An official website of the United States government
Abstract Protein language models (pLMs) trained on a large corpus of protein sequences have shown unprecedented scalability and broad generalizability in a wide range of predictive modeling tasks, but their power has not yet been harnessed for predicting protein–nucleic acid binding sites, critical for characterizing the interactions between proteins and nucleic acids. Here, we present EquiPNAS, a new pLM-informed E(3) equivariant deep graph neural network framework for improved protein–nucleic acid binding site prediction. By combining the strengths of pLM and symmetry-aware deep graph learning, EquiPNAS consistently outperforms the state-of-the-art methods for both protein–DNA and protein–RNA binding site prediction on multiple datasets across a diverse set of predictive modeling scenarios ranging from using experimental input to AlphaFold2 predictions. Our ablation study reveals that the pLM embeddings used in EquiPNAS are sufficiently powerful to dramatically reduce the dependence on the availability of evolutionary information without compromising on accuracy, and that the symmetry-aware nature of the E(3) equivariant graph-based neural architecture offers remarkable robustness and performance resilience. EquiPNAS is freely available at https://github.com/Bhattacharya-Lab/EquiPNAS.
more »
« less
This content will become publicly available on December 18, 2025
Protein-Nucleic Acid Complex Modeling with Frame Averaging Transformer
Nucleic acid-based drugs like aptamers have recently demonstrated great therapeutic potential. However, experimental platforms for aptamer screening are costly, and the scarcity of labeled data presents a challenge for supervised methods to learn protein-aptamer binding. To this end, we develop an unsupervised learning approach based on the predicted pairwise contact map between a protein and a nucleic acid and demonstrate its effectiveness in protein-aptamer binding prediction. Our model is based on FAFormer, a novel equivariant transformer architecture that seamlessly integrates frame averaging (FA) within each transformer block. This integration allows our model to infuse geometric information into node features while preserving the spatial semantics of coordinates, leading to greater expressive power than standard FA models. Our results show that FAFormer outperforms existing equivariant models in contact map prediction across three protein complex datasets, with over 10% relative improvement. Moreover, we curate five real-world protein-aptamer interaction datasets and show that the contact map predicted by FAFormer serves as a strong binding indicator for aptamer screening.
more »
« less
- Award ID(s):
- 2403317
- PAR ID:
- 10601458
- Publisher / Repository:
- Neural Information Processing Systems
- Date Published:
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract MotivationQuality assessment (QA) of predicted protein tertiary structure models plays an important role in ranking and using them. With the recent development of deep learning end-to-end protein structure prediction techniques for generating highly confident tertiary structures for most proteins, it is important to explore corresponding QA strategies to evaluate and select the structural models predicted by them since these models have better quality and different properties than the models predicted by traditional tertiary structure prediction methods. ResultsWe develop EnQA, a novel graph-based 3D-equivariant neural network method that is equivariant to rotation and translation of 3D objects to estimate the accuracy of protein structural models by leveraging the structural features acquired from the state-of-the-art tertiary structure prediction method—AlphaFold2. We train and test the method on both traditional model datasets (e.g. the datasets of the Critical Assessment of Techniques for Protein Structure Prediction) and a new dataset of high-quality structural models predicted only by AlphaFold2 for the proteins whose experimental structures were released recently. Our approach achieves state-of-the-art performance on protein structural models predicted by both traditional protein structure prediction methods and the latest end-to-end deep learning method—AlphaFold2. It performs even better than the model QA scores provided by AlphaFold2 itself. The results illustrate that the 3D-equivariant graph neural network is a promising approach to the evaluation of protein structural models. Integrating AlphaFold2 features with other complementary sequence and structural features is important for improving protein model QA. Availability and implementationThe source code is available at https://github.com/BioinfoMachineLearning/EnQA. Supplementary informationSupplementary data are available at Bioinformatics online.more » « less
-
Abstract MotivationNucleic acid binding proteins (NABPs) play critical roles in various and essential biological processes. Many machine learning-based methods have been developed to predict different types of NABPs. However, most of these studies have limited applications in predicting the types of NABPs for any given protein with unknown functions, due to several factors such as dataset construction, prediction scope and features used for training and testing. In addition, single-stranded DNA binding proteins (DBP) (SSBs) have not been extensively investigated for identifying novel SSBs from proteins with unknown functions. ResultsTo improve prediction accuracy of different types of NABPs for any given protein, we developed hierarchical and multi-class models with machine learning-based methods and a feature extracted from protein language model ESM2. Our results show that by combining the feature from ESM2 and machine learning methods, we can achieve high prediction accuracy up to 95% for each stage in the hierarchical approach, and 85% for overall prediction accuracy from the multi-class approach. More importantly, besides the much improved prediction of other types of NABPs, the models can be used to accurately predict single-stranded DBPs, which is underexplored. Availability and implementationThe datasets and code can be found at https://figshare.com/projects/Prediction_of_nucleic_acid_binding_proteins_using_protein_language_model/211555.more » « less
-
Aptamers have many useful attributes including specific binding to molecular targets. After aptamers are identified, their target binding must be characterized. Fluorescence anisotropy (FA) is one technique that can be used to characterize affinity and to optimize aptamer–target interactions. Efforts to make FA assays more efficient by reducing assay volume and time from mixing to measurement may save time and resources by minimizing consumption of costly reagents. Here, we use thrombin and two thrombin-binding aptamers as a model system to show that plate-based FA experiments can be performed in volumes as low as 2 μL per well with 20 minute incubations with minimal loss in assay precision. We demonstrate that the aptamer–thrombin interaction is best modelled with the Hill equation, indicating cooperative binding. The miniaturization of this assay has implications in drug development, as well as in the efficiency of aptamer selection workflows by allowing for higher throughput aptamer analysis.more » « less
-
null (Ed.)Compound-protein pairs dominate FDA-approved drug-target pairs and the prediction of compound-protein affinity and contact (CPAC) could help accelerate drug discovery. In this study we consider proteins as multi-modal data including 1D amino-acid sequences and (sequence-predicted) 2D residue-pair contact maps. We empirically evaluate the embeddings of the two single modalities in their accuracyand generalizability of CPAC prediction (i.e. structure-free interpretable compound-protein affinity prediction). And we rationalize their performances in both challenges of embedding individual modalities and learning generalizable embedding-label relationship. We further propose two models involving cross-modality protein embedding and establish that the one with cross interaction (thus capturing correlations among modalities) outperforms SOTAs and our single modality models in affinity, contact, and binding-site predictions for proteins never seen in the training set.more » « less
