Biology subjects

Godinez, W. J.

Publications and source records attributed to Godinez, W. J..

6 recordsLinked to original sources

bioRxiv2024

A diffusion model conditioned on compound bioactivity profiles for predicting high-content images

High-content imaging (HCI) provides a rich snapshot of compound-induced phenotypic outcomes that augment our understanding of compound mechanisms in cellular systems. Generative imaging models for HCI provide a route towards anticipating the phenotypic outcomes of chemical perturbations in silico at unprecedented scale and speed. Here, we developed Profile-Diffusion (pDIFF), a generative method leveraging a latent diffusion model conditioned on in silico bioactivity profiles to predict high-content images displaying the cellular outcomes induced by compound treatment. We trained and evaluated a pDIFF model using high-content images from a Cell Painting assay profiling 3750 molecules with corresponding in silico bioactivity profiles. Using a realistic held-out set, we demonstrate that pDIFF provides improved predictions of phenotypic responses of compounds with low chemical similarity to compounds in the training set compared to generative models trained on chemical fingerprints only. In a virtual hit expansion scenario, pDIFF yielded significantly improved expansion outcomes, thus showcasing the potential of the methodology to speed up and improve the search for novel phenotypically active molecules.

bioinformatics↗

bioRxiv2023

Compound activity prediction with dose-dependent transcriptomic profiles and deep learning

Predicting compound activity in assays is a long-standing challenge in drug discovery. Computational models based on compound-induced gene-expression signatures from a single profiling assay have shown promise towards predicting compound activity in other, seemingly unrelated, assays. Applications of such models include predicting mechanisms-of-action (MoA) for phenotypic hits, identifying off-target activities, and identifying polypharmacologies. Here, we introduce Transcriptomics-to-Activity Transformer (TAT) models that leverage gene-expression profiles observed over compound treatment at multiple concentrations to predict compound activity in other biochemical or cellular assays. We built TAT models based on gene-expression data from a RASL-Seq assay to predict the activity of 2,692 compounds in 262 dose response assays. We obtained useful models for 51% of the assays as determined through a realistic held-out set. Prospectively, we experimentally validated the activity predictions of a TAT model in a malaria inhibition assay. With a 63% hit rate, TAT successfully identified several sub-micromolar malaria inhibitors. Our results thus demonstrate the potential of transcriptomic responses over compound concentration and the TAT modeling framework as a cost-efficient way to identify the bioactivities of promising compounds across many assays.

bioinformatics↗

bioRxiv2021

De-novo generation of novel phenotypically active molecules for Chagas disease from biological signatures using AI-driven generative chemistry

Designing novel molecules with targeted biological activities and optimized physicochemical properties is a challenging endeavor in drug discovery. Recent developments in artificial intelligence have enhanced the early steps of de novo drug design and compound optimization. Herein, we present a generative adversarial network trained to design new chemical matter that satisfies a given biological signature. Our model, called pqsar2cpd, is based on the activity of compounds across multiple assays obtained via pQSAR (profile-quantitative structure-activity relationships). We applied pqsar2cpd to Chagas disease and designed a novel molecule that was experimentally confirmed to inhibit growth of parasites in vitro at low micromolar concentrations. Altogether, this approach bridges chemistry and biology into one single framework for the design of novel molecules with promising biological activity.

pharmacology and toxicology↗

bioRxiv2019

Morphological deconvolution of beta-lactam polyspecificity in E. coli

Beta-lactam antibiotics comprise one of the earliest known classes of antibiotic therapies. These molsecules covalently inhibit enzymes from the family of penicillin-binding proteins, which are essential to the construction of the bacterial cell wall. As a result, beta-lactams have long been known to cause striking changes to cellular morphology. The exact nature of the changes tend to vary by the precise PBPs engaged in the cell since beta-lactams exhibit a range of PBP enzyme specificity. The traditional method for exploring beta-lactam polyspecificity is a gel-based binding assay which is low-throughput and typically run ex situ in cell extracts. Here, we describe a medium-throughput, image-based assay combined with machine learning methods to automatically profile the activity of beta-lactams in E. coli cells. By testing for morphological change across a panel of strains with perturbations to individual PBP enzymes, our approach automatically and quantifiably relates different beta-lactam antibiotics according to their preferences for individual PBPs in cells. We show the potential of our approach for guiding the design of novel inhibitors towards different PBP-binding profiles by recapitulating the activity of two recently-reported PBP inhibitors.

systems biology↗

bioRxiv2018

Unsupervised phenotypic analysis of cellular images with multi-scale convolutional neural networks

Large-scale cellular imaging and phenotyping is a widely adopted strategy for understanding biological systems and chemical perturbations. Quantitative analysis of cellular images for identifying phenotypic changes is a key challenge within this strategy, and has recently seen promising progress with approaches based on deep neural networks. However, studies so far require either pre-segmented images as input or manual phenotype annotations for training, or both. To address these limitations, we have developed an unsupervised approach that exploits the inherent groupings within cellular imaging datasets to define surrogate classes that are used to train a multi-scale convolutional neural network. The trained network takes as input full-resolution microscopy images, and, without the need for segmentation, yields as output feature vectors that support phenotypic profiling. Benchmarked on two diverse benchmark datasets, the proposed approach yields accurate phenotypic predictions as well as compound potency estimates comparable to the state-of-the-art. More importantly, we show that the approach identifies novel cellular phenotypes not included in the manual annotation nor detected by previous studies.\n\nAuthor summaryCellular microscopy images provide detailed information about how cells respond to genetic or chemical treatments, and have been widely and successfully used in basic research and drug discovery. The recent breakthrough of deep learning methods for natural imaging recognition tasks has triggered the development and application of deep learning methods to cellular images to understand how cells change upon perturbation. Although successful, deep learning studies so far either can only take images of individual cells as input or require human experts to label a large amount of images. In this paper, we present an unsupervised deep learning approach that, without any human annotation, analyzes directly full-resolution microscopy images displaying typically hundreds of cells. We apply the approach to two benchmark datasets, and show that the approach identifies novel visual phenotypes not detected by previous studies.

bioinformatics↗

bioRxiv2017

Flat-to-curved transition during clathrin-mediated endocytosis correlates with a change in clathrin-adaptor ratio and is regulated by membrane tension

Although essential for many cellular processes, the sequence of structural and molecular events during clathrin-mediated endocytosis remains elusive. While it was believed that clathrin-coated pits grow with a constant curvature, it was recently suggested that clathrin first assembles to form a flat structure and then bends while maintaining a constant surface area. Here, we combine correlative electron and light microscopy and mathematical modelling to quantify the sequence of ultrastructural rearrangements of the clathrin coat during endocytosis in mammalian cells. We confirm that clathrin-coated structures can initially grow flat and that lattice curvature does not show a direct correlation with clathrin coat assembly. We demonstrate that curvature begins when 70% of the final clathrin content is acquired. We find that this transition is marked by a change in the clathrin to clathrin-adaptor protein AP2 ratio and that membrane tension suppresses this transition. Our results support the model that mammalian cells dynamically regulate the flat-to-curved transition in clathrin-mediated endocytosis by both biochemical and mechanical factors.

cell biology↗