Graph neural networks for conditional de novo drug design

Abate, Carlo; Decherchi, Sergio; Cavalli, Andrea

doi:10.1002/wcms.1651

Cited by 8 publications

(10 citation statements)

References 108 publications

(240 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This quantitative definition cannot be easily estimated, yet, at least, it provides a common reference, as this ideal quantity could be the object of an estimation process. While experimentally possibly unfeasible, we expect this process to become more and more accessible thanks to generative learning, 16 neural networks for property prediction, 17 retro‐synthesis prediction 18 and possibly through large language models 19 customization for freedom to operate investigation.…”

Section: Machine Learning Methods For Ligandability Predictionmentioning

confidence: 99%

“…These last two types of representation were used in two methodologies presented in SHREC 2022, 48 a protein–ligand binding site recognition competition. From a more general perspective, graphs have been used to represent molecules to tackle several problems in the last decade: from protein structure and function prediction to drug‐target interaction prediction 16,114 . In the case of proteins they are represented as a graph, in which two nodes (atoms) are connected if their distance is below a certain threshold.…”

Section: Machine Learning Methods For Ligandability Predictionmentioning

confidence: 99%

“…From a more general perspective, graphs have been used to represent molecules to tackle several problems in the last decade: from protein structure and function prediction to drug-target interaction prediction. 16,114 In the case of proteins they are represented as a graph, in which two nodes (atoms) are connected if their distance is below a certain threshold. The graph is then fed to a graph neural network (GNN) (e.g., graph convolutional 115 ), a type of deep learning architecture specifically designed for graphs, to process the information according to the previously defined graph topology.…”

Section: Direct Ligandability Estimationmentioning

confidence: 99%

See 2 more Smart Citations

Ligandability and druggability assessment via machine learning

Palma

Abate

Decherchi

et al. 2023

WIREs Comput Mol Sci

Self Cite

View full text Add to dashboard Cite

Drug discovery is a daunting and failure‐prone task. A critical process in this research field is represented by the biological target and pocket identification steps as they heavily determine the subsequent efforts in selecting a putative ligand, most often a small molecule. Finding “ligandable” pockets, namely protein cavities that may accept a drug‐like binder is instrumental to the more general and drug discovery oriented “druggability” estimation process. While high‐throughput experimental techniques exist to identify putative binding sites other than the orthosteric one, these techniques are relatively expensive and not so commonly available in labs. In this regard, computational means of detecting ligandable pockets are advisable for their inexpensiveness and speed. These methods can become, in principle, particularly predictive when supported by machine learning methodologies that provide the modeling framework. As with any data‐driven effort, the outcome critically depends on the input data, its featurization process and possible associated biases. Also, the machine learning task, (supervised/unsupervised) the learning method, and the possible usage of molecular dynamics data considerably shape the inherent assumptions of the modeling step. Defining a proper quantitative thermodynamic and/or kinetic score (or label) is key to the modeling process; here we revise literature and propose residence time as a novel ideal indicator of ligandability. Interestingly the vast majority of the methods does not keep into consideration kinetics nor thermodynamics when devising predictors.This article is categorized under: Data Science > Artificial Intelligence/Machine Learning Structure and Mechanism > Computational Biochemistry and Biophysics Data Science > Chemoinformatics

show abstract

Section: Machine Learning Methods For Ligandability Predictionmentioning

confidence: 99%

Section: Machine Learning Methods For Ligandability Predictionmentioning

confidence: 99%

Section: Direct Ligandability Estimationmentioning

confidence: 99%

See 1 more Smart Citation

Ligandability and druggability assessment via machine learning

Palma

Abate

Decherchi

et al. 2023

WIREs Comput Mol Sci

Self Cite

View full text Add to dashboard Cite

show abstract

“…Applications of GNNs in drug discovery are rapidly expanding, particularly in conditional de novo drug design. GNNs excel at processing graph-structured data and have played a pivotal role in predicting drugtarget interactions and designing new candidate molecules efficiently [68,69]. The fusion of GNNs with deep learning techniques is revolutionizing graph generation for molecular structures, offering promising applications in drug discovery by optimizing resource utilization and improving the efficiency of generating new bioactive molecules.…”

Section: Generative Graph Neural Network (Gnns)mentioning

confidence: 99%

Unlocking the Future of Drug Development: Generative AI, Digital Twins, and Beyond

Mariam,

Niazi,

Magoola

2024

Preprint

View full text Add to dashboard Cite

This article delves into the intersection of generative AI and digital twins within drug discovery, exploring their synergistic potential to revolutionize pharmaceutical research and development. Through various instances and examples, we illuminate how generative AI algorithms, capable of simulating vast chemical spaces and predicting molecular properties, are increasingly integrated with digital twins of biological systems to expedite drug discovery. By harnessing the power of computational models and machine learning, researchers can design novel compounds tailored to specific targets, optimize drug candidates, and simulate their behavior within virtual biological environments. This paradigm shift offers unprecedented opportunities for accelerating drug development, reducing costs, and, ultimately, improving patient outcomes. As we navigate this rapidly evolving landscape, collaboration between interdisciplinary teams and continued innovation will be paramount in realizing the promise of generative AI and digital twins in advancing drug discovery.

show abstract

“…These molecular strings are then used for model training and subsequent generation of molecules in textual form. Compared to generative methods based on molecular graphs [12], CLMs can learn more complex molecular properties better [8], and generate increasingly larger molecules more efficiently [13,14]. These aspects have made CLMs become one of the de facto approaches for de novo drug design.…”

Section: Introductionmentioning

confidence: 99%

Chemical Language Modeling with Structured State Spaces

Özçelik,

de Ruiter,

Criscuolo

et al. 2024

Preprint

View full text Add to dashboard Cite

Generative deep learning is reshaping drug design. Chemical language models (CLMs) – which generate molecules in the form of molecular strings – bear particular promise for this endeavor. Here, we introduce a recent deep learning architecture, termed Structured State-Space Sequence (S4) model, into de novo drug design. In addition to its unprecedented performance in various fields, S4 has a remarkable capability to capture the global properties of long sequences. This aspect is key for chemical language modeling, where complex molecular properties like bioactivity can 'emerge' from distant positions in the molecular strings. This observation gives rise to the following question: Can S4 advance chemical language modeling for de novo design? To provide an answer, we systematically benchmark S4 with state-of-the-art CLMs on an array of drug discovery tasks, such as the identification of bioactive compounds, and the design of drug-like molecules and natural products. S4 showed a superior capacity to learn complex molecular properties, while at the same time exploring diverse scaffolds. Finally, when applied prospectively to kinase inhibition, S4 designed eight of out ten molecules that were predicted as highly active by molecular dynamics simulations. Taken together, these findings advocate for the introduction of S4 into chemical language modeling -- uncovering its untapped potential in the molecular sciences.

show abstract

Graph neural networks for conditional de novo drug design

Cited by 8 publications

References 108 publications

Ligandability and druggability assessment via machine learning

Ligandability and druggability assessment via machine learning

Unlocking the Future of Drug Development: Generative AI, Digital Twins, and Beyond

Chemical Language Modeling with Structured State Spaces

Contact Info

Product

Resources

About