This data consist in the models, topology files, parameters files and trajectories of molecular dynamics simulations of the catalytic domain of human HDAC9 and small molecule inhibitor "4-(2-(1H-pyrrol-2-yl)oxazol-5-yl) aniline", also known as compound 9 from publication: "Investigation of the histone deacetylase inhibitor potential of phorbazole analogues" by IA Nystad Reierth et al., see related publication field.

The goal of this study was to discriminate between two most likely binding modes (either aniline group "up" toward the solvent or aniline group "down" toward the protein core) of compound 9 in the HDAC9 catalytic site. Five molecular dynamic (MD) runs were started with the ligand in either orientation "up" or "down" and the average movement of compound 9 over the course of the simulation and formation and duration of hydrogen bonds between compound 9 and HDAC9 was analysed to assess the stability of each of compound 9 starting orientations. The starting orientations of compound 9 were determined through docking studies, which datasets are available there: "https://doi.org/10.18710/3DNRGM".

The data contains all files used for the 10 different molecular dynamic runs (5 "up" and 5 "down", including the instruction and input files for minimization, heating, pre-production and production compatible with Amber 2024, as well as parameter files for compound 9 (forcefield modification, S66_ligand.frcmod). The rest of the data is split across two folders, one for runs of compound 9 in the "up" orientation (run_ligandUp), the other for the "down" orientation (run_ligandDown). Each folder contains results from 5 independent MD runs (run 1 to run 5), including the trajectory (filename.nc), last restart file (filename.ncrst) and output file (filename.out) which contains the log for the runs. Each runs is split into a short preproduction run on CPU (prod_CPU_equilibration) followed by a longer GPU production run (prod_GPU). Finally, the coordinates (filename.crd), topology (filename.top), as well as the restart file from the heating run (filename_heat.ncrst), which is the same for each of the 5 runs, are found in the head folders (run_ligandUP or run_ligandDown). The folder structure might be easiest to understand when the files are set to "tree" representation on the main page of the dataset.

We recommend consulting the method section from the main article (see related publication field) to get more details regarding how the results were generated.

Amber 24, 2024

PMEMD CUDA, 18.0.0