Browse
Submit Data
Databases
API
Help

Dataset Information

0 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

Explaining and avoiding failure modes in goal-directed generation of small molecules.

ABSTRACT: Despite growing interest and success in automated in-silico molecular design, questions remain regarding the ability of goal-directed generation algorithms to perform unbiased exploration of novel chemical spaces. A specific phenomenon has recently been highlighted: goal-directed generation guided with machine learning models produce molecules with high scores according to the optimization model, but low scores according to control models, even when trained on the same data distribution and the same target. In this work, we show that this worrisome behavior is actually due to issues with the predictive models and not the goal-directed generation algorithms. We show that with appropriate predictive models, this issue can be resolved, and molecules generated have high scores according to both the optimization and the control models.

SUBMITTER: Langevin M

PROVIDER: S-EPMC8973583 | biostudies-literature | 2022 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Publications

Explaining and avoiding failure modes in goal-directed generation of small molecules.

Langevin Maxime M Vuilleumier Rodolphe R Bianciotto Marc M

Journal of cheminformatics 20220401 1

Despite growing interest and success in automated in-silico molecular design, questions remain regarding the ability of goal-directed generation algorithms to perform unbiased exploration of novel chemical spaces. A specific phenomenon has recently been highlighted: goal-directed generation guided with machine learning models produce molecules with high scores according to the optimization model, but low scores according to control models, even when trained on the same data distribution and the ...[more]

PMID: 35365218

Similar Datasets

Neural correlates of goal-directed and non-goal-directed movements.

Project description:What are the cortical neural correlates that distinguish goal-directed and non-goal-directed movements? We investigated this question in the monkey frontal eye field (FEF), which is implicated in voluntary control of saccades. Here, we compared FEF activity associated with goal-directed (G) saccades and non-goal-directed (nG) saccades made by the monkey. Although the FEF neurons discharged before these nG saccades, there were three major differences in the neural activity: First, the variability in spike rate across trials decreased only for G saccades. Second, the local field potential beta-band power decreased during G saccades but did not change during nG saccades. Third, the time from saccade direction selection to the saccade onset was significantly longer for G saccades compared with nG saccades. Overall, our results reveal unexpected differences in neural signatures for G versus nG saccades in a brain area that has been implicated selectively in voluntary control. Taken together, these data add critical constraints to the way we think about saccade generation in the brain.

| S-EPMC8017686 | biostudies-literature

Critical care ultrasound goal-directed versus early goal-directed therapy in septic shock.

Project description: Not available

| S-EPMC8724081 | biostudies-literature

Mol-AIR: Molecular Reinforcement Learning with Adaptive Intrinsic Rewards for Goal-Directed Molecular Generation.

Project description:Optimizing techniques for discovering molecular structures with desired properties is crucial in artificial intelligence (AI)-based drug discovery. Combining deep generative models with reinforcement learning has emerged as an effective strategy for generating molecules with specific properties. Despite its potential, this approach is ineffective in exploring the vast chemical space and optimizing particular chemical properties. To overcome these limitations, we present Mol-AIR, a reinforcement learning-based framework using adaptive intrinsic rewards for effective goal-directed molecular generation. Mol-AIR leverages the strengths of both history-based and learning-based intrinsic rewards by exploiting random distillation network and counting-based strategies. In benchmark tests, Mol-AIR demonstrates improved performance over existing approaches in generating molecules having the desired properties, including penalized LogP, QED, and celecoxib similarity, without any prior knowledge. We believe that Mol-AIR represents a significant advancement in drug discovery, offering a more efficient path to discovering novel therapeutics.

| S-EPMC11898073 | biostudies-literature

Generation of host-directed and virus-specific antivirals using targeted protein degradation promoted by small molecules and viral RNA mimics.

Project description:Targeted protein degradation (TPD), as exemplified by proteolysis-targeting chimera (PROTAC), is an emerging drug discovery platform. PROTAC molecules, which typically contain a target protein ligand linked to an E3 ligase ligand, recruit a target protein to the E3 ligase to induce its ubiquitination and degradation. Here, we applied PROTAC approaches to develop broad-spectrum antivirals targeting key host factors for many viruses and virus-specific antivirals targeting unique viral proteins. For host-directed antivirals, we identified a small-molecule degrader, FM-74-103, that elicits selective degradation of human GSPT1, a translation termination factor. FM-74-103-mediated GSPT1 degradation inhibits both RNA and DNA viruses. Among virus-specific antivirals, we developed viral RNA oligonucleotide-based bifunctional molecules (Destroyers). As a proof of principle, RNA mimics of viral promoter sequences were used as heterobifunctional molecules to recruit and target influenza viral polymerase for degradation. This work highlights the broad utility of TPD to rationally design and develop next-generation antivirals.

