AHoJ: Rapid, tailored search and retrieval of apo and holo protein structures

C. P. Feidakis1, R. Krivák2, D. Hoksza2, M. Novotný1

1Department of Cell Biology, Faculty of Science, Charles University, Viničná 7, 128 00 Praha 2, Czech Republic

2Department of Software Engineering, Faculty of Mathematics and Physics, Charles University, Malostranské náměstí 25, 118 00 Praha 1, Czech Republic


Studying protein – small molecule interactions can provide insights into diverse but fundamental topics in structural biology. Understanding the induced fit model, observing conformational shifts upon binding as well as exploring the promiscuity of a binding site are situations where both apo and holo snapshots of a protein are required. Machine learning applications that rely on structural information to predict such interactions, also stand to benefit from the availability of both bound and unbound states of the same protein.

There is currently no dataset or tool available, capable of generating a list of apo and holo conformations on demand, for a given protein structure.

Here we present Apo-Holo Juxtaposition (AHoJ), a web-based tool, that given a user-specified holo protein structure and one or more ligands that bind to it, identifies other structures that belong to the same protein, and classifies them as apo or holo, considering the users’ preferences. A reverse search can also be conducted, in the case of starting with a structure that does not bind any ligands. The tool performs a search within the experimentally solved structures in the PDB and provides the user with a list of structures that are superimposed and visualised. The results are processed and displayed chain by chain, where each chain is annotated with a list of bound ligands and scored for its similarity with the query. AHoJ also features a multiple input mode, allowing it to generate customised apo-holo datasets.