided to try building one of the nine FDA-approved protease inhibitors.

I picked TPV, otherwise known as tipranavir.

Here's what TPV looks like in JMol:

Tipranavir - TPV: TPV as seen in JMol with atom numbers added.Tipranavir - TPV: TPV as seen in JMol with atom numbers added.

I've added the atom numbers used in the PDB definition file.

LociOiling's picture
User offline. Last seen 1 hour 29 min ago. Offline
Joined: 12/27/2012
Groups: Beta Folders
Here's a first attempt to build TPV in Foldit

Here's my initial attempt to build TPV in Foldit. I haven't gotten to the point of counting atoms yet. This version gets a substantial cLogP penalty, -800 or -900 points. From what I can find, cLogP is a measure of hydrophobicity, the higher the value, the more hydrophobic the molecule.

From previous puzzles, we've seen that Foldit shows the bonding patterns in rings a little differently than other tools. This has something to do with resonance.

Attempt to build TPV in FolditAttempt to build TPV in Foldit

LociOiling's picture
User offline. Last seen 1 hour 29 min ago. Offline
Joined: 12/27/2012
Groups: Beta Folders
One area of difference....

The sulfur at S30 is shown double-bonded to two oxygens in JMol and in chemical diagrams.

So far, I haven't been able to reproduce this bonding in Foldit, getting "Modification is not chemically feasible!" when I try.

Here are the JMol and Foldit views. The sulfur is the yellow atom, the oxygens are the red atoms on either side.

Detail of TPV showing sulfur double-bonded to two oxygen atomsDetail of TPV showing sulfur double-bonded to two oxygen atoms

toshiue's picture
User offline. Last seen 2 hours 23 min ago. Offline
Joined: 01/31/2016
Groups: Go Science
thank you!

These are wonderful, thank you. Now I just need to get the knack of the Ligand Design tool.

LociOiling's picture
User offline. Last seen 1 hour 29 min ago. Offline
Joined: 12/27/2012
Groups: Beta Folders
maybe a video...

I'll see if I can do it again, time for yet another Foldit video.

Meanwhile, it looks like I managed to get all the atoms, my count matches the formula C31 H33 F3 N2 O5 S for the heavy atoms at least. (Haven't counted the hydrogens.)

I've updated the picture with the atom numbers. A couple of the carbons are hiding behind other carbons, as indicated by the curved arrows. They are present when I rotate the molecule in Foldit.

TPV in Foldit with numbered atomsTPV in Foldit with numbered atoms

Joined: 09/22/2017
Groups: Beta Folders
Sulfonamides are broken in foldit

I ran into the same issue when I tried to add a sulfonamide group to something I was working on

LociOiling's picture
User offline. Last seen 1 hour 29 min ago. Offline
Joined: 12/27/2012
Groups: Beta Folders
I think I have a procedure for the double bonds...

Still working on a video demo, but here are the steps to get a sulfur double-bonded with two oxygens, as discussed above.

It's easiest to work in the selection interface. I use Ligand Specific coloring and Cartoon Ligand view.

Select the ligand (segment 199) and click on the Design Ligand (L) tool. The ligand gets deselected easily, so you'll most likely need to repeat this step several times. The instructions below assume you're in ligand design mode.

The starting structure has three carbons, with a red oxygen attached to the middle carbon.

Click on one of the white hydrogens attached to one of the carbons on the end. The click on N for nitrogen. The hydrogen is replaced by a blue nitrogen with three attached hydrogens.

Click on one of the hydrogens attached to the nitrogen you just added. Click on C for carbon. You'll get a new carbon, again with three hydrogens.

Click on one of the new hydrogens and click on O for oxygen. Do the same for one of the other new hydrogens.

You'll now have a carbon that's bonded to the nitrogen you added, plus two oxygens and hydrogen.

Click on the carbon and then click S for sulfur. The carbon is replaced by a yellow sulfur, but everything else stays the same.

Click on the sulfur, then on one of the attached oxygen atoms. The click on the double bond button. The oxygen bond is now double-bonded to the sulfur. Repeat this for the other oxygen atom.

When you're done, you'll have a sulfur bonded to the nitrogen, and double-bonded to two oxygen atoms. There's also an extra hydrogen attached to the sulfur, which allows you to continue the molecule.

It's easy to select the wrong atom. You can click an atom again to deselect, or use the "clear selection" broom icon to clear all selected atoms.

Unlike other types of selection, right-clicking, control-clicking and shift-clicking have no effect. It's a simple (left) click to select and deselect atoms in ligand design mode.

You can also get to the same configuration without going through adding a carbon. Just add the sulfur. You'll get a sulfur with one hydrogen. Change the hydrogen to oxygen. The oxygen is now double-bonded to the sulfur, and there's another hydrogen. Change the second hydrogen to oxygen, then convert to a double bond. You'll end up with the two double-bonded oxygens and an available hydrogen for expansion.

One other source of confusion is "show bondable atoms". This gives you spheres around the oxygens and nitrogens that resemble the selection sphere in ligand design mode. Turn off show bonds and show bondable atoms while you're designing your ligand.

LociOiling's picture
User offline. Last seen 1 hour 29 min ago. Offline
Joined: 12/27/2012
Groups: Beta Folders
Video...

Here's a video showing the steps...

https://www.youtube.com/watch?v=qQIdtGQLDmI

Joined: 09/22/2017
Groups: Beta Folders
This is a compound I came up with on the spot if anybody wants

https://opsin.ch.cam.ac.uk/opsin/4%E2%80%90%7B4%E2%80%90[(5%E2%80%90amino%E2%80%906%E2%80%90hydroxy%E2%80%901%E2%80%90methoxyhexan%E2%80%903%E2%80%90%20yl)amino]but%E2%80%902%E2%80%90en%E2%80%901%E2%80%90yl%7D%E2%80%901%E2%80%90(isopropylsulfanyl)%E2%80%909%E2%80%90%20methoxy%E2%80%906,8%E2%80%90dimethyldecan%E2%80%901%E2%80%90one.png

jeff101's picture
User offline. Last seen 1 hour 30 min ago. Offline
Joined: 04/20/2012
Groups: Go Science
JellyJump's ligand:

Click HERE for the original image of JellyJump's ligand below:

Sitemap

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, RosettaCommons