I have worked in healthcare and life sciences for the past eighteen years, thirteen years in human health and the past five years in animal health.

Regardless of human or animal, the drug development process is very similar, covering discovery and development, preclinical research, clinical research and approvals. The end-to-end process can take between ten to fifteen years, commonly requiring a significant investment in resources and money.

Therefore, it is fair to state that drug development is considered a very high-risk venture, but if successful, it can deliver world-changing results.

The first phase of this process is arguably the most challenging, where researchers must identify (discover) a specific molecule (e.g., DNA sequence, RNA molecule, protein, metabolite), that plays a role in a disease state and therefore, could be targeted by a drug to produce a therapeutic effect.

Once discovered, researchers search for a compound or compounds that interact with the target molecule and that have the potential to become drug candidates.

A variety of compounds are usually identified, triggering a series of experiments to understand their performance and viability. Typically, this assessment includes factors such as absorption, administration, side effects, and potential interactions. With the completion of these experiments, the most promising compounds are isolated and the preclinical research can begin.

Even from this description, it is clear that the drug discovery and development process is extremely uncertain, and can fail at any point. Therefore, scientists and technologists are continuously looking for new capabilities and/or techniques to improve the chances of success or at minimum, to reduce the time, effort and cost.

Therefore, I thought I would highlight a few companies that are doing interesting things with technology, specifically Artificial Intelligence (AI), to support drug discovery and development.

Insilico Medicine

  • Insilico Medicine uses AI for target identification, drug discovery, and biomarker development. Their end-to-end AI platform integrates various AI models to accelerate drug discovery and development.

Recursion

  • Recursion uses AI to analyze cellular imaging data, uncovering new biological insights and drug candidates. Their approach integrates high-throughput experiments with machine learning.

Benevolent

  • BenevolentAI integrates AI to explore the vast chemical space and identify promising drug candidates. Their platform combines biomedical data with machine learning to discover and develop new treatments.

310 AI

  • 310.ai is building an AI operating system for life sciences, search protein databases, visualise, and run models all in one user-friendly interface.

Exscientia

  • Exscientia focuses on AI-driven drug discovery and design. Their AI platform accelerates the creation of new drug candidates, optimising properties such as potency and selectivity.

Schrödinger

  • Schrödinger combines physics-based computational methods with machine learning to design and optimise new drugs. Their platform is used for both small molecule and biologics drug discovery.

Atomwise

  • Atomwise employs deep learning for structure-based drug design. Their AI technology, AtomNet, is designed to predict the bioactivity of small molecules for drug discovery applications.

Cyclica

  • Cyclica includes an AI platform, Ligand Express, which predicts polypharmacology and off-target effects of drug candidates, aiding in the design of safer and more effective drugs.

Verge Genomics

  • Verge Genomics uses AI to map out disease-relevant gene networks and discover drug targets. Their AI platform helps identify novel treatments for neurodegenerative diseases.

Relay Therapeutics (ZebiAI)

  • Relay Therapeutics uses AI to predict the effects of small molecule drugs on biological targets, helping to identify potential drug candidates faster and more accurately.

Insitro

  • Insitro applies machine learning to biological data to discover and develop new drugs. Their approach combines data generation at scale with cutting-edge machine learning techniques.

There is no guarantee that these companies will succeed and it is likely many will be acquired as they attempt to sustain their high burn rate through experimentation.

However, I believe it is the convergence of traditional and computational science that will deliver the next wave of innovation within healthcare and life sciences.

enterprise architecture ea enterprise_architecture artificial_intelligence ai machine_learning ml generative_ai large_language_model llm natural_language_processing nlp transformer data prompt_engineering life_sciences health healthcare lifeinwork