Our work

Enabling genetic sequencing using an AI-driven, web-based app

Illustration of a DNA helix, a sequencing screen and phone screens
The client

Leading biotechnology company

Our client is a market-leading biotechnology company headquartered in the UK that sells to more than 100 markets across the globe. They have a proven track record of delivering disruptive technologies to the markets that they serve and are focused on automation and increasing ease of use through the provision of more cost effective solutions and an increasing number of products that can be used without internet connectivity. Our clients’ technologies are used by research scientists worldwide as well as to support ‘real-life’ decision making in healthcare, industry and other applied settings.

Key outcomes

We helped a leading biotechnology company to build a new browser-based DNA sequencing application through the use of AI. Working together, we designed and built a solution that has significantly improved customer experience and data privacy whilst simultaneously reducing costs.

  • 1.1 million

    DNA sequencing samples per second achieved on a consumer-grade MacBook Pro.

  • 97% - 99.8%

    Accuracy of machine learning models, depending on the model chosen by the user.

  • Improved

    Customer experience and data privacy significantly, whilst simultaneously reducing costs.

Illustration of a person presenting a plan on a board
The challenge

Tooling that was difficult to use and expensive to support

Our client offers DNA sequencing as a product to their customers. Previously, this was provided through the provision of a sequencing device and application which functioned as a basecaller. A basecaller is a software tool that deciphers the building blocks of DNA and RNA strands, providing a confidence rating for each piece of genetic information.

Whilst the outputs of their current basecaller were technically strong, there were several usability aspects that they wished to improve in order to provide a better customer experience. The current application required a high level of technical proficiency to set up correctly. Due to this technicality, many customers had to pay server farms or cloud providers for additional infrastructure on which they could run it. There was a high customer support requirement from our client, incurring significant time and cost, and customers also needed a high-end PC in order to meet the minimum requirements for a very resource intensive process. Lastly, many customers wished to retain sensitive data, which was not possible with the current application.

As a result, our client approached us to help them design and build a browser-based basecaller application using the latest cutting-edge AI machine learning (ML) browser APIs available so that data inference could be performed using both of their established machine learning models.

Illustration of two people holding up a large lightbulb
The solution

An AI-driven, web-based app

We worked with the client to design and build a highly performant, AI-driven web-based basecaller that functioned as a standalone app. It features a bespoke user interface (UI) that has been designed to deliver the same feature set and user experience (UX) that their existing customers had come to expect.

In addition, the app features advanced inference error handling - the ability to identify and deal with errors and uncertainties - that works across both of the clients’ existing ML models. As well as advanced analytics and benchmarking functionality that enables the client to observe and understand how well the app is performing and how users are interacting with it.

Illustration of bar charts and pie charts

The results

Improved UX, reduced costs, and increased privacy

The end result of the project has been an AI-powered product that better meets the needs of our clients’ customers. This has been achieved through:

Improved user experience: the user-friendly browser based DNA sequencing app is not only simpler and less resource intensive to use, it’s also simpler for customers to set up by themselves. This has eliminated the need for advanced technical proficiency and high-end hardware, widening the potential customer base for the product.

Reduced costs: the move to a browser-based app means that customers are no longer reliant on outsourced server farms or cloud platform providers, ensuring significant savings on their infrastructure costs. What’s more, our client also reduced their costs in the long-term as the ease of use reduces the requirement for high-levels of customer support.

Increased data privacy: the clients’ customers are now able to retain full control over sensitive data without having to send it to servers controlled by either the client themselves or a third party provider. This has led to increased levels of trust which has helped to make our clients’ product a reliable choice for both new and existing customers.

Key technology

Python, React, HTML5, ES6, Chart.js, WebGL

Terms in this story