Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jul 12, 2026

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

Published on: December 15, 2023

SMDNet: A Self-Training-Aware and Multi-Modal-Adaptive Deep Learning Network for Low-Data Aβ₄₂ Probe Design and

Yanling Wu¹, Feifan Xiang², Menglong Li¹

¹College of Chemistry, Sichuan University, Chengdu 610064, P. R. China.

Journal of Medicinal Chemistry

|July 10, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Targeting MAO-A With Ultrafast Kinetics: Affinity Probes for Precision Imaging of Brain Pathologies and Depression Biomarker Validation.

Advanced healthcare materials·2026

Same author

Evaluation of Salter and Pemberton osteotomies in developmental dysplasia of the hip: the role of the congruity index as a prognostic factor.

Journal of orthopaedic surgery and research·2026

Same author

Comment on Yu et al. Reproductive Lifespan and Reproductive Factors in Relation to Dementia Risk in Postmenopausal Women With Type 2 Diabetes.

Diabetes care·2026

Same author

Motif-Level Graph Learning Enables Interpretable Prediction of Drug-Induced QT Prolongation via Cooperative Substructural Determinants.

International journal of molecular sciences·2026

Same author

The unmeasured role of concomitant medication dynamics and immortal time bias in SGLT2 inhibitors study.

Journal of the Formosan Medical Association = Taiwan yi zhi·2026

Same author

Beyond abscess and inflammatory nodule counts: Reflecting on placebo and pain in the ruxolitinib trial.

Journal of the American Academy of Dermatology·2026

This study introduces SMDNet, a deep learning framework for designing Alzheimer's disease probes. SMDNet accelerates the discovery of effective amyloid-beta (Aβ) imaging agents, improving diagnosis and monitoring.

Area of Science:

Biomedical Imaging
Computational Chemistry
Artificial Intelligence

Background:

Amyloid-beta (Aβ) plaque accumulation is a key indicator of Alzheimer's disease.
Fluorescence imaging aids in Alzheimer's diagnosis, but developing effective Aβ probes is challenging due to complex structure-property relationships and limited data.
Current probe development relies heavily on inefficient trial-and-error methods.

Purpose of the Study:

To develop a deep learning (DL) framework, SMDNet, for efficient, low-data design and optimization of Aβ42 probes.
To improve the accuracy and reduce the experimental burden in developing fluorescent probes for Alzheimer's disease diagnosis.

Main Methods:

Developed SMDNet, a self-training-aware and multimodal-adaptive DL framework integrating molecular graphs, fingerprints, and protein descriptors.

Related Experiment Videos

Last Updated: Jul 12, 2026

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

Published on: December 15, 2023

Employed cross-attention and protein-conditional adaptive layer normalization for multimodal data integration.

Utilized iterative self-training with confidence-aware and distribution-aware sampling to enhance data quality in low-data scenarios.

Main Results:

SMDNet demonstrated strong predictive ability, confirmed by ablation studies, external validation, and generalization analyses.
Interpretability analyses identified chemically significant substructures influencing probe predictions.
SMDNet guided the design of five Thioflavin T (ThT)-derived probe candidates, with TA3 exhibiting excellent binding affinity and high-contrast imaging.

Conclusions:

SMDNet offers a powerful, data-efficient approach for rational design and optimization of Aβ42 fluorescent probes.
The framework's predictive capabilities extend across different fluorescent scaffold classes, including coumarin and naphthalimide derivatives.
SMDNet accelerates the development of novel imaging agents for Alzheimer's disease, potentially improving diagnostic and monitoring capabilities.