Foldit Wiki

Tryptophan is one of twenty amino acids that make up proteins.

Foldit is mainly about shapes. You don't need a science degree to be successful at the game. But you'll see a lot of scientific terms floating around, so it's good to have little background.

Here's a quick introduction to biochemistry, Foldit-style.


Foldit involves folding proteins into a three-dimensional shape.

Proteins are chains of amino acids.

The chain of amino acids is called the backbone of the protein.

The backbone of the protein is held together by strong peptide bonds.

The sequence of amino acids is called the primary structure of the protein. It's what makes a protein unique.

Each amino acid has a sidechain that sticks out from the backbone. The sidechain is what makes each amino acid unique.

Glycine is the only amino acid that has no sidechain. The lack of a sidechain makes glycine the most flexible amino acid.

See the Amino Acid Gallery to get an idea of what each amino acid looks like.


Inside a cell, protein chains are assembled by ribosomes. Ribsomes are a type of "molecular machine", which build proteins based on information found in DNA, with the help of RNA, in a process called translation.

The hard part is what happens next. Proteins are generally expected to have a specific shape, according to Anfinsen's dogma. But all ribosomes do is assemble a chain of amino acids. As the chain leaves the ribosome, the protein folds up on its own. There's no folding machine to give the protein its final shape.

There are lots of ways a protein could fold up, but generally only one way it does fold up. This led to Levinthal's paradox, which points out that it would would take longer than the age of the universe for a small protein to fold up if it tried every possible shape. Small proteins actually fold in a millesecond or less.

Foldit is used to study both how proteins fold, and to determine the shapes they fold into.


Using default view options, hydrophilic sidechains are blue, hydrophobic sidechains are orange.

Water is believed to be one of the major factors in shaping proteins.

Many proteins are normally surrounded a mixture of water and other chemicals. Proteins have core or inside, where the water can't reach.

Amino acids are classified as either hydrophilic or hydrophobic. Foldit shows hydrophilics in blue and hydrophobics in orange (using default view options).

Hydrophobic amino acids are repelled by water. These amino acids should be buried or hidden in the core as much as possible.

Hydrophilic amino acids are OK with water, so they're more likely to be found on the surface of a protein.

Long hydrophilics like arginine and lysine are hard to fit in the core, but shorter hydrophilics like serine are often found in or near the core.

The standard scoring in most Foldit puzzles rewards proteins which have most of their hydrophobic amino acids in the core of the protein. This type of protein is known as a globular protein. Other types of proteins, particular membrane proteins, don't have the same issues with water. Some Foldit puzzles have had modified scoring to deal with membrane proteins.


Various types of chemical bonds hold a protein together.

The backbone of the protein is held together by strong covalent peptide bonds. You can't break peptide bonds in Foldit, although you can insert temporary cutpoints, which allow you to move parts of the protein around.

Hydrogen bonds between sidechains of a protein.

Hydrogen bonds are much weaker than peptide bonds, but they're important in stabilizing the protein. Foldit shows hydrogen bonds as blue-and-white "candy cane" spirals. Unlike peptide bonds, you can make or break hydrogen bonds while playing the game.

Another type of special bond in Foldit is a disulfide bridge, a covalent bond which connects the sulfur tips of two cysteines. Foldit shows disulfide bridge as yellow-and-green spirals. As a covalent bond, a disulfide bridges are much stonger than hydrogen bonds, but you can still make and break them in Foldit.

Other types of bonding, such as salt bridges and pi stacking contribute to holding proteins together, but don't have a special visualization in Foldit.

Secondary structure[]

Hydrogen bonds appear as blue spirals between segments of a helix.

The primary structure is the sequence of amino acids that make up a protein.

Hydrogen bonds can form patterns that shape parts of the protein into secondary structures. The default "cartoon" view emphasizes the secondary structure.

A helix is one type of secondary structure. A helix is spiral. The amino acids in a helix form hydrogen bonds with each other.

A sheet is another type of secondary structure. Foldit shows sheets as flat structure with zig-zag or lighthning bolt pattern. Each sheet bonds edgewise with one or more other sheets.

Any part of the protein that's not a helix or a sheet is called a loop.

Helix, sheet, and loop are the three types of secondary structure.

In Foldit, you can normally change the secondary structure assigned to any part of the protein. Simply changing the secondary structure doesn't change the shape of the protein, but it may influence how tools like rebuild work.

Foldit also has an auto structures tool, which looks at patterns of hydrogen bonds, and assigns the correct secondary structures for the current shape of the protein.

More atoms and molecules[]

Amino acids are made of up of just five types of atoms:

  • hydrogen
  • carbon
  • nitrogen
  • oxygen
  • sulfur

(Once in a while, selenium may slip in, replacing sulfur.)

Proteins interact with other types of chemical compounds. They may grab on to calcium or copper, either temporarily or more permanently. They may bond to other proteins or other types of molecule.

Foldit has several ways of working with these other types of interactions, most commonly in a design puzzle.

Foldit symmetry puzzles involve a protein bonding with one or more identical copies of itself.

Some design puzzles feature a ligand, another type of atom or molecule that bonds to the protein. In this type of puzzle, the ligand normally appears as one or more segments of the protein, which have secondary structure type "M" for molecule.

In "docking" puzzles, Foldit players design a small protein intended to bond or "dock" with a specific area on a larger target. Viruses like Ebola and Marburg have been one type of target. In this type of puzzle, much of the target is normally locked.

One area under development in Foldit is drug design. The drug design feature would allow Foldit players to work with general biochemistry, designing non-protein molecules to bind to proteins.

Scientific goals of Foldit[]

Scientists have various techniques to study the shape of proteins, such as X-ray crystallography and Nuclear Magnetic Resonance.

Both X-ray crystallography and NMR have limitations. Foldit can help refine results to get more accurate structures. For example, the protease from the Mason-Pfizer monkey virus had been "solved" by NMR, but the various solutions didn't agree very well with each other, or the results from X-ray crystallography. Foldit players were able to refine NMR models, which led scientists to improved solution for this protein.

Foldit is also used to study how proteins fold up. One of the major goals of the Foldit science team is to create proteins that fold up on their own in the same way that natural ones do. Many Foldit design puzzles involve desiging completely new proteins from scratch. The Foldit team evaluates the results of these puzzles, and design which seem promising are tested in the lab, by growing the new proteins in bacteria.