x

Artificial Intelligence in Drug Discovery

Drug discovery and development is a complex and time-consuming process with a high attrition rate. For decades, target-based drug discovery has focused on immortalized cells to identify and optimize inhibitory or activating molecules. Testing with more complex biological systems typically takes place during late stages in the drug discovery pipeline. Novel in vitro and in silico technologies have great potential to decrease the duration and cost of drug discovery, in which artificial intelligence (AI) is emerging as a new approach. Applying advanced tools early in drug discovery enables timely identification of compounds with suitable therapeutic activities and exclusion of compounds with unsuitable safety profiles, thereby reducing attrition risk.

AI and the drug discovery pipeline

As a consequence of the known chemical space, which comprises >1060 molecules, finding successful new therapeutic entities is one of the most difficult aspects of drug discovery. The drug discovery process has thus far been limited by the lack of advanced technologies to refine compound selection. This can now be addressed by applying AI in various stages of drug discovery. The technology can assist with target identification and drug selection, de novo drug design based on drug-target interaction, and the prediction of a compound’s biological activity, toxicity, and physiochemical properties. AI can thus expedite the discovery of novel hits, or lead compounds, and identify new therapeutic uses for existing drugs.

Optimizing success rates

Existing computational models based on AI are relatively precise in identifying targets or predicting compound effects. This emerging field is growing rapidly and will increase productivity, thus reducing lead times for drug discovery. However, hit and lead compounds emerging through AI pipelines still require testing in physiologically relevant in vitro models, before progressing to regulatory approval and clinical trials. Researchers need to collaborate, combining state-of-the-art laboratory technologies with new software, digital tools, and hybrid scientific workflows to ensure generation of high-quality, robust and well-structured data. Partnering is a way to efficiently acquire these broad capabilities.

Our experienced scientists can utilize hiPSC technology to help validate your in silico results. Contact us to explore the possibilities:

Contact us

 

AI holds great potential for target identification and assessment of the effects of new or existing compounds. At Ncardia, we utilize hiPSC technology to validate your AI-generated predictions and to evaluate preclinical safety and proof-of-efficacy. You can trust us to maximize your project’s physiological relevance and value, and to help mitigate attrition risks when progressing along the drug discovery and development pipeline.

Validation of targets identified through AI

Potential targets identified in silico need to be validated in physiologically relevant in vitro models. hiPSC-derived models are highly suitable for target validation, as they can provide scalable and predictive solutions to assess compound efficacy and safety at high throughput. In this application note, we describe how we use hiPSC technology and collaborate with AI companies for target validation.

Assays

Drug Discovery assays

In every phase of drug discovery - whether it is target identification, lead finding, or safety assessment - a reliable assay is the key to success. We work together with you to accelerate your research and successfully bring your project to the next stage - by meeting your milestones and ensuring compliant, on-time assay results through several stages of your drug discovery process.

Assay  

Multi-Electrode Array (MEA) Analysis

Assay  

Contractility

Assay  

High Content Imaging

Assay  

Metabolism Assays

Assay  

Calcium signaling

Assay  

Biomarker Detection

Assay  

Angiogenesis

Assay  

Endothelial Permeability