Scalable deeper graph neural networks for high-performance materials property prediction.
Ontology highlight
ABSTRACT:
SUBMITTER: Omee SS
PROVIDER: S-EPMC9122959 | biostudies-literature | 2022 May
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Graph neural networks (GNNs) have received increasing attention because of their expressive power on topological data, but they are still criticized for their lack of interpretability. To interpret GNN models, explainable artificial intelligence (XAI) methods have been developed. However, these methods are limited to qualitative analyses without quantitative assessments from the real-world datasets due to a lack of ground truths. In this study, we have established five XAI-specific molecular property benchmarks, including two synthetic and three experimental datasets. Through the datasets, we quantitatively assessed six XAI methods on four GNN models and made comparisons with seven medicinal chemists of different experience levels. The results demonstrated that XAI methods could deliver reliable and informative answers for medicinal chemists in identifying the key substructures. Moreover, the identified substructures were shown to complement existing classical fingerprints to improve molecular property predictions, and the improvements increased with the growth of training data.
| S-EPMC9782255 | biostudies-literature
Project description:In this paper, we propose DGCL, a dual-graph neural networks (GNNs)-based contrastive learning (CL) integrated with mixed molecular fingerprints (MFPs) for molecular property prediction. The DGCL-MFP method contains two stages. In the first pretraining stage, we utilize two different GNNs as encoders to construct CL, rather than using the method of generating enhanced graphs as before. Precisely, DGCL aggregates and enhances features of the same molecule by the Graph Isomorphism Network and the Graph Attention Network, with representations extracted from the same molecule serving as positive samples, and others marked as negative ones. In the downstream tasks training stage, features extracted from the two above pretrained graph networks and the meticulously selected MFPs are concated together to predict molecular properties. Our experiments show that DGCL enhances the performance of existing GNNs by achieving or surpassing the state-of-the-art self-supervised learning models on multiple benchmark datasets. Specifically, DGCL increases the average performance of classification tasks by 3.73$\%$ and improves the performance of regression task Lipo by 0.126. Through ablation studies, we validate the impact of network fusion strategies and MFPs on model performance. In addition, DGCL's predictive performance is further enhanced by weighting different molecular features based on the Extended Connectivity Fingerprint. The code and datasets of DGCL will be made publicly available.
| S-EPMC11428321 | biostudies-literature
Project description:The local prediction of fatigue damage within polycrystals in a high-cycle fatigue setting is a long-lasting and challenging task. It requires identifying grains tending to accumulate plastic deformation under cyclic loading. We address this task by transcribing ferritic steel microtexture and damage maps from experiments into a microstructure graph. Here, grains constitute graph nodes connected by edges whenever grains share a common boundary. Fatigue loading causes some grains to develop slip markings, which can evolve into microcracks and lead to failure. This data set enables applying graph neural network variants on the task of binary grain-wise damage classification. The objective is to identify suitable data representations and models with an appropriate inductive bias to learn the underlying damage formation causes. Here, graph convolutional networks yielded the best performance with a balanced accuracy of 0.72 and a F1-score of 0.34, outperforming phenomenological crystal plasticity (+ 68%) and conventional machine learning (+ 17%) models by large margins. Further, we present an interpretability analysis that highlights the grains along with features that are considered important by the graph model for the prediction of fatigue damage initiation, thus demonstrating the potential of such techniques to reveal underlying mechanisms and microstructural driving forces in critical grain ensembles.
| S-EPMC10397301 | biostudies-literature
Project description:Recently, graph neural networks (GNNs) have revolutionized the field of chemical property prediction and achieved state-of-the-art results on benchmark data sets. Compared with the traditional descriptor- and fingerprint-based QSAR models, GNNs can learn task related representations, which completely gets rid of the rules defined by experts. However, due to the lack of useful prior knowledge, the prediction performance and interpretability of the GNNs may be affected. In this study, we introduced a new GNN model called RG-MPNN for chemical property prediction that integrated pharmacophore information hierarchically into message-passing neural network (MPNN) architecture, specifically, in the way of pharmacophore-based reduced-graph (RG) pooling. RG-MPNN absorbed not only the information of atoms and bonds from the atom-level message-passing phase, but also the information of pharmacophores from the RG-level message-passing phase. Our experimental results on eleven benchmark and ten kinase data sets showed that our model consistently matched or outperformed other existing GNN models. Furthermore, we demonstrated that applying pharmacophore-based RG pooling to MPNN architecture can generally help GNN models improve the predictive power. The cluster analysis of RG-MPNN representations and the importance analysis of pharmacophore nodes will help chemists gain insights for hit discovery and lead optimization.
