[Booth Report] Can You Discover Drugs on AWS!? We Visited the "Amazon Bio Discovery" Booth!

Hi there! I'm Kusunoki!

I joined Classmethod as a new graduate, and it's almost been 3 months now.

I attended AWS Summit Japan 2026, and since I'm still fresh from the experience, I'd like to write a blog post about it quickly!

This time, I want to share something that gave me an "I didn't know AWS was doing this too!" moment.

That is "Amazon Bio Discovery".

Actually, I was in the Faculty of Pharmacy in university. So this service really resonated with me!

What is Amazon Bio Discovery

First, let's read the official AWS description.

https://aws.amazon.com/jp/biodiscovery/

Translating the original text (English), it goes like this:

Amazon Bio Discovery is a service that gives researchers direct access to "biological AI models" trained on vast amounts of biological data. These specialized AI models generate and evaluate antibody drug candidates. But simply having access isn't enough. AI agents help select the right models for your research goals, optimize input data, and evaluate candidates, then seamlessly send them to partner labs for synthesis and testing. Test results are automatically returned to the application for analysis and model refinement. This creates a "lab-in-the-loop" experimental cycle where your organization's knowledge accumulates with each iteration.

…Yes. Honestly, I think just reading this alone might leave you wondering what it's all talking about, so let me organize a few things:

• What does making a drug actually mean
• What is AI-driven drug discovery
• How does having this service change things
• Is it actually useful

First of All, What Does It Mean to Make a Drug (Antibody Drug)?

This service is designed for creating a type of drug called an "antibody." Simply put, an antibody is "a protein that sticks precisely to the cause of a disease and defeats it." It's a word you've probably heard in the news about COVID-19.

To turn this into a drug, you need to endlessly repeat the cycle of "design → actually make and test → review results and redesign."

Note that doing design and prediction inside a computer is called in silico, while testing with actual cells or test tubes is called in vitro. Drug discovery is a much broader process, but along the way, candidates are narrowed down by going back and forth between these two approaches. Understanding this will help the following discussion flow more smoothly.

In Recent Years, "AI Drug Discovery" Has Been Advancing

Traditionally, this "design" step involved researchers using their experience and knowledge to think "this shape might bind well," then actually making and testing it... over and over. But even just making and testing one candidate takes a lot of time and money, so the number of candidates that can be tested is limited.

This is where AI comes in. That said, it's not the same as "generative AI" like ChatGPT or image generation — the main players here are machine learning models trained on vast amounts of biological data.

Specifically:

• AI predicts the three-dimensional shape of proteins
  (The famous example is AlphaFold, which won the 2024 Nobel Prize in Chemistry!)
• AI designs large numbers of molecules likely to bind to a target
• Whether those candidates "actually bind" is evaluated computationally (in silico) before synthesis

In other words, "design → evaluation → narrowing down" is run entirely inside a computer before going to the lab, and only the most promising candidates are actually made and tested. This is AI drug discovery.
It means the "design" part of the loop has shifted significantly from human hands to AI.

Amazon Bio Discovery: Before and After

Before

Researchers needed to prepare everything themselves:

• Computing environment … Build the infrastructure to run models and operate them with machine learning expertise
• Model selection expertise … Choose the right models from the many available and combine them
• A place to experiment … Actually make and test designed candidates in a physical lab (wet lab)

The bar was quite high, as you can see.

After

Amazon Bio Discovery bundles all of these scattered, high-barrier elements together.
Mapped against the Before, the correspondence looks like this:

• Computing environment → Already provided
  Infrastructure and 40+ models are available from the start, with no need to build your own.
• Model selection expertise → Supported by AI agents
  It suggests which models are appropriate, where to target binding, based on industry best practices.
• A place to experiment → Can be outsourced
  Even without your own lab, you can place orders directly from the screen through a network of partner research institutions (CRO network). Pricing and turnaround times are visible too.

As a result,

What personally excited me the most was: AI drug discovery, which previously required having a dedicated wet lab, a computational infrastructure, and specialized personnel all in place, might now be accessible to much smaller teams.

For example, a drug discovery startup — even without massive research facilities — could potentially enter this loop and pursue drug discovery with just ideas and expertise! That's the kind of service I felt this was.

And It Has Actually Been Published as a Paper

In fact, this approach has already been published as a paper.

https://www.amazon.science/publications/agent-guided-de-novo-design-of-nanobody-binders-against-a-novel-cancer-target

The paper describes how nanobodies (small antibodies) designed from scratch using this workflow against a novel cancer target were confirmed to actually bind in real experiments.

In other words, it has been properly demonstrated that this actually works. That was personally the most convincing point for me.

And according to AWS officially, the speed of such development has improved dramatically — a process that previously took up to a year with conventional design methods was completed in just a few weeks.

Thoughts

I never really had a biotech image of AWS, but I was surprised to find that such an impressive service — one that has already been published in papers — had already been released. It would be exciting to see research accelerate even further with this.
This was something I wouldn't have known about without attending AWS Summit, so I'm glad I went.

Apparently there's a trial available now, so I'd like to take my time and try it out sometime!

Thank you so much for reading to the end!