Samri Gebre




Protein Folding

Research with Professor Joseph O'Rourke

Summer '05



Week 1

We started working on the different research projects on May 16. Gillian and I started out working on the protein folding project. Our initial point was trying to find a locked 45 degree unit chain, using connects. For the moment we have deferred from the protein folding project until we figure out what to do next and have joined Diana on working with the 3D printer. We have spent some time learning to use Alias Studio 12 through the tutorials. We have been watching the tutorial videos to help us with creating basic shapes and objects. Our project was to create different shapes of Borromean Rings. To see examples of Borromean Rings created using Alias Studio 12 see Gillian's Page.

May 20

Today we set up the 3D printer, to print the five sets of Borromean Rings. We also learned to use Dreamweaver so we can keep web logs of our research. One of our goals is to print out an image of a protein backbone in the 3D printer. I spent sometime browsing the protein explorer web site. At the moment I am looking into how to save the 3D protein models into STL files. Stereolithography (STL) is a format that uses a mesh of triangles to form solid objects. In this format the triangles share both common sides and vertices.

Figure 1. A backbone model of the protein deoxyhemoglobin. This was taken from the protein explorer web site and edited using SnagIt.

Week 2

May 23

We've been struggling to extract 3D protein models from Protein Explorer. Joe thought this might not be our only option and so I've been looking at a couple of other web sites to help me along. One of the software I have been using is SnagIt. SnagIt allows people to capture images from their screens and is compatible with various imaging formats. A 30 day trial of the SnagIt software and a detailed description of its functions can be found online.

May 24

I am still looking for STL protein model. I have been looking at a web site by Michael Sanner. He is a professor at Scripps Research Institute. He uses a number of software, in particular the Python programming language to help him with 3D visualization. I also started exploring trefoil knots. Joe wants to print out a circular trefoil knot, and has given me a web site to look into.

May 25

I spent most of today exploring a web site by Herbert Edelsbrunner. In the site there are numerous kinds of beautiful trefoil knots. In the site there are numerous kinds of beautiful trefoil knots. Although they were all in STL format, I wasn't able to extract a model that has the right size for printing. I finally found trefoil knot in Mathematica that I was able to adjust the size for printing.

Figure 2. A circular trefoil knot.


Week 3

May 31

At the moment I am giving up on the stick model. Joe thought it would probably be fragile anyway. And now I am looking for a sphere model of any protein and a stick model of the same protein, because Joe wants to make a 2D picture of it. Today we also went back to looking at locked 45-deg chain for a brief moment.


June 1

I bought (with Joe's money) two boxes of Chemistry molecular models sets. I explored the Protein Data Bank and found some really cool 3D models of proteins. I looked at Acetylcholine receptor M2 and found its bond angles and bond length. Diana is trying to draw the molecules in Alias using the information about the angles and the length.

June 2

I found out the Acetylcholine Receptor M2 model is in VRML format. It is exciting to finally find a Z-print compatible model. I was able to extract it from the Protein Data Bank in the form I want it in using the table set up they have on the site. I also made a model of the protein, using the Chemistry molecular model sets.

Figure 3. Sphere, backbone and ball & stick models of Acetylcholine Receptor M2.

June 3

We had another exciting print run today. Gillian, Diana and I cleaned our workstation. We had visitors coming in to see the 3D printer and our models. I spent sometime looking at pictures of Knight's Visor on line.

We had a wonderful picnic. I played soccer and a little bit of football.


Week 4

June 6

In the morning Gillian and I spent sometime taking the models out from Friday's run. I sifted the powder while Gillian de-powdered and cleaned the models. I was a bit disappointed…we broke the protein model while Z-bonding…oh well we'll just print it again in our next run.

June 7

I started looking at the Dowel-brass eyelet model Gillian and Meghan made for Joe. Joe would like the model to be made using 11-chains. We took an inventory of available materials. We realized we need a few extra materials and so we'll be going to the hardware store to get them.

June 8

I started looking up more into coordinates and radii of the atoms from the Protein Data Bank. Today was an awfully warm day. Diana, Gillian and I walked to the hardware store to shop for parts. It was a huge mistake…we didn't know how to get to the sore and so we ended up walking extra…we were tired and dehydrated by the time we got there. Thanks to Kitu we didn't have to walk back.

June 9

I am further investigating to see if we can build sphere model of the protein using either Alias or Mathematica.

June 10

I found all the coordinates and the radii of each atom for the Acetylcholine Receptor M2. Joe was able to edit the coordinates and so we can use them in Mathematica. We finally have a sphere model was created using Mathematic. We'll be printing both the backbone and sphere models in our next run.


Week 5

June 13

I spent some time editing the sphere model. Joe wants it to be the same size as the backbone model. I also started making my first model of the 11-chain with the new dowels & eyelets.

June 14

I have finished building the 11- chain model. There are a few minor things I need to fix. One of the threads is bent and so I need to adjust the size on one of the pieces on the 3/4 chain.

June 15

We had a print run today. We are 2 inches short in powder and so we're hoping everything will work out.

June 16 & 17

I have finished painting the 11-chain and 2-chains. Joe has proved that nearly-unit chains can lock for 60 degrees. I am starting to look into > 60 degree angles and see if they lock.

figure 4. The 11-chain locked model built using dowels and eyelets.


Week 6

June 20

We found out the powder for the 3D printer is getting low in quality. We need to order new powder before we start seeing the effect on our models.

Today Joe brought in images of 60-deg (equilateral triangle) and 108-deg (regular pentagon) that he wants models of. It was harder than I thought to make a 60-deg angle with dowels.


June 21

The sphere model was too heavy and it broke while we were de-powdering it. In our next run we'll have to make a much smaller size (about 6 inches.) I am also trying to figure out the bond angles for the backbone model. The PDB has information about the geometry of the protein and so I need to figure out a way to understand and interpret the numbers.


June 22

I Z-bonded some of our models and painted the backbone model. I also spent some time working with Anna as she helped me edit the 11/2 chain photos in photo shop.

June 23

I spent most of the time working on my summer abstract. Joe went over on possible ideas I can write about and helped me with the editing along the way. Also thanks Faith for editing my page and Anna & Nell for helping me with the pictures.

June 24

Today is my last day of research. In the morning all of us went to AWIS conference to hear Rita Colwell's talk.