Transforming drug discovery with AI-driven active learning

Approximately 12-14 years and $2.6 billion dollars are needed to create and launch a new drug. The first step to achieving this is the drug discovery stage. It usually takes about four years and, in the case of small molecule therapeutics, necessitates more than 4,000 compounds to be manufactured and screened.

Organizations need a transformational method for drug discovery to speed up the innovation cycle, advance drug candidates to the clinical stage more rapidly, and boost competitiveness.

They must create advanced new therapeutics for rare diseases faster, reduce experimental timelines and costs, and decrease the failure rate of drug candidates in the clinical stages.

Integrated drug discovery workflow

Drug Discovery transformation necessitates an integrated drug discovery workflow that blends in silico and experimental methods (or the “Virtual” and the “Real” (“V+R”)).

BIOVIA has developed an integrated, dedicated solution for Small Molecule Therapeutics Design that incorporates all capabilities for V and R relevant to an advanced discovery organization.

It includes capabilities to design and carry out physical lab experiments, record results gathered, and register the molecules efficiently and seamlessly. This boosts lab productivity and reduces R-cycles. Incorporating in silico approaches like molecular modeling and machine learning (ML) will offer the required insight into intermolecular systems.

Artificial intelligence (AI) helps scientists identify which compounds to synthesize next to help them reach a desired target product profile; while taking into account activity, ADME, anti-target effect, and toxicology profiles. Sharing validated models to the discovery team extends modeling specialists' reach.

The V-cycle — the computerized virtual creation, testing, and selection of unique small molecules — helps to design and create drug candidates with superior efficiency, safety profiles, and lower costs.

To launch AI-driven Drug Design as a practice within a discovery organization, all facets of this process and people involved must use fully integrated applications built on a backbone of collective informatics.

Transforming drug discovery with AI-driven active learning

Figure 1. By combining physical testing in the lab (real) with in silico (virtual) in an iterative way, organizations can speed the drug discovery process by as much as 50%, by 1) reducing the number of active learning cycles and 2) shortening the time for each active learning cycle.  Image Credit: BIOVIA, Dassault Systèmes

Transforming drug discovery with AI-driven active learning

Figure 2. The Small Molecule Therapeutics Design workflow involves a series of activities done in various applications to manage the V+R active learning process.  Image Credit: BIOVIA, Dassault Systèmes

The virtual

Using in silico capabilities, BIOVIA’s solution for Small Molecule Therapeutics Design adds an extra dimension to Drug Discovery by allowing researchers to virtually detect and enhance small therapeutic molecules before and in combination with physical testing.

AI enables medicinal chemists to rapidly produce ideas for new compounds to reach their target product profile quickly. The chemists can employ ML models and physics-based approaches designed and configured by in-house professionals without needing to learn how to build them.

This method helps decrease the number of physical trials needed to find a lead candidate. It also helps researchers boost the number of leads and discover unique, non-obvious molecules via de novo exploration of chemical space.

This process is complex as it necessitates numerous properties in a Target Product Profile (TPP) to be simultaneously optimized. A multi-objective optimization algorithm will balance properties such as off-target selectivity, ADME, on-target activity, developability, safety, and toxicity profiles, and even ease of synthesis to form new molecules with a higher chance of matching the TPP.

Approaches that are invaluable in the generative design process are pharmacophore scoring and docking simulations. These methods can be costly, but using them in a scalable cloud framework makes it possible to apply them to massive virtual chemistry data sets. These 3D approaches are vital for producing better quality molecules.

Molecular modeling and simulation methods enable computational chemists to assist design chemists by engineering and publishing extremely accurate models assembled for their particular areas of interest. These models can be life-cycled and managed in the cloud and need no involvement from IT or software developers.

Transforming drug discovery with AI-driven active learning

Figure 3. (left) A candidate drug docked in a protein’s binding pocket; (right) a map of the intermolecular interactions between the protein and ligand. Image Credit: BIOVIA, Dassault Systèmes

Transforming drug discovery with AI-driven active learning

Figure 4. The system helps you monitor the progress of the multi-objective optimization process, as it is balancing multiple properties that are often competing. Image Credit: BIOVIA, Dassault Systèmes

Transforming drug discovery with AI-driven active learning

Figure 5. A cloud-based electronic lab notebook helps scientists to design, plan, record, and analyze experiments, draw conclusions, and document the outcome. Automated capturing of results, signatures, and countersignatures ensures IP protection. Image Credit: BIOVIA, Dassault Systèmes

The computerized virtual creation and testing of unique small molecules will improve lead molecule design in advance of physical testing, making physical testing more targeted and decreasing drug discovery cycle times and costs.

