2026-10-10T21:58:04Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34036394 2026-09-30T17:51:28Z category_1 category_4 category_8 category_21 category_734 category_931 category_132 portal_5 item_type_6 month_year_09_2026

<p>Evaluation indicators.</p> Shouzhen Song (25145501) Hua Shi (7282) Hongfeng Wu (10227425) Dachen Liu (23461465) Yihang Lin (10198502) Nor Ashidi Mat Isa (25145504) Quan Zou (157931) Leyi Wei (3959444) Biophysics Biochemistry Microbiology Biotechnology Biological Sciences not elsewhere classified Information Systems not elsewhere classified Infectious Diseases t6se prediction tasks substantial application potential spatial conformational patterns maintaining stable performance local spatial topology linear functional motifs grained token level developing precise anti module enhances consistency existing leading models effector proteins secreted although existing methods model integrates sequence enable effective cross models associations effector functions based methods attention module virulence functions tokenized self synergistic modeling residue orientations protein sequence negative bacteria modal fusion modal alignment infective strategies geometric spaces fully capture extensive evaluations extensible architecture contrastive learning contextual embeddings <div><p>Accurate prediction of effector proteins secreted by Gram-negative bacteria is important for elucidating bacterial pathogenic mechanisms and developing precise anti-infective strategies. Although existing methods have benefited from the strong sequence feature extraction capacity of pretrained protein language models, reliance on linear sequence information alone often fails to fully capture the three-dimensional conformational signals required for virulence functions. Meanwhile, conventional structure-based methods are limited by the scarcity of experimentally resolved protein structures. To address these challenges, we propose GeoEPred, a multimodal deep learning framework designed for the synergistic modeling of protein sequence and structure to identify Gram-negative bacterial effector proteins. Specifically, the model integrates sequence-contextual embeddings from a pretrained protein language model with three-dimensional structural representations predicted by ESMFold. A feature projection network refines fine-grained sequence signals associated with effector functions, while geometric vector perceptrons characterize inter-residue orientations, distances, and local spatial topology to capture potential structural conformational motifs. To further enable effective cross-modal fusion, we design a cross-modal alignment and feature-tokenized self-attention module. This module enhances consistency between the sequence-semantic and structural-geometric spaces through contrastive learning and models associations between linear functional motifs and spatial conformational patterns at a fine-grained token level. Extensive evaluations on multiple benchmark datasets show that GeoEPred achieves better predictive performance than existing leading models in T3SE, T4SE, and T6SE prediction tasks, while maintaining stable performance in remote homolog recognition scenarios. Moreover, the modular and extensible architecture of GeoEPred demonstrates strong generalization ability and substantial application potential for genome-scale effector protein discovery.</p></div> 2026-09-30T18:01:24Z Text Journal contribution 10.1371/journal.pcbi.1014344.s001 https://figshare.com/articles/journal_contribution/_p_Evaluation_indicators_p_/34036394 CC BY 4.0