Valence: creating a world-leading AI chemistry design company

Written Dec 15, 2021. Wrote blog post for Amplitude Ventures' investment in Valence AI. You can find the original here.

This week, Valence announced the close of a $7M oversubscribed series seed financing. Amplitude is investing to support the company advance its AI-first chemistry design platform, scale its partnerships with leading drug discovery organizations, further expand the team and advance its own internal drug discovery programs.

AI is an increasingly important tool for life sciences and a significant part of our targeted investing area. Valence, who is developing a strong AI platform, is a great example of Amplitude’s approach to creating, building and growing world-leading, innovative therapeutics companies. We’re thrilled to help Daniel Cohen, Therence Bois, and the rest of the Valence team realize their vision of creating the world’s most powerful chemistry engine.

We first met the Valence team several years ago when they had raised their first round of financing and had just begun to build their AI platform. From those early days, it was evident they were on the path to building a leading generative drug design platform. Recent milestones such as the partnership deals with Repare, Servier, and Charles River represented a tipping point in the maturation of the technology and validation that the time was now to accelerate their growth.

Why does AI matter in drug discovery?

Discovering new medicines is difficult and time-consuming, historically taking decades and billions of dollars. If scientists understand the mechanism of the disease, work begins to identify ‘druggable’ targets using existing data points. If the targets are known, scientists can screen thousands of compounds to find one that desirably affects the target. However, many times the drivers of disease are not known and scientists have trouble identifying new compounds.

Many AI-focused drug discovery companies to date have been focused on these problems — either interpreting complex datasets for the prediction of new targets in diseases or sorting through existing chemical structures to predict which compound in physical or computational chemistry libraries might have the best features for development. Both of these are important yet limiting in that they can’t design new chemical matter.

The application of machine learning and artificial intelligence to the drug design process has the potential to make it quicker, cheaper and more effective. While these approaches have typically required millions of data points to train and develop these advanced models, the Valence team has leveraged cutting-edge AI algorithms built upon much smaller data sets. The development of these “lean” generative AI models make them an ideal fit for chemistry design, enabling data and biology scientists to work together in a data loop where 100s or 1000s of data points guide the precise design multiple parameters of a drug compound.

What makes Valence unique?

The field of drug discovery AI is slowly evolving. Historically, companies used generation 1 technology where machine learning approaches are used to score existing compounds and chemical libraries to identify good chemical hits. However, Valence is pioneering generation 2 technology through its generative AI capabilities where new chemical matter can be created based on either existing data or based on information around new targets.

In the last several years, Valence has made significant commercial and R&D progress. Building on research done by the company’s founding team at Mila, the world’s largest deep learning research institute, Valence has developed proprietary AI technologies for low/no data drug design. This is unique from other companies as Valence can create brand new chemical matter, something which has historically been impossible to do. Valence has been able to integrate several unique AI approaches such as few-shot learning, graph-based molecular representation learning, structure and ligand-based de novo design and chemical reaction prediction into a chemistry engine capable of rapid design of new chemical matter against a variety of diseases.

Valence is now deploying its engine across multiple target classes and indications through drug design collaborations with leading biotech and pharmaceutical companies, CROs and academic centers, including Charles River Labratories, Repare Therapeutics and Servier.

Canada can create and foster world-leading AI companies

Valence is another example of how Canada is becoming a hotbed for innovation. Leveraging the exceptional AI talent out of the Mila institute in Montreal, Daniel, Therence and their team have been able to create a world-leading novel chemistry design platform that is rapidly scaling to become a full-stack biotech company.

Valence is building a culture that combines generative AI with novel chemistry design to create a unique company with the ability to change the drug development process from discovery to design.

Becoming a world-leading company is no easy task but with brilliant founders, a unique chemistry design platform and a knowledgeable group of investors, Valence has demonstrated its potential to become something remarkable.

We are thrilled this company is in our portfolio and look forward to translating Valence’s AI chemistry platform into new drugs that can make a huge difference to patient’s lives.

-Amplitude team.

‍