PUResNet: prediction of protein-ligand binding sites using deep residual neural network

Kandel, Jeevan; Tayara, Hilal; Chong, Kil To

doi:10.1186/s13321-021-00547-7

Cited by 72 publications

(87 citation statements)

References 28 publications

(39 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some utilize 3D grids to capture the spatial distribution of the properties within molecular conformers, where 3DCNN [72,40] have been the method of choice. Other studies use 3D voxel-based surface representations as inputs to 3DCNN [55,66] for the prediction of protein-ligand binding sites [42]. However, none of those approaches consider a temporal perspective and take advantage of molecular dynamics simulations to describe the joint flexibility of proteins and ligands [36].…”

Section: Related Workmentioning

confidence: 99%

Pre-training of Equivariant Graph Matching Networks with Conformation Flexibility for Drug Binding

Wu¹,

Jiang²,

Jin³

et al. 2022

Preprint

View full text Add to dashboard Cite

The latest biological findings discover that the motionless 'lock-and-key' theory is no longer applicable and the flexibility of both the receptor and ligand plays a significant role in helping understand the principles of the binding affinity prediction. Based on this mechanism, molecular dynamics (MD) simulations have been invented as a useful tool to investigate the dynamical properties of this molecular system. However, the computational expenditure prohibits the growth of reported protein trajectories. To address this insufficiency, we present a novel spatial-temporal pre-training protocol, PretrainMD, to grant the protein encoder the capacity to capture the time-dependent geometric mobility along MD trajectories. Specifically, we introduce two sorts of self-supervised learning tasks: an atom-level denoising generative task and a protein-level snapshot ordering task. We validate the effectiveness of PretrainMD through the PDBbind dataset for both linear-probing and fine-tuning. Extensive experiments show that our PretrainMD exceeds most state-of-the-art methods and achieves comparable performance. More importantly, through visualization we discover that the learned representations by pre-training on MD trajectories without any label from the downstream task follow similar patterns of the magnitude of binding affinities. This strongly aligns with the fact that the motion of the interactions of protein and ligand maintains the key information of their binding. Our work provides a promising perspective of self-supervised pre-training for protein representations with very fine temporal * The corresponding authors.Preprint. Under review.

show abstract

Section: Related Workmentioning

confidence: 99%

Pre-training of Equivariant Graph Matching Networks with Conformation Flexibility for Drug Binding

Wu¹,

Jiang²,

Jin³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…This work and others [e.g., Stepniewska-Dziubinska et al (2020) ] demonstrated that both prediction and other activities, such as segmentation, are beneficial, so one can devise a more complex framework than a pure predictor. Along these lines, PUResNet ( Kandel et al, 2021 ) uses an interesting deep residual (skip connections) decoder/encoder architecture derived from the U-net concept. This work presented both an architecture and a cleanup procedure for the training set.…”

Section: Introductionmentioning

confidence: 99%

Probabilistic Pocket Druggability Prediction via One-Class Learning

et al. 2022

View full text Add to dashboard Cite

The choice of target pocket is a key step in a drug discovery campaign. This step can be supported by in silico druggability prediction. In the literature, druggability prediction is often approached as a two-class classification task that distinguishes between druggable and non-druggable (or less druggable) pockets (or voxels). Apart from obvious cases, however, the non-druggable class is conceptually ambiguous. This is because any pocket (or target) is only non-druggable until a drug is found for it. It is therefore more appropriate to adopt a one-class approach, which uses only unambiguous information, namely, druggable pockets. Here, we propose using the import vector domain description (IVDD) algorithm to support this task. IVDD is a one-class probabilistic kernel machine that we previously introduced. To feed the algorithm, we use customized DrugPred descriptors computed via NanoShaper. Our results demonstrate the feasibility and effectiveness of the approach. In particular, we can remove or mitigate biases chiefly due to the labeling.

show abstract

“…Analyzing the vast extent of available drug and target data in existing databases, emerging and revolutionary computer technologies and deep learning concepts can lower drug development expenses. Currently, neural networks are considered to be relatively beneficial in bioinformatics applications [ 6 , 7 , 8 , 9 , 10 ].…”

Section: Introductionmentioning

confidence: 99%

CSatDTA: Prediction of Drug–Target Binding Affinity Using Convolution Model with Self-Attention

Ghimire

Tayara

Xuan

et al. 2022

IJMS

Self Cite

View full text Add to dashboard Cite

Drug discovery, which aids to identify potential novel treatments, entails a broad range of fields of science, including chemistry, pharmacology, and biology. In the early stages of drug development, predicting drug–target affinity is crucial. The proposed model, the prediction of drug–target affinity using a convolution model with self-attention (CSatDTA), applies convolution-based self-attention mechanisms to the molecular drug and target sequences to predict drug–target affinity (DTA) effectively, unlike previous convolution methods, which exhibit significant limitations related to this aspect. The convolutional neural network (CNN) only works on a particular region of information, excluding comprehensive details. Self-attention, on the other hand, is a relatively recent technique for capturing long-range interactions that has been used primarily in sequence modeling tasks. The results of comparative experiments show that CSatDTA surpasses previous sequence-based or other approaches and has outstanding retention abilities.

show abstract

PUResNet: prediction of protein-ligand binding sites using deep residual neural network

Cited by 72 publications

References 28 publications

Pre-training of Equivariant Graph Matching Networks with Conformation Flexibility for Drug Binding

Pre-training of Equivariant Graph Matching Networks with Conformation Flexibility for Drug Binding

Probabilistic Pocket Druggability Prediction via One-Class Learning

CSatDTA: Prediction of Drug–Target Binding Affinity Using Convolution Model with Self-Attention

Contact Info

Product

Resources

About