Bayesian exploration of the composition space of CuZrAl metallic glasses for mechanical properties

This dataset corresponds to the analyses reported in the npj paper: Bayesian exploration of the composition space of CuZrAl metallic glasses for mechanical properties.npj Computational Materials, 11(1), 96. 2025.Authors: Mäkinen, T., Parmar, A. D., Bonfanti, S., & Alava, M. J. Please refer to the paper for full methodology and analysis details. What this dataset contains--------------------------Simulation outputs and derived metrics used in the npj paper. The structure mirrors our otherCu–Zr(–Al) metallic‑glass datasets: results are grouped by preparation/cooling rate and then byindependent replicate (iteration). Top‑level layout----------------OA_npj/ ├── 10_11/ ├── 10_12/ ├── 10_13/ └── figure/ Naming convention-----------------10_{NN} → preparation/cooling rate 10^{NN} (e.g., 10_11 → 10^11 K/s).iter_{k} → independent replicate run / iteration index (k = 2 … 12 in 10_11 shown below). Example: 10_11/---------------10_11/ ├── iter_2/ ├── iter_3/ ├── iter_4/ ├── iter_5/ ├── iter_6/ ├── iter_7/ ├── iter_8/ ├── iter_9/ ├── iter_10/ ├── iter_11/ └── iter_12/ Typical per‑iteration contents (filenames may vary by workflow)---------------------------------------------------------------• processed data used in figures/tables (e.g., stress/strain series, detected events)• logs/metadata describing the run and parameters• optional raw simulation snapshots/dumps if retained Coverage summary----------------Cooling‑rate folders present: 10_11, 10_12, 10_13.Each contains multiple iter_{k} subfolders with independent seeds/replicates. How to cite-----------Please cite the npj article above and the Zenodo record DOI for this dataset. ----------- SB acknowledges support from the National Science Center in Poland through the SONATA BIS grant DEC-2023/50/E/ST3/00569 (PI: SB). SB acknowledges support from the Foundation for Polish Science in Poland through the FIRST TEAM FENG.02.02-IP.05-0177/23 project (PI: SB).