DeepScaffold: A Comprehensive Tool for Scaffold-Based De Novo Drug Discovery Using Deep Learning

Li, Yibo; Hu, Jianxing; Wang, Yanxing; Zhou, Jun; Zhang, Liangren; Liu, Zhenming

doi:10.1021/acs.jcim.9b00727

Cited by 112 publications

(109 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This particularity is noticeable in scaffolds (6), (8), and especially (10) and may point to applicability domain limitations of the APM. Moreover, the absolute number of different molecules decorated varied greatly among scaffolds, and the total number of molecules (26,140) was approximately two times greater than that of the previous section, possibly because the model had less information from the scaffolds and was less focused. Lastly, the overlap between the decoys and the generated molecules was also calculated and yielded higher results to those in the previous section (Additional file 2: Table S3).…”

Section: Generating Molecules From New Scaffoldsmentioning

confidence: 77%

“…Two approaches have been published [25,26] that use Graph Generative Neural Networks instead of SMILES to represent molecules. They train only one model each and use a general drug-like molecular set as training data, which is preprocessed.…”

Section: Comparing Smiles-based and Ggnn Scaffold Decoratorsmentioning

confidence: 99%

“…An approach was published that could generate molecules by completing a SMILES string from both sides using a bidirectional RNN [24], but it is limited to two attachment points. Furthermore, other approaches have been reported that use graph generative neural networks (GGNN), which are even able to decorate a scaffold without the need of specifying attachment points [25,26]. Unfortunately, GGNNs are experimental architectures that require far more resources to train and sample than SMILES-based architectures [20].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SMILES-based deep generative scaffold decorator for de-novo drug design

et al. 2020

View full text Add to dashboard Cite

Molecular generative models trained with small sets of molecules represented as SMILES strings can generate large regions of the chemical space. Unfortunately, due to the sequential nature of SMILES strings, these models are not able to generate molecules given a scaffold (i.e., partially-built molecules with explicit attachment points). Herein we report a new SMILES-based molecular generative architecture that generates molecules from scaffolds and can be trained from any arbitrary molecular set. This approach is possible thanks to a new molecular set pre-processing algorithm that exhaustively slices all possible combinations of acyclic bonds of every molecule, combinatorically obtaining a large number of scaffolds with their respective decorations. Moreover, it serves as a data augmentation technique and can be readily coupled with randomized SMILES to obtain even better results with small sets. Two examples showcasing the potential of the architecture in medicinal and synthetic chemistry are described: First, models were trained with a training set obtained from a small set of Dopamine Receptor D2 (DRD2) active modulators and were able to meaningfully decorate a wide range of scaffolds and obtain molecular series predicted active on DRD2. Second, a larger set of drug-like molecules from ChEMBL was selectively sliced using synthetic chemistry constraints (RECAP rules). In this case, the resulting scaffolds with decorations were filtered only to allow those that included fragment-like decorations. This filtering process allowed models trained with this dataset to selectively decorate diverse scaffolds with fragments that were generally predicted to be synthesizable and attachable to the scaffold using known synthetic approaches. In both cases, the models were already able to decorate molecules using specific knowledge without the need to add it with other techniques, such as reinforcement learning. We envision that this architecture will become a useful addition to the already existent architectures for de novo molecular generation.

show abstract

Section: Generating Molecules From New Scaffoldsmentioning

confidence: 77%

Section: Comparing Smiles-based and Ggnn Scaffold Decoratorsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SMILES-based deep generative scaffold decorator for de-novo drug design

et al. 2020

View full text Add to dashboard Cite

show abstract

“…a partially-built molecule with explicit attachment points) mainly due to the restrictions of the SMILES syntax. Some approaches have been reported that use graph generative neural networks (GGNN) which are even able to decorate a scaffold without the need of specifying attachment points [24,25].…”

Section: Introductionmentioning

confidence: 99%

SMILES-Based Deep Generative Scaffold Decorator for De-Novo Drug Design

Arús‐Pous

Patronov²,

Bjerrum³

et al. 2020

Preprint

View full text Add to dashboard Cite

Molecular generative models trained with small sets of molecules represented as SMILES strings are able to generate large regions of the chemical space. Unfortunately, due to the sequential nature of SMILES strings, these models are not able to generate molecules given a scaffold (i.e. partially-built molecules with explicit attachment points). Herein we report a new SMILES-based molecular generative architecture that generates molecules from scaffolds and can be trained from any arbitrary molecular set. This is possible thanks to a new molecular set pre-processing algorithm that exhaustively cuts all combinations of acyclic bonds of every molecule, obtaining a large number of scaffold-decorations combinations. Moreover, it serves as a data augmentation technique and can be readily coupled with randomized SMILES to obtain even better results with small sets. Two examples showcasing the potential of the architecture in medicinal and synthetic chemistry are described: First, models were trained with a training set obtained from a small set of Dopamine Receptor D2 (DRD2) active modulators and were able to meaningfully decorate a wide range of scaffolds and obtain molecular series predicted active on DRD2. Second, a larger set of drug-like molecules from ChEMBL was selectively sliced using synthetic chemistry constraints (RECAP rules). Moreover, the resulting scaffold-decorations were filtered to only allow decorations that were fragment-like. This allowed models trained with this dataset to selectively decorate diverse scaffolds with fragments that were generally predicted to be synthesizable and attachable to the scaffold using known synthetic approaches. In both cases, the models were already able to decorate molecules using specific knowledge without the need to add it with other techniques, such as reinforcement learning. We envision that this architecture will become a useful addition to the already existent architectures for de-novo molecular generation.

show abstract

“…More recently, it was recognised that various kinds of architectures could in fact permit the reversal of this numerical encoding so as to return a molecule (or its SMILES string encoding a unique structure). These are known as generative methods [41][42][43][44][45][46][47][48][49][50], and at heart their aim to generate a suitable and computationally useful representation [51] of the input data. It is common (but cf.…”

mentioning

confidence: 99%

VAE-Sim: a novel molecular similarity measure based on a variational autoencoder

Samanta

O’Hagan

Swainston

et al. 2020

Preprint

View full text Add to dashboard Cite

Molecular similarity is an elusive but core ‘unsupervised’ cheminformatics concept, yet different ‘fingerprint’ encodings of molecular structures return very different similarity values even when using the same similarity metric. Each encoding may be of value when applied to other problems with objective or target functions, implying that a priori none is ‘better’ than the others, nor than encoding-free metrics such as maximum common substructure (MCSS). We here introduce a novel approach to molecular similarity, in the form of a variational autoencoder (VAE). This learns the joint distribution p(z|x) where z is a latent vector and x are the (same) input/output data. It takes the form of a ‘bowtie’-shaped artificial neural network. In the middle is a ‘bottleneck layer’ or latent vector in which inputs are transformed into, and represented as, a vector of numbers (encoding), with a reverse process (decoding) seeking to return the SMILES string that was the input. We train a VAE on over 6 million druglike molecules and natural products (including over one million in the final holdout set). The VAE vector distances provide a rapid and novel metric for molecular similarity that is both easily and rapidly calculated. We describe the method and its application to a typical similarity problem in cheminformatics.

show abstract

DeepScaffold: A Comprehensive Tool for Scaffold-Based De Novo Drug Discovery Using Deep Learning

Cited by 112 publications

References 38 publications

SMILES-based deep generative scaffold decorator for de-novo drug design

SMILES-based deep generative scaffold decorator for de-novo drug design

SMILES-Based Deep Generative Scaffold Decorator for De-Novo Drug Design

VAE-Sim: a novel molecular similarity measure based on a variational autoencoder

Contact Info

Product

Resources

About