ADCP 1.0 documentation

You are viewing ADCP 1.0 (deprecated)

These pages describe version 1.0, distributed with ADFRsuite 1.0. For new work use ADCP 1.1.

AutoDock CrankPep (ADCP) is a program for docking flexible peptides into receptors. The source of this version, as published in 2019, is archived on GitHub.

Tutorials

FAQ

How many steps and replicas are enough for my peptide?: The longer the peptide, the harder the docking problem, so run more replicas with more steps if resources allow. As a guideline, run 100–300 replicas with a maximum of 1–3 million Monte Carlo steps per amino acid (10–30 million steps for a 10-mer).
Should I start from a helix, an extended structure, or a mixture?: If you know something about the peptide’s conformation, start from that conformation. Starting from one conformation does not prevent sampling of others with different secondary structure. If nothing is known, start from a mixture of 80% extended and 20% helical conformations with -p 20 (the percentage of runs that start as a helix). Without -p, the case of the sequence given with -s sets the start: lower case for extended (coil), UPPER case for helix.
Why are there clashes in the output structure?: ADCP builds side chains from a discrete rotamer library and uses a coarse-grained scoring function for the internal energy. The output can therefore have minor clashes and non-ideal bond angles and lengths. Minimize the structure with any molecular mechanics force field to fix them. (Version 1.1 adds OpenMM minimization.)
Where are the caps at the peptide termini?: Peptide termini come in different types, and adding the proper caps did not conclusively help docking. ADCP therefore docks without caps. You can add caps to the output structure with software such as VMD or PyMOL.