What it does
A structure file lists where every atom of a protein (or DNA, RNA and any bound molecules) sits in space. This viewer reads that file inside your browser and lets you look at it and measure it:
- 3D view. Turn it with a finger or the mouse, pinch or scroll to zoom. Pick a cartoon (thick tubes for helices, arrows for strands, thin lines elsewhere), a backbone trace (straight lines between C-alpha atoms) or all atoms as sticks. Ligands are drawn as balls and sticks; water is hidden until you tick it.
- Colours by secondary structure, chain, B-factor (how much an atom moves or how uncertain it is: blue low, red high), position along the chain (blue at the N end to red at the C end) or element.
- Chains, residues and ligands: how many of each, the numbering, every chain’s sequence with helix and strand letters coloured, and which residues sit within 4 Å of each ligand.
- Distances: tap two atoms in the view, tap two letters in a sequence, or pick two residues from the lists.
- Contacts: every pair of residues with any atoms closer than a cutoff, with the closest pair of atoms, as a table and a CSV.
How to use it
- Press Open a PDB or mmCIF file (or drop the file on the view, or paste its text). Files ending .pdb, .ent, .cif and gzipped .gz all work.
- Turn and zoom the view. Reset view brings back the starting angle, which lays the longest side of the molecule across the screen.
- Tap an atom, then a second one, to see the distance. Clear picks removes the marks.
- Set the contact cutoff and how many neighbours along the chain to skip, then download the list.
Options
- Secondary structure. Most files from crystallography or cryo-EM carry HELIX and SHEET records (in mmCIF, struct_conf and struct_sheet_range) and the page uses them. When a file has none, such as a model straight out of a prediction or simulation program, the page makes a rough guess from the C-alpha atoms: five residues in a row whose C-alpha spacings fit a helix are called helix; extended stretches are called strand only when another extended stretch lies alongside within 5.6 Å. You can switch to the guess for any file to compare. It is not a full hydrogen-bond assignment such as DSSP.
- Cutoff. 4 Å is a common choice for side-chain contacts and hydrogen bonds, 3.5 Å keeps mostly hydrogen bonds and salt bridges, and 6 to 8 Å is used for looser residue contact maps.
- Skip neighbours. Residues next to each other along a chain always touch through the peptide bond, so they are left out. The default 3 skips the next two residues; raise it to see only long-range contacts.
- Show all contacts, only those between chains (an interface), or only those with ligands and water (a binding site). Tick Count water to include water, and Count hydrogens if your file has them and you want them counted.
Worked example
The page opens with a made-up model: chain A has a helix, a short loop and a two-strand hairpin, chain B is a short helix lying against chain A’s helix, and an invented ring compound called LIG sits by the hairpin. It is drawn from 240 atoms (only backbone and C-beta atoms, no full side chains). The page reads 2 chains, 46 residues, 1 ligand, with 57 % helix, 26 % strand from the file’s HELIX and SHEET records.
The residue boxes start on A GLU 4 and B ALA 10: the closest atoms are A GLU 4 O to B ALA 10 CB, 3.316 Å, and the two C-alpha atoms are 4.468 Å apart. At a 4 Å cutoff skipping 3 neighbours there are 64 residue pairs; choose Between chains only and the list drops to 7 pairs, the interface between the two helices. Choose With ligands and water only and tick Count water: 4 pairs are left, A TYR 29 next to LIG (3.356 Å) and three waters.
Tips
- On a big structure the all-atoms style draws thousands of sticks; the cartoon is quicker to turn.
- Only the first model of an NMR ensemble is read, and where an atom has alternate positions only the first is kept. The page says so under the chain table.
- The contacts CSV has every pair with chain, residue name and number, closest distance, the two atoms and the number of atom pairs within the cutoff, ready for a spreadsheet.
- Download the view as a PNG on white for slides, or the whole report as an image, PDF or Word file.
FAQ
Is my file uploaded? No. It is read by the page in your browser and stays on your device. Nothing is stored after you close the page.
Why does the guess differ from the file’s records? The guess only looks at C-alpha spacing, so it often runs a helix one residue longer and trims the ends of strands next to turns. The records in a deposited file come from the depositors’ own assignment.
Can it open AlphaFold or other predicted models? Yes, they are ordinary PDB or mmCIF files. In predicted models the B-factor column usually holds the confidence score, so colouring by B-factor shows confident parts in red and uncertain parts in blue.
Why are some residues missing from the trace? Gaps in the numbering or C-alpha atoms more than 4.2 Å apart break the line, because part of the chain was not seen in the experiment.
Where the numbers come from
- File reading (PDB columns and mmCIF tables), the secondary-structure guess, distances, contacts and the 3D drawing are written by fibub. The guess uses C-alpha distance windows in the style of the published P-SEA method, with ranges chosen by fibub.
- Checked in the tests against Biopython 1.88 on the example: atoms, residues, chains, distances, every residue contact at four settings (NeighborSearch), mmCIF tokens (MMCIF2Dict) and the same structure written as mmCIF; the principal axes against NumPy; an ideal helix worked by hand.
- The example structure is made up by fibub (built from chosen backbone angles, with an invented ligand and made-up B-factors). It is not an experimental structure or a Protein Data Bank entry.
- A file you open is read on the page only. It is never uploaded or stored.
Standard scientific values, such as constants, atomic weights and melting points, are facts found in every textbook; we name where ours were checked. If you are an author or publisher and want a value, a credit or a table changed or removed, tell us and we will act quickly.
Last updated 2026-11-09.