| S-EPMC9351086 | biostudies-literature
Project description:Graph neural networks (GNNs) have been widely used in molecular property prediction, but explaining their black-box predictions is still a challenge. Most existing explanation methods for GNNs in chemistry focus on attributing model predictions to individual nodes, edges or fragments that are not necessarily derived from a chemically meaningful segmentation of molecules. To address this challenge, we propose a method named substructure mask explanation (SME). SME is based on well-established molecular segmentation methods and provides an interpretation that aligns with the understanding of chemists. We apply SME to elucidate how GNNs learn to predict aqueous solubility, genotoxicity, cardiotoxicity and blood-brain barrier permeation for small molecules. SME provides interpretation that is consistent with the understanding of chemists, alerts them to unreliable performance, and guides them in structural optimization for target properties. Hence, we believe that SME empowers chemists to confidently mine structure-activity relationship (SAR) from reliable GNNs through a transparent inspection on how GNNs pick up useful signals when learning from data.
| S-EPMC10160109 | biostudies-literature
Project description:We investigate the potential of graph neural networks for transfer learning and improving molecular property prediction on sparse and expensive to acquire high-fidelity data by leveraging low-fidelity measurements as an inexpensive proxy for a targeted property of interest. This problem arises in discovery processes that rely on screening funnels for trading off the overall costs against throughput and accuracy. Typically, individual stages in these processes are loosely connected and each one generates data at different scale and fidelity. We consider this setup holistically and demonstrate empirically that existing transfer learning techniques for graph neural networks are generally unable to harness the information from multi-fidelity cascades. Here, we propose several effective transfer learning strategies and study them in transductive and inductive settings. Our analysis involves a collection of more than 28 million unique experimental protein-ligand interactions across 37 targets from drug discovery by high-throughput screening and 12 quantum properties from the dataset QMugs. The results indicate that transfer learning can improve the performance on sparse tasks by up to eight times while using an order of magnitude less high-fidelity training data. Moreover, the proposed methods consistently outperform existing transfer learning strategies for graph-structured data on drug discovery and quantum mechanics datasets.
| S-EPMC11258334 | biostudies-literature
Project description:Predictive modeling of toxicity is a crucial step in the drug discovery pipeline. It can help filter out molecules with a high probability of failing in the early stages of de novo drug design. Thus, several machine learning (ML) models have been developed to predict the toxicity of molecules by combining classical ML techniques or deep neural networks with well-known molecular representations such as fingerprints or 2D graphs. But the more natural, accurate representation of molecules is expected to be defined in physical 3D space like in ab initio methods. Recent studies successfully used equivariant graph neural networks (EGNNs) for representation learning based on 3D structures to predict quantum-mechanical properties of molecules. Inspired by this, we investigated the performance of EGNNs to construct reliable ML models for toxicity prediction. We used the equivariant transformer (ET) model in TorchMD-NET for this. Eleven toxicity data sets taken from MoleculeNet, TDCommons, and ToxBenchmark have been considered to evaluate the capability of ET for toxicity prediction. Our results show that ET adequately learns 3D representations of molecules that can successfully correlate with toxicity activity, achieving good accuracies on most data sets comparable to state-of-the-art models. We also test a physicochemical property, namely, the total energy of a molecule, to inform the toxicity prediction with a physical prior. However, our work suggests that these two properties can not be related. We also provide an attention weight analysis for helping to understand the toxicity prediction in 3D space and thus increase the explainability of the ML model. In summary, our findings offer promising insights considering 3D geometry information via EGNNs and provide a straightforward way to integrate molecular conformers into ML-based pipelines for predicting and investigating toxicity prediction in physical space. We expect that in the future, especially for larger, more diverse data sets, EGNNs will be an essential tool in this domain.
