Predicted and experimental protein-ligand coordinates with distance labels and evaluation splits
- 1Leibniz University of Hannover
Abstract
Structure prediction tools now provide protein-ligand geometry for pairs no laboratory experiment has resolved thus far. This data is typically provided without calibrated, practically usable per-system measures of whether a given complex is correct and how much to trust it. Agreement between independently constructed methods is an established signal for this type of problem. What has been missing is a way to derive what a given level of agreement is worth. We present PLI-Parallax, which deposits predicted geometry together with experimental data needed to calibrate it. Chai-1, two Boltz-2 configurations, and the docking engine smina were run over shared inputs across an experimentally resolved crystal tier of 19,350 complexes and a corpus tier of 31,746 cross-docked pairs without experimental ground truth on predicted receptors, yielding 307,314,646 residue-to-ligand-atom distance records, which the stored coordinates allow a consumer to recompute at a cutoff of their own choosing. Their mutual agreement is fitted against observed accuracy where experimental data permits it and carried to where it does not. This way 30,567 systems carry predicted label accuracy together with a split-conformal interval. On the crystal tier that interval covers observed accuracy at the stated rate. On the corpus tier it ranks systems by expected label quality, since both distributions differ. The deposit is accompanied by 646 evaluation configurations across seven data split families, and 631 of them report how far their training and test entities separate under a two-sample test. The 906 protein accessions were partitioned to control sequence leakage, so a protein-cold split here tests generalisation across sequence space. The two tiers support evaluation against experimental data and, where this is absent, supervision weighted by how far the configurations agree.
Related
- Bioinformatic and experimental analyses revealed pathogen-derived pathoPEPs candidates predicted to control plant miRNAs expression and infection
- Dynamic membrane organization coordinates protein-O-mannosylation and proteostasis in mycobacteria
- Predicted structure of the 89-kDa invasion-tip protein of Holospora obtusa, an endonuclear symbiont of Paramecium caudatum, suggests a pH-dependent conformational change
- Predicted condensation-related signatures localize to two C-terminal regions of the prion protein across cervid PRNP variants associated with chronic wasting disease
- The choice of structural template can create apparent links between predicted protein stability and longevity in tetrapods
BibTeX
@misc{klamt2026predicted,
title = {{Predicted and experimental protein-ligand coordinates with distance labels and evaluation splits}},
author = {T. Klamt},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.06.749709},
url = {https://www.biorxiv.org/content/10.64898/2026.09.06.749709v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Predicted and experimental protein-ligand coordinates with distance labels and evaluation splits
AU - T. Klamt
PY - 2026
DA - 2026/09/06
PB - bioRxiv
DO - 10.64898/2026.09.06.749709
UR - https://www.biorxiv.org/content/10.64898/2026.09.06.749709v1
AB - Structure prediction tools now provide protein-ligand geometry for pairs no laboratory experiment has resolved thus far. This data is typically provided without calibrated, practically usable per-system measures of whether a given complex is correct and how much to trust it. Agreement between independently constructed methods is an established signal for this type of problem. What has been missing is a way to derive what a given level of agreement is worth. We present PLI-Parallax, which deposits predicted geometry together with experimental data needed to calibrate it. Chai-1, two Boltz-2 configurations, and the docking engine smina were run over shared inputs across an experimentally resolved crystal tier of 19,350 complexes and a corpus tier of 31,746 cross-docked pairs without experimental ground truth on predicted receptors, yielding 307,314,646 residue-to-ligand-atom distance records, which the stored coordinates allow a consumer to recompute at a cutoff of their own choosing. Their mutual agreement is fitted against observed accuracy where experimental data permits it and carried to where it does not. This way 30,567 systems carry predicted label accuracy together with a split-conformal interval. On the crystal tier that interval covers observed accuracy at the stated rate. On the corpus tier it ranks systems by expected label quality, since both distributions differ. The deposit is accompanied by 646 evaluation configurations across seven data split families, and 631 of them report how far their training and test entities separate under a two-sample test. The 906 protein accessions were partitioned to control sequence leakage, so a protein-cold split here tests generalisation across sequence space. The two tiers support evaluation against experimental data and, where this is absent, supervision weighted by how far the configurations agree.
ER -