Dataset "Inverse Theoretical Spectroscopy via Monte-Carlo Sampling — From Spectra to Ensembles of Molecular Structures"

Dataset: Inverse Theoretical Spectroscopy via Monte-Carlo Sampling — From Spectra to Ensembles of Molecular Structures Authors: Erik Tsvetaev and Christoph R. Jacob, TU Braunschweig ## Overview This dataset accompanies the publication 'Inverse Theoretical Spectroscopy via Monte-Carlo Sampling — From Spectra to Ensembles of Molecular Structures'. It contains Monte Carlo trajectories, simulation runscripts, and scripts for reconstructing the data from xyz coordinates. All simulations were performed using the `imce` framework in combination with `XTB` and `STD2`. The dataset covers three simulation strategies (tMC, sMC and w-sMC) across multiple temperatures and parameter settings, each replicated five times. ## Requirements The following software must be installed to reproduce or extend the results: - **imce** — Monte Carlo simulation and analysis framework (see https://github.com/erik-tsvetaev/imce/tree/v1.0.0; DOI: 10.5281/zenodo.20931648)- **xTB** — Semiempirical tight-binding code (see https://github.com/grimme-lab/xtb/releases/tag/v6.7.1)- **STD2** — Excited-state calculation backend (see https://github.com/grimme-lab/std2) Please refer to the imce installation manual for detailed setup instructions, including how to configure XTB and STD2 as backends. ## Directory Structure ```dataset/│├── ReadMe.md # This file|├── 00_dfhbi.xyz # Reference geometry of DFHBI|├── 01_analysis_scripts| ├── dfhbi.xyz, dfhbi.svg # reference structure| ├── recompute_trajectory.py # Recomputes properties along a trajectory from xyz format| ├── sim_uq_short.py # Runs a short simulation for verification of the code | └── analyse_trajectory.py # Analyses a single simulation (HDF5, obtained e.g. from sim_uq_short.py) and produces figures|├── 02_runscripts/ # Python runscripts for all simulations│ ├── sim_mrt2_100_{1..5}.py # MRT2 sampling at 100 K, 5 independent runs│ ├── sim_mrt2_300_{1..5}.py # MRT2 sampling at 300 K, 5 independent runs│ ├── sim_mrt2_500_{1..5}.py # MRT2 sampling at 500 K, 5 independent runs│ ├── sim_uq_015_{1..5}.py # UQ sampling, τ = 0.15, 5 independent runs│ ├── sim_uq_02_{1..5}.py # UQ sampling, τ = 0.20, 5 independent runs│ ├── sim_uq_025_{1..5}.py # UQ sampling, τ = 0.25, 5 independent runs│ ├── sim_windows_high_{1..5}.py # Windowed sampling, high-energy window, 5 independent runs│ ├── sim_windows_low_{1..5}.py # Windowed sampling, low-energy window, 5 independent runs│ └── sim_windows_mixed_{1..5}.py # Windowed sampling, mixed window, 5 independent runs│└── 03_trajectories/ # Concatenated xyz trajectories (stride = 500, n_sim = 5) ├── mrt2_100_traj.xyz.gz ├── mrt2_300_traj.xyz.gz ├── mrt2_500_traj.xyz.gz ├── uq_015_traj.xyz.gz ├── uq_02_traj.xyz.gz ├── uq_025_traj.xyz.gz ├── windows_high_traj.xyz.gz ├── windows_low_traj.xyz.gz └── windows_mixed_traj.xyz.gz```