Cell morphology-guided <i>de novo</i> hit design by conditioning GANs on phenotypic image features

Zapata, Paula Andrea Marin; Méndez‐Lucio, Oscar; Le, Tuan; Beese, Carsten J; Wichard, Jörg D.; Rouquié, David; Clevert, Djork-Arné

doi:10.1039/d2dd00081d

Cited by 8 publications

(7 citation statements)

References 75 publications

(134 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Next, we use CNN module to provide a representation of the compound that contains voxel and electronic density information in three-dimension space 62 . And, the cGAN module is designed to generate compounds that target PPI interfaces using features from the protein complex interface region to regulate the inputs 63 . The cGAN module consists of a generator, a discriminator, and a conditional network.…”

Section: Resultsmentioning

confidence: 99%

Interface-aware molecular generative framework for protein-protein interaction modulators

Wang,

Mao,

et al. 2023

Preprint

View full text Add to dashboard Cite

Protein-protein interactions (PPIs) play a crucial role in many biochemical processes and biological processes. Recently, many structure-based molecular generative models have been proposed. However, PPI sites and compounds targeting PPIs have distinguished physicochemical properties compared to traditional binding pockets and drugs, it is still a challenging task to generate compounds targeting PPIs by considering PPI complexes or interface hotspot residues. In this work, we propose a specifically molecular generative framework based on PPI interfaces, named GENiPPI. We evaluated the framework and found it can capture the implicit relationship between the PPI interface and the active molecules, and can generate novel compounds that target the PPI interface. Furthermore, the framework is able to generate diverse novel compounds with limited PPI interface inhibitors. The results show that PPI interface-based molecular generative model enriches structure-based molecular generative models and facilitates the design of inhibitors based on PPI structures.

show abstract

Section: Resultsmentioning

confidence: 99%

Interface-aware molecular generative framework for protein-protein interaction modulators

Wang,

Mao,

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Beyond optimizing representations that are hard to biologically-interpret, experiment treatment-based weak supervision may lead to undesired consequences where the representations of treatments that lead to a similar phenotype are pushed away from one another in the latent space because they do not share the same label, possibly even pushing one representation closer to the control ( Supplementary Figure 4A ). In turn, representations with reduced cross-treatment phenotypic similarity may induce errors in downstream analyses, especially in unsupervised interpretation of biological function such as lead optimization [ 53 ] and identifying unknown MoA signatures [ 14 , 19 ]. In contrast, our anomaly-based method is self-supervised, i.e., does not use treatment labels (or any other assumptions on the underlying data) to guide the representation, rather treatment profiles’ latent representations are encoded solely according to their deviation from the control in-distribution ( Supplementary Fig.…”

Section: Discussionmentioning

confidence: 99%

Anomaly detection for high-content image-based phenotypic cell profiling

Shpigler,

Kolet,

Golan

et al. 2024

Preprint

View full text Add to dashboard Cite

High-content image-based phenotypic profiling combines automated microscopy and analysis to identify phenotypic alterations in cell morphology and provide insight into the cell’s physiological state. Classical representations of the phenotypic profile can not capture the full underlying complexity in cell organization, while recent weakly machine-learning based representation-learning methods are hard to biologically interpret. We used the abundance of control wells to learn the in-distribution of control experiments and use it to formulate a self-supervised reconstruction anomaly-based representation that encodes the intricate morphological inter-feature dependencies while preserving the representation interpretability. The performance of our anomaly-based representations was evaluated for downstream tasks with respect to two classical representations across four public Cell Painting datasets. Anomaly-based representations improved reproducibility, Mechanism of Action classification, and complemented classical representations. Unsupervised explainability of autoencoder-based anomalies identified specific inter-feature dependencies causing anomalies. The general concept of anomaly-based representations can be adapted to other applications in cell biology.

show abstract

“…Although the typical downstream applications of morphological profiling have been focused on clustering or classification tasks (section An Overview of Image-Based Profiling Data Analysis ), Zapata et al . proposed to leverage morphological profiles to guide de novo molecular design with GANs [ 124 ]. Compared to using transcriptional profiling for compound de novo design [ 125 ], morphological profiling provides higher throughput with less cost.…”

Section: Novel Applications Of Morphological Profiling In Drug Discoverymentioning

confidence: 99%

Morphological profiling for drug discovery in the era of deep learning

Tang,

Ratnayake,

Seabra

et al. 2024

Briefings in Bioinformatics

View full text Add to dashboard Cite

Morphological profiling is a valuable tool in phenotypic drug discovery. The advent of high-throughput automated imaging has enabled the capturing of a wide range of morphological features of cells or organisms in response to perturbations at the single-cell resolution. Concurrently, significant advances in machine learning and deep learning, especially in computer vision, have led to substantial improvements in analyzing large-scale high-content images at high throughput. These efforts have facilitated understanding of compound mechanism of action, drug repurposing, characterization of cell morphodynamics under perturbation, and ultimately contributing to the development of novel therapeutics. In this review, we provide a comprehensive overview of the recent advances in the field of morphological profiling. We summarize the image profiling analysis workflow, survey a broad spectrum of analysis strategies encompassing feature engineering– and deep learning–based approaches, and introduce publicly available benchmark datasets. We place a particular emphasis on the application of deep learning in this pipeline, covering cell segmentation, image representation learning, and multimodal learning. Additionally, we illuminate the application of morphological profiling in phenotypic drug discovery and highlight potential challenges and opportunities in this field.

show abstract

Cell morphology-guided de novo hit design by conditioning GANs on phenotypic image features

Abstract: Developing novel bioactive molecules is time-consuming, costly and rarely successful. As a mitigation strategy, we utilize, for the first time, cellular morphology to directly guide the de novo design of...

Cited by 8 publications

References 75 publications

Interface-aware molecular generative framework for protein-protein interaction modulators

Interface-aware molecular generative framework for protein-protein interaction modulators

Anomaly detection for high-content image-based phenotypic cell profiling

Morphological profiling for drug discovery in the era of deep learning

Contact Info

Product

Resources

About