ProDive methods data and precomputed validation results

Data inputs, fixed sampling resources, parameter-sensitivity outputs, and structural validation results for ProDive, a GPU-accelerated workflow for detecting local similarities between profile hidden Markov models using symmetric KL divergence. The archive includes Pfam seed profiles and alignments, final Pfam–Pfam segment correspondences, HHsearch comparison resources at 20% and 70% probability thresholds, RMSD results, and resources for reconstructing the Pfam structure collection. Source code is distributed separately in the ProDive methods repository.