Recovering molecules from coarse-grained beads: free-energy-conditioned generative backmapping across chemical space
Abstract
Transferable coarse-grained (CG) force fields compress chemical space: by aggregating atoms into a reduced set of interaction beads, models such as MARTINI reduce the number of distinguishable compounds by roughly three orders of magnitude, making high-throughput screening of thermodynamic properties tractable across soft matter, with drug--membrane permeability as a well-developed example. The compression is lossy and, so far, one-way: a screen returns a combination of beads, with no established route back to the compounds it stands for. Recovering those compounds--compositional backmapping--is a one-to-many inverse map, distinct from the better-studied conformational problem of rebuilding atomic coordinates from a known mapping. Here we formulate compositional backmapping as conditional graph generation by introducing juniper, a discrete denoising diffusion model over molecular graphs conditioned on the octanol--water partition free energy ΔGW ↦ O, the principal driver of MARTINI bead type assignment and hence a proxy for bead identity. Trained on molecules of up to 9 heavy atoms mapped onto one or two beads, juniper generates molecules that are 93\% valid and 92\% unique for two-bead targets, and whose ΔGW ↦ O distributions track the target ΔGCGW ↦ O linearly (r2 ≥ 0.96), departing only in the hydrophobic and hydrophilic tails. Although the model receives no chemical information beyond a single scalar, the functional groups shift systematically with the imposed free energy, from branched hydrocarbons at the apolar end to amides, imides, and isocyanates at the polar end. A bead combination flagged by a CG screen can therefore be turned into candidate molecules for atomistic study or synthesis.
Related
- VIBE-QC: an open-source, general-purpose electronic-structure code for molecules and solids developed with generative AI
- Sampling a trillion-sized GDB-20 database of drug-like molecules by generative artificial intelligence.
- Applicability-Domain-Aware Evaluation of Machine-Learning NMR Chemical-Shift Prediction in Small drug Molecules: GT-NMR for 13C and CASCADE for 1H
- GenGLem as a generative framework for capturing chemical cliffs in energetic materials
- QALPA: Property-guided diffusion modeling for efficient exploration of chemical spaces of flexible molecules
BibTeX
@misc{vazquezsalazar2026recovering,
title = {{Recovering molecules from coarse-grained beads: free-energy-conditioned generative backmapping across chemical space}},
author = {Luis Itza Vazquez-Salazar and Tristan Bereau},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.04432},
eprint = {2609.04432},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.04432},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - Recovering molecules from coarse-grained beads: free-energy-conditioned generative backmapping across chemical space
AU - Luis Itza Vazquez-Salazar
AU - Tristan Bereau
PY - 2026
DA - 2026/09/03
PB - arXiv
DO - 10.48550/arxiv.2609.04432
UR - https://arxiv.org/abs/2609.04432
AB - Transferable coarse-grained (CG) force fields compress chemical space: by aggregating atoms into a reduced set of interaction beads, models such as MARTINI reduce the number of distinguishable compounds by roughly three orders of magnitude, making high-throughput screening of thermodynamic properties tractable across soft matter, with drug--membrane permeability as a well-developed example. The compression is lossy and, so far, one-way: a screen returns a combination of beads, with no established route back to the compounds it stands for. Recovering those compounds--compositional backmapping--is a one-to-many inverse map, distinct from the better-studied conformational problem of rebuilding atomic coordinates from a known mapping. Here we formulate compositional backmapping as conditional graph generation by introducing juniper, a discrete denoising diffusion model over molecular graphs conditioned on the octanol--water partition free energy ΔG_W ↦ O, the principal driver of MARTINI bead type assignment and hence a proxy for bead identity. Trained on molecules of up to 9 heavy atoms mapped onto one or two beads, juniper generates molecules that are 93\% valid and 92\% unique for two-bead targets, and whose ΔG_W ↦ O distributions track the target ΔG^CG_W ↦ O linearly (r^2 ≥ 0.96), departing only in the hydrophobic and hydrophilic tails. Although the model receives no chemical information beyond a single scalar, the functional groups shift systematically with the imposed free energy, from branched hydrocarbons at the apolar end to amides, imides, and isocyanates at the polar end. A bead combination flagged by a CG screen can therefore be turned into candidate molecules for atomistic study or synthesis.
ER -