Artificial intelligence (AI) technologies hold huge promise for the biotechnology industry, which hopes analysis of massive datasets could unlock incredible breakthroughs in disease treatment. However, in order to fullfil such potential, biotech businesses will need to build databases and models large enough to generate meaningful insight; that is expensive and time-consuming.

Houston-based start-up Biostate AI is one company that thinks it can help solve this problem. It has developed a low-cost way to sequence RNA, the molecule found in all living cells that translates the genetic information contained in DNA into proteins that do the work of those cells. Academics, biotechs, pharmaceutical companies and researchers can all use Biostate to cut the cost of analysing the RNA in the human blood and tissue samples they take and hold. But in return, Biostate gets to retain this data – and add it to the models it is building.

Ultimately, the company’s ambition is to build a computational model that captures the behaviours of every molecule inside a living cell and to predict how each of these molecules will change with health, disease or exposure to drugs. That would unlock major leaps forward in drug development, enabling the design of medicines personalised to each patient.

“We want to build an AI that really understands humans,” explains Ashwin Gopinath, who founded Biostate last year with David Zhang. “That’s how we’ll understand exactly what the best treatment is to give an individual patient for a specific disease at any given time.”

The reality of many diseases – including cancers and auto-immune conditions, Biostate’s areas of focus – is that while we may give them a single name, there are multiple variations; lung cancer, says, comes in many different forms. In addition, people respond to drugs differently – and they may even respond to the same drug differently at different times, depending on what else is going in their bodies.

Developing treatment plans that take account of all these variables to provide personalised medicines could make a dramatic difference to the outcome for each patient. But that requires researchers to build data models that cover every possible eventuality. That’s ultimately what Biostate hopes to achieve. “It’s an endless frontier,” adds Zhang. “The more data you have, the better your outcomes are going to be.”

One early collaboration is particularly dear to the heart of Gopinath, whose interest in personalised medicine grew following his wife’s diagnosis with leukemia. Biostate has agreed a partnership with Weill Cornell Medicine focused on this condition, with the firm leveraging its repository of bone marrow and blood samples. “This agreement reflects Cornell’s commitment to fostering industry partnerships that translate academic research into real-world impact,” explained Lisa Placanica, senior managing director of Cornell University’s Center for Technology Licensing, when the collaboration was announced.

Biostate now hopes to make further progress in the coming months courtesy of a $12 million fundraising that the company is announcing today. The Series A round is led by Accel, with participation from Gaingels, Mana Ventures, InfoEdge Ventures and existing investors Matter Venture Partners, Vision Plus Capital and Catapult Ventures. The round takes the total amount of funding raised by the company to just over $20 million.

It’s a fascinating field where AI technologies are evolving at speed. Last year, for example, saw the Sanger Institute launch the Generative and Synthetic Genomics research programme, with the aim of building foundational datasets and models to engineer biology.

“Our goal is to produce data at scale in a fast and cost-effective way, which can then be used to train predictive and generative models," said Ben Lehrer, head of generative and synthetic genomics at the Wellcome Sanger Institute, in a recent blog about the launch. “By producing large-scale genetic sequences and predicting the impact of genetic changes, gen AI tools can help accelerate our understanding of genome biology.”

Elsewhere, Exact Sciences acquired Genomic Health for $2.8 billion five years ago to build a presence in this market, while firms such as Tempus, Foundation Medicine, Veracyte and NanoString Technologies are all exploring aspects of the field.

It’s a potentially huge and lucrative market. The McKinsey Global Institute has estimated that the technology could generate $60 billion to $110 billion a year in economic value for the pharma and medical-product industries.

"Just as OpenAI used massive datasets to decode language, Biostate is decoding the molecular signals that govern human health," says Shekhar Kirani, a partner at Accel, of its investment in the company. Kirani believes AI heralds “a new era of diagnostics and therapeutics”.