Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher.
Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?
Some links on this page may take you to non-federal websites. Their policies may differ from this site.
-
Protein structure is central to biological function, and enabling multimodal protein models requires joint reasoning over sequence, structure, and function. A key barrier is the lack of principled protein structure tokenizers (PSTs): existing approaches fix token size or rely on continuous vector codebooks, limiting interpretability, multi-scale control, and transfer across architectures. We introduce GeoBPE, a geometry-grounded PST that transforms continuous, noisy, multi-scale backbone conformations into discrete ``sentences'' of geometry while enforcing global constraints. Analogous to byte-pair encoding, GeoBPE generates a hierarchical vocabulary of geometric primitives by iteratively (i) clustering Geo-Pair occurrences with k-medoids to yield a resolution-controllable vocabulary; (ii) quantizing each Geo-Pair to its closest medoid prototype; and (iii) reducing drift through differentiable inverse kinematics that optimizes boundary glue angles under an SE(3) end-frame loss. GeoBPE offers compression (>10x reduction in bits-per-residue at similar distortion rate), data efficiency (>10x less training data), and generalization (maintains test/train distortion ratio of 1.0−1.1). It is architecture-agnostic: (a) its hierarchical vocabulary provides a strong inductive bias for coarsening residue-level embeddings from large PLMs into motif- and protein-level representations, consistently outperforming leading PSTs across 12 tasks and 24 test splits; (b) paired with a transformer, GeoBPE supports unconditional backbone generation via language modeling; and (c) tokens align with CATH functional families and support expert-interpretable case studies, offering functional meaning absent in prior PSTs.more » « lessFree, publicly-accessible full text available May 22, 2027
-
Graphene oxide/polymer composite water filtration membranes were developed via coalescence of graphene oxide (GO) stabilized Pickering emulsions around a porosity-generating polymer. Triptycene poly(ether ether sulfone)-CH2NH2:HCl polymer interacts with the GO at the water−oil interface, resulting in stable Pickering emulsions. When they are deposited and dried on polytetrafluoroethylene substrate, the emulsions fuse to form a continuous GO/polymer composite membrane. X-ray diffraction and scanning electron microscopy demonstrate that the intersheet spacing and thickness of the membranes increased with increasing polymer concentration, confirming the polymer as the spacer between the GO sheets. The water filtration capability of the composite membranes was tested by removing Rose Bengal from water, mimicking separations of weak black liquor waste. The composite membrane achieved 65% rejection and 2500 g m−2 h−1 bar−1. With high polymer and GO loading, composite membranes give superior rejection and permeance performance when compared with a GO membrane. This methodology for fabrication membranes via GO/polymer Pickering emulsions produces membranes with a homogeneous morphology and robust chemical separation strength.more » « less
An official website of the United States government
