An official website of the United States government
This content will become publicly available on March 10, 2026
T-ALPHA: A Hierarchical Transformer-Based Deep Neural Network for Protein–Ligand Binding Affinity Prediction with Uncertainty-Aware Self-Learning for Protein-Specific Alignment
- Award ID(s):
- 2124511
- PAR ID:
- 10628292
- Publisher / Repository:
- American Chemical Society
- Date Published:
- Journal Name:
- Journal of Chemical Information and Modeling
- Volume:
- 65
- Issue:
- 5
- ISSN:
- 1549-9596
- Page Range / eLocation ID:
- 2395 to 2415
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
Protein language models trained on evolutionary data have emerged as powerful tools for predictive problems involving protein sequence, structure and function. However, these models overlook decades of research into biophysical factors governing protein function. We propose mutational effect transfer learning (METL), a protein language model framework that unites advanced machine learning and biophysical modeling. Using the METL framework, we pretrain transformer-based neural networks on biophysical simulation data to capture fundamental relationships between protein sequence, structure and energetics. We fine-tune METL on experimental sequence–function data to harness these biophysical signals and apply them when predicting protein properties like thermostability, catalytic activity and fluorescence. METL excels in challenging protein engineering tasks like generalizing from small training sets and position extrapolation, although existing methods that train on evolutionary signals remain powerful for many types of experimental assays. We demonstrate METL’s ability to design functional green fluorescent protein variants when trained on only 64 examples, showcasing the potential of biophysics-based protein language models for protein engineering.more » « less
-
Proteolysis-targeting chimera (PROTAC) has emerged as a groundbreaking therapeutic strategy by hijacking the endogenous ubiquitin proteasome system (UPS) for targeted protein degradation. These heterobifunctional molecules recruit E3 ligases to recognize the protein of interest (POI) and facilitate its ubiquitination, leading to subsequent proteasomal degradation. Compared to conventional protein inhibitors, PROTACs offer a broader range of target degradation and remain effective even against proteins with drug-resistant mutations. Moreover, PROTACs function in a catalytic manner to degrade POIs, allowing for significantly lower administration dosages. In recent years, PROTACs have shown great promise in cancer therapy due to their high efficiency and broad applicability. However, their clinical applications remain challenging due to low bioavailability, limited tumor-targeting ability, and potential side effects. Utilizing nanomedicine for the delivery of PROTACs offers a promising strategy to enhance bioavailability, improve tumor selectivity, and minimize toxicity, thereby advancing their applications in cancer treatment. In this review, we outline the fundamental design principles of PROTACs, summarize the latest progress of nanomedicines from molecular design to drug delivery for improved tumor treatment, introduce PROTAC-based combination therapies and emerging design strategies, and discuss current challenges and future prospects of PROTAC nanomedicines toward clinical translation.more » « less