| S-EPMC10528416 | biostudies-literature

Small molecules facilitate rapid and synchronous iPSC generation.

| S-EPMC4326224 | biostudies-literature

Small molecules facilitate rapid and synchronous iPSC generation

Project description:The reprogramming of somatic cells into induced pluripotent stem cells (iPSCs) upon overexpression of OCT4, KLF4, SOX2 and c-MYC (OKSM) provides a powerful system to interrogate basic mechanisms of cell fate change. However, iPSC formation with standard methods is typically protracted and inefficient, resulting in heterogeneous cell populations. We show that exposure of OKSM-expressing cells to both ascorbic acid and a GSK3-β inhibitor (AGi) facilitates more synchronous and rapid iPSC formation from several mouse cell types. AGi treatment restored the ability of refractory cell populations to yield iPSC colonies, and it attenuated the activation of developmental regulators commonly observed during the reprogramming process. Moreover, AGi supplementation gave rise to chimera-competent iPSCs after as little as 48 h of OKSM expression. Our results offer a simple modification to the reprogramming protocol, facilitating iPSC induction at unparalleled efficiencies and enabling dissection of the underlying mechanisms in more homogeneous cell populations. 18 samples were analyzed in total, samples represent bulk cultures of reprogrammable-MEFs (Rep-MEFs) that express OKSM and were supplemented with ascorbic acid and GSK3i during iPS cell induction. Control samples represent similar bulk cultures of reprogrammable-MEFs that express OKSM but were not supplemented with ascorbic acid and GSK3i during iPS cell induction. GMP-iPSCs were generated using 48 hours of OKSM+AGi induction. Fibroblast-iPSCs were generated using 96 hours of OKSM+AGi induction.

2014-09-23 | E-GEOD-57774 | biostudies-arrayexpress

Goal-directed or goal-misdirected - how should we interpret the literature?

Project description:Goal-directed therapy (GDT) can be a vague term, meaning different things to different people and, depending on the clinical environment, sometimes even different things to the same person. It can refer to perioperative fluid management, clinicians driving oxygen delivery to supramaximal values, early treatment of sepsis in the emergency department, and even to restriction of perioperative crystalloids with the goal of maintaining preadmission body weight. Understandably, strong opinions about GDT vary; some clinicians consider it essential for perioperative care, others completely ineffective in critically ill patients. This commentary aims to further position the excellent review by Lees and colleagues in the context of the critical care and perioperative setting.

| S-EPMC2887118 | biostudies-literature

Generation of host-directed and virus-specific antivirals using targeted protein degradation promoted by small molecules and viral RNA mimics

Project description:Targeted protein degradation (TPD), as exemplified by proteolysis-targeting chimera (PROTAC), is a prominent paradigm in drug discovery. In this study, we applied conventional and novel PROTAC approaches to develop broad-spectrum antivirals (targeting key host factors for many different viruses) and virus-specific antivirals (targeting unique viral proteins). With respect to host-directed antivirals, we identified a small molecule degrader, FM-74-103, with pan-antiviral activities, as shown by inhibiting both RNA and DNA viruses. FM-74-103 elicits the selective degradation of human GSPT1, a translation termination factor, which further shuts down translation initiation. With respect to virus-specific antivirals, we developed viral RNA oligonucleotide-based bifunctional molecules (Destroyers). We provided proof-ofprinciple that RNA mimics of viral promoter sequences can be used as molecular glues to recruit viral polymerase (vPOL) and target it for degradation. Collectively, this study shows that TPD can be exploited to rationally design and develop the next generations of antivirals.

2023-06-15 | PXD042352 | Pride

A Multi-Omics Interpretable Machine Learning Model Reveals Modes of Action of Small Molecules.

Project description:High-throughput screening and gene signature analyses frequently identify lead therapeutic compounds with unknown modes of action (MoAs), and the resulting uncertainties can lead to the failure of clinical trials. We developed an approach for uncovering MoAs through an interpretable machine learning model of transcriptomics, epigenomics, metabolomics, and proteomics. Examining compounds with beneficial effects in models of Huntington's Disease, we found common MoAs for compounds with unrelated structures, connectivity scores, and binding targets. The approach also predicted highly divergent MoAs for two FDA-approved antihistamines. We experimentally validated these effects, demonstrating that one antihistamine activates autophagy, while the other targets bioenergetics. The use of multiple omics was essential, as some MoAs were virtually undetectable in specific assays. Our approach does not require reference compounds or large databases of experimental data in related systems and thus can be applied to the study of agents with uncharacterized MoAs and to rare or understudied diseases.

| S-EPMC6976599 | biostudies-literature

Small molecules facilitate rapid and synchronous iPSC generation

2014-09-23 | GSE57774 | GEO