The real

BIOVIA’s solution for Small Molecule Therapeutics Design assists researchers’ workflows in the lab end-to-end. The researchers can effectively strategize, record, plan, and analyze physical experiments, derive conclusions, and record outcomes through an intuitive user interface.

It supports and records the creation, organization, and utilization of samples and lab materials, offering an instant inventory overview. Integrating data capture and analytics, experiment authoring, compound registration, and review guarantees the safe capture of essential data, documentation, and expert scientific knowledge.

This permits its reuse throughout projects and inside R&D communities and protects intellectual property (IP). Team members can easily detect biological and chemical substances and share the data across and within teams.

A team of researchers can work together and instantly contribute to the same experiment. Overall, it accelerates physical lab experimentation and assists teamwork and knowledge management.

Producing safer, more efficacious therapeutics faster

BIOVIA offers an agile, integrated and robust solution for Small Molecule Therapeutics Design that integrates virtual and real operations in a collaborative cloud environment.

The synergy between in silico techniques for virtual identification and enhancement of compounds and physical experimentation at the bench significantly improves R&D productivity, helping teams to deliver safer, more effective treatments to patients more rapidly than ever before.

With the BIOVIA solution for Small Molecule Therapeutics Design, drug discovery teams can:

  • Collaborate to design lead candidates in a single unified environment
  • Improve the quality and number of lead therapeutic candidates with better safety profiles
  • Decrease physical experimentation with corroborated predictive models and in silico experiments
  • • Democratize high-quality, up-to-date predictive models
  • Boost success rate by engineering molecules with preferred properties

About BIOVIA, Dassault Systèmes

BIOVIA™ provides global, collaborative product lifecycle experiences to transform scientific innovation. Our solutions create an unmatched scientific management environment that can help science-based organizations create and connect biological, chemical and material innovations to improve the way we live.

The industry-leading BIOVIA portfolio integrates the diversity of science, experimental processes and information requirements, end-to-end, across research, development, QA/QC and manufacturing. Capabilities include Scientific Informatics, Molecular Modeling/Simulation, Data Science, Laboratory Informatics, Formulation Design, BioPharma Quality & Compliance and Manufacturing Analytics.

BIOVIA is committed to enhancing and speeding innovation, increasing productivity, improving quality and compliance, reducing costs and accelerating product development for customers in multiple industries.


Sponsored Content Policy: News-Medical.net publishes articles and related content that may be derived from sources where we have existing commercial relationships, provided such content adds value to the core editorial ethos of News-Medical.Net which is to educate and inform site visitors interested in medical research, science, medical devices and treatments.

Last updated: Oct 11, 2022 at 9:45 AM

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    BIOVIA, Dassault Systèmes. (2022, October 11). Transforming drug discovery with AI-driven active learning. News-Medical. Retrieved on February 03, 2023 from https://www.news-medical.net/whitepaper/20221011/Transforming-drug-discovery-with-AI-driven-active-learning.aspx.

  • MLA

    BIOVIA, Dassault Systèmes. "Transforming drug discovery with AI-driven active learning". News-Medical. 03 February 2023. <https://www.news-medical.net/whitepaper/20221011/Transforming-drug-discovery-with-AI-driven-active-learning.aspx>.

  • Chicago

    BIOVIA, Dassault Systèmes. "Transforming drug discovery with AI-driven active learning". News-Medical. https://www.news-medical.net/whitepaper/20221011/Transforming-drug-discovery-with-AI-driven-active-learning.aspx. (accessed February 03, 2023).

  • Harvard

    BIOVIA, Dassault Systèmes. 2022. Transforming drug discovery with AI-driven active learning. News-Medical, viewed 03 February 2023, https://www.news-medical.net/whitepaper/20221011/Transforming-drug-discovery-with-AI-driven-active-learning.aspx.

Other White Papers by this Supplier