UMA-Inverse: Ligand-Conditioned Protein Inverse Folding with a Distogram-Supervised Dense Pair Encoder

William Sobolewski

Home/Archive/arXiv

arXiv · 2607.07866

UMA-Inverse: Ligand-Conditioned Protein Inverse Folding with a Distogram-Supervised Dense Pair Encoder

William Sobolewski

View original source ↗Read source document ↗Export metadata ↓

Abstract

Designing protein sequences that bind specific ligands benefits from an inverse-folding model conditioned on full ligand geometry. We present UMA-Inverse, which replaces the sparse graph backbone of LigandMPNN with a dense pair-representation encoder: a six-block PairMixer (triangle multiplication, no triangle self-attention or sequence track) refines all residue-residue and residue-ligand atom pairs, supervised by an auxiliary distogram objective, and an autoregressive decoder attends over ligand atoms through a learned, position-specific readout of the pair tensor. The model is compact ($\sim$3.3 M parameters). On the LigandMPNN test splits it reaches 56.1%/55.1%/35.3% interface recovery (small-molecule/metal/nucleotide). It trails LigandMPNN, but by less than the published numbers suggest: re-run under our identical protocol, LigandMPNN scores 59.8/64.4/53.3 (vs. published 63.3/77.5/50.5). In a pocket-fixed setting the redesigns are confidently folded and ligand-binding-competent under Boltz-2 cofolding, again modestly behind LigandMPNN. Its distinctive property is representational: the dense encoder propagates ligand identity to residues far beyond the interface, where LigandMPNN's signal decays. We offer UMA-Inverse as a compact baseline for ligand-conditioned inverse folding that trails LigandMPNN in accuracy, together with a characterization of how a dense all-pairs encoder distributes ligand information.

Explore related subjects

q-bio.BM

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

William Sobolewski. 2026-07-08. UMA-Inverse: Ligand-Conditioned Protein Inverse Folding with a Distogram-Supervised Dense Pair Encoder. https://arxiv.org/abs/2607.07866

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Decoding enzyme-substrate interaction topology reveals principles underlying catalytic efficiency and mutational outcomes

The enzyme turnover number (kcat) defines catalytic efficiency and constrains quantitative models of metabolism, yet the molecular determinants governing kcat and its response to mutation remain poorly understood. Measurements are sparse and labor-intensive, and most computational approaches provide numerical predictions without explaining how enzyme-substrate interactions shape catalytic outcomes. A central challenge is therefore to identify the topological principles that determine where mutations act and how their functional outcomes are encoded within the enzyme-substrate interaction network. Here, we show that catalytic efficiency and mutational effects can be interpreted through enzyme-substrate interaction topology. We developed Interkcat, an interpretable bidirectional cross-attention framework that captures reciprocal coordination between protein residues and substrate atoms. Optimized on a unified benchmark, Interkcat achieves state-of-the-art predictive performance (R2 = 0.701). From its learned representations, we derive an Interaction Topology Score (ITS) that identifies sequence regions statistically enriched for mutation-sensitive sites without explicit structural inputs. We further demonstrate that higher-order topological features distinguish opposing mutational outcomes: lethal mutations disrupt coordinated networks, whereas activity-preserving or enhancing mutations retain sparse, globally organized coupling. These findings establish interaction topology as a unifying principle linking enzyme sequence, catalytic efficiency, and evolutionary perturbation.

q-bio.BM↗

arXiv2026

Adapting Boltz-2 with limited experimental activity data improves early enrichment in virtual screening

Virtual screening aims to prioritize active compounds from large chemical libraries within a limited experimental budget. When applying Boltz-2 to virtual screening, a key challenge is how to use limited experimental data from the target assay to improve the prioritization of active compounds. We investigated whether fine-tuning the Boltz-2 affinity heads with a small number of binary activity labels could improve early enrichment of active compounds in hit discovery. We compared fine-tuning with 40-300 labels in a retrospective evaluation on eight MF-PCBA targets. With 300 activity measurements, fine-tuning increased the number of actives in the top 1% by a geometric mean of 1.77-fold across the eight targets and improved average precision (AP) by 2.14-fold relative to the control without fine-tuning. We also investigated whether rescoring a subset of candidates could retain the improvement in hit recovery by reranking only the top-ranked Boltz-2 candidates with the fine-tuned head. Restricting rescoring to approximately 10% of the evaluation set retained hit recovery comparable to full rescoring. These findings show that affinity-head fine-tuning with limited activity labels improves early enrichment with Boltz-2 and that this benefit can be retained when rescoring a restricted set of candidates.

q-bio.BM↗

arXiv2026

SaltyMeta: a curated benchmark and protein language model-informed web tool for salty peptide prediction

Excess sodium intake remains a major public health challenge, while salty and saltiness-enhancing peptides offer a potential route to preserve sensory saltiness in reduced-sodium foods. Machine-learning studies of salty peptides, however, are constrained by small datasets, heterogeneous evidence standards, uncertain negative labels, and sequence similarity leakage. Here we present SaltyMeta, a curated benchmark and web-accessible screening framework for salty or saltiness-enhancing short peptides. The benchmark contains 580 peptides, including 280 positive peptides and 300 negative peptides, all standardized to 2-15 residue one-letter amino-acid sequences. Quality control found no non-standard residues, exact duplicates, or positive-negative overlaps. A similarity-grouped split retained 456 peptides for training and 124 for held-out testing. We evaluated 548 interpretable peptide descriptors, frozen ESM2 embeddings at 8M, 35M, and 150M parameter scales, and descriptor-embedding fusion models under grouped cross-validation. The traditional ExtraTrees baseline selected by training-set grouped cross-validation achieved ROC-AUC=0.693 in cross-validation and ROC-AUC=0.704, PR-AUC=0.702, F1=0.626, and MCC=0.304 on the held-out test set. The best initial protein-language-model fusion was traditional descriptors plus ESM2-8M embeddings, with grouped CV ROC-AUC=0.696 and test ROC-AUC=0.700. Advanced optimization using PCA95 dimensionality reduction and ExtraTrees feature-importance filtering yielded a practical ESM2-8M PCA95 top-300 model with test ROC-AUC=0.715 and PR-AUC=0.703, although repeated-CV gains remained modest. SaltyMeta is a transparent prioritization tool, not a sensory validation substitute. We provide benchmark, models, scripts, GitHub, and Streamlit for reproducible screening of food-derived peptides.

q-bio.BM↗