| S-EPMC10583285 | biostudies-literature
Project description:Bitter taste is an unpleasant taste modality that affects food consumption. Bitter peptides are generated during enzymatic processes that produce functional, bioactive protein hydrolysates or during the aging process of fermented products such as cheese, soybean protein, and wine. Understanding the underlying peptide sequences responsible for bitter taste can pave the way for more efficient identification of these peptides. This paper presents BitterPep-GCN, a feature-agnostic graph convolution network for bitter peptide prediction. The graph-based model learns the embedding of amino acids in the bitter peptide sequences and uses mixed pooling for bitter classification. BitterPep-GCN was benchmarked using BTP640, a publicly available bitter peptide dataset. The latent peptide embeddings generated by the trained model were used to analyze the activity of sequence motifs responsible for the bitter taste of the peptides. Particularly, we calculated the activity for individual amino acids and dipeptide, tripeptide, and tetrapeptide sequence motifs present in the peptides. Our analyses pinpoint specific amino acids, such as F, G, P, and R, as well as sequence motifs, notably tripeptide and tetrapeptide motifs containing FF, as key bitter signatures in peptides. This work not only provides a new predictor of bitter taste for a more efficient identification of bitter peptides in various food products but also gives a hint into the molecular basis of bitterness.Scientific ContributionOur work provides the first application of Graph Neural Networks for the prediction of peptide bitter taste. The best-developed model, BitterPep-GCN, learns the embedding of amino acids in the bitter peptide sequences and uses mixed pooling for bitter classification. The embeddings were used to analyze the sequence motifs responsible for the bitter taste.
| S-EPMC11459932 | biostudies-literature
Project description:High-throughput sequencing technologies have generated massive protein sequences, but the annotations of protein sequences highly rely on the low-throughput and expensive biological experiments. Therefore, accurate and fast computational alternatives are needed to infer functional knowledge from protein sequences. The gene ontology (GO) directed acyclic graph (DAG) contains the hierarchical relationships between GO terms but is hard to be integrated into machine learning algorithms for functional predictions. We developed a deep learning system named PANDA2 to predict protein functions, which used the cutting-edge graph neural network to model the topology of the GO DAG and integrated the features generated by transformer protein language models. Compared with the top 10 methods in CAFA3, PANDA2 ranked first in cellular component ontology (CCO), tied first in biological process ontology (BPO) but had a higher coverage rate, and second in molecular function ontology (MFO). Compared with other recently-developed cutting-edge predictors DeepGOPlus, GOLabeler, and DeepText2GO, and benchmarked on another independent dataset, PANDA2 ranked first in CCO, first in BPO, and second in MFO. PANDA2 can be freely accessed from http://dna.cs.miami.edu/PANDA2/.
| S-EPMC8808544 | biostudies-literature
Project description:Machine learning has advanced the rapid prediction of inorganic materials properties, yet data scarcity for specific properties and capturing thermodynamic stability remains challenging. We propose a framework utilizing a Graph Neural Network with composition-based and crystal structure-based architectures, combined with a transfer learning scheme. This approach accurately predicts energy-related properties (e.g., total energy, energy above the convex hull, energy band gap) and data-scarce mechanical properties (e.g., bulk and shear modulus). Our model incorporates four-body interactions, capturing periodicity and structural characteristics. It outperforms state-of-the-art models in 8 materials property regression tasks. Also, this model predicts local atomic environments and global structural features better than several models. Transfer learning addresses mechanical property data scarcity, while separate architecture analysis allows application to materials lacking crystal structure information. Our framework's interpretability aids in understanding elemental contributions, enhancing material design and discovery. Continuous advancements promise further performance improvements, driving efficient and accurate materials property prediction.
| S-EPMC11742729 | biostudies-literature