Illustrate: Non-photorealistic Biomolecular Illustration

Illustrate is a small Fortran program for creating non-photorealistic illustrations of molecules, with cartoony colors and outlines, and soft ambient shadows. I originally developed it during postdoctoral training in the laboratory of Arthur J. Olson at the Scripps Research Institute, and have tinkered with it ever since. I have used it to create materials for my books and for PDB-101, the outreach portal of the RCSB Protein Data Bank, most notably, for the “Molecule of the Month.”

With the help of Ludovic Autin, I am finally releasing a version of the program that is generally useable. Ludo has designed a great web interface (below) that allows you to position a molecule from the PDB using NGL, and then create an illustration. I am also releasing the Fortran code (for free use and with many apologies) for users who want to get their hands dirty and customize their illustrations.

I hope you enjoy creating these illustrations as much as I do!

David S. Goodsell


Please cite:

Goodsell DS, Autin L, Olson AJ (2019) Illustrate: Software for Biomolecular Illustration. Structure (Epub ahead of print), doi: 10.1016/j.str.2019.08.011.

Development of Illustrate has been kindly supported by Damon Runyon-Walter Winchell Cancer Research Fund Fellowship DRG 972, the US National Institutes of Health R01-GM120604 and the RCSB Protein Data Bank.


Known issues:

–the web interface is having trouble with biological assemblies in some cases, so you might get surprising results–we’re sorting out the problems now

–the web interface will fail with very large structures

–if you load in your own PDB files, the web interface may have trouble assigning protein/nucleic acid chains to create the default view. Try using a few of the other “style” options if you don’t get an image.

–the web and standalone versions currently have problems with parsing the chain IDs for structures with many different chains. If you’re getting pictures that only include a few of the chains, that may be the problem. I’m currently working on a fix, but a kludgey solution is to delete all of the header information from your PDB file. Ugh.

Complex of tRNA with EF-Tu, PDB entry 1ttt.