GBPNet

Aykent, Sarp; Xia, Tian

doi:10.1145/3534678.3539441

Cited by 14 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Baseline comparison methods for this task include a variety of state-of-the-art CNNs, recurrent neural networks (RNNs), GNNs, and ENNs, with additional baselines utilizing explicit protein-ligand interaction information listed in Supplementary Table S2 . Using the same dataset and dataset splits, results for these methods are reported as in Wang et al (2023b ), Aykent and Xia (2022) , and Liu et al (2023) , respectively. Note, however, that due to their lack of official publicly-available PyTorch Geometric ( Fey and Lenssen 2019 ) source code, for this task we include simple PyTorch Geometric reproductions of PaiNN ( Schütt et al 2021 ) and the Equivariant Transformer (ET) ( Thölke and De Fabritiis 2022 ) as additional equivariant GNN and Transformer baselines, respectively.…”

Section: Resultsmentioning

confidence: 99%

“…Baseline comparison methods for this task include a composition of state-of-the-art CNNs, GNNs, and ENNs (including our reproductions of PaiNN and ET), as well as previous statistics-based methods. Using the same dataset and dataset splits, results for these methods are reported as in Aykent and Xia (2022) and Townshend et al (2020) , respectively.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Geometry-complete perceptron networks for 3D molecular graphs

Morehead,

Cheng

2024

Bioinformatics

View full text Add to dashboard Cite

Motivation The field of geometric deep learning has recently had a profound impact on several scientific domains such as protein structure prediction and design, leading to methodological advancements within and outside of the realm of traditional machine learning. Within this spirit, in this work, we introduce GCPNet, a new chirality-aware SE(3)-equivariant graph neural network designed for representation learning of 3D biomolecular graphs. We show that GCPNet, unlike previous representation learning methods for 3D biomolecules, is widely applicable to a variety of invariant or equivariant node-level, edge-level, and graph-level tasks on biomolecular structures while being able to (1) learn important chiral properties of 3D molecules and (2) detect external force fields. Results Across four distinct molecular-geometric tasks, we demonstrate that GCPNet’s predictions (1) for protein–ligand binding affinity achieve a statistically significant correlation of 0.608, more than 5%, greater than current state-of-the-art methods; (2) for protein structure ranking achieve statistically significant target-local and dataset-global correlations of 0.616 and 0.871, respectively; (3) for Newtownian many-body systems modeling achieve a task-averaged mean squared error less than 0.01, more than 15% better than current methods; and (4) for molecular chirality recognition achieve a state-of-the-art prediction accuracy of 98.7%, better than any other machine learning method to date. Availability and implementation The source code, data, and instructions to train new models or reproduce our results are freely available at https://github.com/BioinfoMachineLearning/GCPNet.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Geometry-complete perceptron networks for 3D molecular graphs

Morehead,

Cheng

2024

Bioinformatics

View full text Add to dashboard Cite

show abstract

“…where [Gainza et al 2020[Gainza et al , 2023, dMaSIF [Sverrisson et al 2021], ProteinMPNN [Dauparas et al 2022], GearNet [Zhang et al 2023d], ProNet , PiFold , and CDConv , and equivariant networks, including GVP-GNN and GBPNet [Aykent and Xia 2022]. For protein backbone generation, the methods are grouped in terms of structure representations they use.…”

Section: Overviewmentioning

confidence: 99%

“…As shown in Table 21, most existing methods consider only scalar features, and the feature order is 0. For GVP-GNN and GBPNet [Aykent and Xia 2022], the feature order is 1, as it considers directional vectors as node and edge features. The directional features are used to update the learned features for each node.…”

Section: Datasetsmentioning

confidence: 99%

“…The methods for protein folding can be categorized into two classes, before and after AlphaFold2 [Jumper et al 2021]: (1) two-stage learning: RaptorX-Contact [Wang et al 2017a], AlphaFold1 [Senior et al 2020], trRoseTTA [Du et al 2021], (2) end-to-end learning: AlphaFold2 [Jumper et al 2021], RoseTTAFold [Baek et al 2021], ESMFold [Lin et al 2023a], OpenFold [Ahdritz et al 2022]. The methods for protein representation learning are grouped into the invariant networks, including IEConv [Hermosilla et al 2021], HoloProt [Somnath et al 2021], MaSIF[Gainza et al 2020[Gainza et al , 2023, dMaSIF[Sverrisson et al 2021], ProteinMPNN[Dauparas et al 2022], GearNet[Zhang et al 2023d], ProNet, PiFold, and CDConv, and equivariant networks, including GVP-GNN [Jing et al 2021] and GBPNet[Aykent and Xia 2022]. For protein backbone generation, the methods are grouped in terms of structure representations they use.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Artificial intelligence in recommender systems

Zhang

Jin

2020

Complex Intell. Syst.

217

View full text Add to dashboard Cite

Recommender systems provide personalized service support to users by learning their previous behaviors and predicting their current preferences for particular products. Artificial intelligence (AI), particularly computational intelligence and machine learning methods and algorithms, has been naturally applied in the development of recommender systems to improve prediction accuracy and solve data sparsity and cold start problems. This position paper systematically discusses the basic methodologies and prevailing techniques in recommender systems and how AI can effectively improve the technological development and application of recommender systems. The paper not only reviews cutting-edge theoretical and practical contributions, but also identifies current research issues and indicates new research directions. It carefully surveys various issues related to recommender systems that use AI, and also reviews the improvements made to these systems through the use of such AI approaches as fuzzy techniques, transfer learning, genetic algorithms, evolutionary algorithms, neural networks and deep learning, and active learning. The observations in this paper will directly support researchers and professionals to better understand current developments and new directions in the field of recommender systems using AI.

show abstract

Contextual AI models for single-cell protein biology

Li,

Huang,

Sumathipala

et al. 2024

Nat Methods

View full text Add to dashboard Cite

Understanding protein function and developing molecular therapies require deciphering the cell types in which proteins act as well as the interactions between proteins. However, modeling protein interactions across biological contexts remains challenging for existing algorithms. Here we introduce PINNACLE, a geometric deep learning approach that generates context-aware protein representations. Leveraging a multiorgan single-cell atlas, PINNACLE learns on contextualized protein interaction networks to produce 394,760 protein representations from 156 cell type contexts across 24 tissues. PINNACLE’s embedding space reflects cellular and tissue organization, enabling zero-shot retrieval of the tissue hierarchy. Pretrained protein representations can be adapted for downstream tasks: enhancing 3D structure-based representations for resolving immuno-oncological protein interactions, and investigating drugs’ effects across cell types. PINNACLE outperforms state-of-the-art models in nominating therapeutic targets for rheumatoid arthritis and inflammatory bowel diseases and pinpoints cell type contexts with higher predictive capability than context-free models. PINNACLE’s ability to adjust its outputs on the basis of the context in which it operates paves the way for large-scale context-specific predictions in biology.

show abstract

GBPNet

Cited by 14 publications

References 17 publications

Geometry-complete perceptron networks for 3D molecular graphs

Geometry-complete perceptron networks for 3D molecular graphs

Artificial intelligence in recommender systems

Contextual AI models for single-cell protein biology

Contact Info

Product

Resources

About