Accurately predicting peptide secondary structures remains a challenging task due to the lack of discriminative information in short peptides. In this study, PHAT is proposed, a deep hypergraph learning framework for the prediction of peptide secondary structures and the exploration of downstream tasks. The framework includes a novel interpretable deep hypergraph multi-head attention network that uses residue-based reasoning for structure prediction. The algorithm can incorporate sequential semantic information from large-scale biological corpus and structural semantic information from multi-scale structural segmentation, leading to better accuracy and interpretability even with extremely short peptides. The interpretable models are able to highlight the reasoning of structural feature representations and the classification of secondary substructures. The importance of secondary structures in peptide tertiary structure reconstruction and downstream functional analysis is further demonstrated, highlighting the versatility of our models. To facilitate the use of the model, an online server is established which is accessible via http://inner.wei-group.net/PHAT/. The work is expected to assist in the design of functional peptides and contribute to the advancement of structural biology research.

A bottleneck limiting the practical application of lithium metal anodes is the uncontrolled growth of lithium dendrites caused by gradient distributed Li+ from separators to collectors. Herein, 3D-printed frameworks with...

Leveraging artificial intelligence for automatic retrosynthesis speeds up organic pathway planning in digital laboratories. However, existing deep learning approaches are unexplainable, like "black box" with few insights, notably limiting their applications in real retrosynthesis scenarios. Here, we propose MechRetro, a chemical-mechanism-driven graph learning framework for interpretable retrosynthetic prediction and pathway planning, which learns several retrosynthetic actions to simulate a reverse reaction via elaborate self-adaptive joint learning. By integrating chemical knowledge as prior information, we design a novel Graph Transformer architecture to adaptively learn discriminative and chemically meaningful molecule representations, highlighting the strong capacity in molecule feature representation learning. We demonstrate that MechRetro outperforms the state-of-the-art approaches for retrosynthetic prediction with a large margin on large-scale benchmark datasets. Extending MechRetro to the multi-step retrosynthesis analysis, we identify efficient synthetic routes via an interpretable reasoning mechanism, leading to a better understanding in the realm of knowledgeable synthetic chemists. We also showcase that MechRetro discovers a novel pathway for protokylol, along with energy scores for uncertainty assessment, broadening the applicability for practical scenarios. Overall, we expect MechRetro to provide meaningful insights for high-throughput automated organic synthesis in drug discovery.

Peptides have recently emerged as therapeutic molecules against various diseases, and the secondary structure of peptides is a crucial determinant of their bioactivity. However, accurately predicting peptide secondary structures remains a challenging task due to the lack of peptide sequence data and low prediction efficiency caused by limitations in feature engineering. Therefore, we developed PHAT, a deep learning framework based on a hypergraph multi-head attention network and transfer learning for the prediction of peptide secondary structures. Comparative results demonstrated the outstanding performance and robustness of PHAT. In particular, PHAT automatically learns a set of biologically meaningful knowledge on secondary sub-structures, overcoming the limitations of "black-box" in deep learning-based models and providing good interpretability. Additionally, we demonstrated that the structure information derived from PHAT significantly improved the performance of downstream tasks such as the prediction of peptide toxicity and protein-peptide binding sites. Importantly, we further explored the applicability of PHAT for contact map prediction, which can aid in the reconstruction of peptide 3-D structures, thus highlighting the versatility of our model. To facilitate the use of PHAT, we establish an online server which is accessible via https://server.wei-group.net/PHAT/. We expect our work to assist in the design of functional peptides and contribute to the advancement of structural biology research.

No abstract