Bystro AI, a Boston-based genomics platform, is making genetic data searchable through conversational AI, allowing researchers and clinicians to query DNA information as easily as they would search the web. The company transforms how scientists interact with massive genomic datasets by eliminating the need for specialized computational training, addressing a critical bottleneck that has plagued biological research for years.

What's the Real Problem With Genomic Data Today?

Modern DNA sequencing generates enormous amounts of information, but most scientists lack the computational expertise to analyze it efficiently. According to founder Alex Kotlar, PhD, this disconnect between data generation and data interpretation has created a massive inefficiency in research.

"When I entered my PhD program, I expected that scientific research was a very efficient process. I imagined brilliant scientists asking questions, running experiments, analyzing data, and discovering answers. What I found instead was a massive bottleneck in data analysis," explained Kotlar.

Alex Kotlar, Founder and PhD, Bystro AI

Kotlar noted that during his entire PhD program in genetics at Emory University, students took only one bioinformatics class. This meant researchers often spent years trying to interpret their datasets, frequently forming massive collaborative teams just to analyze results. The real bottleneck wasn't generating data; it was making that data understandable.

How Does Bystro's Technology Actually Work?

Bystro operates as a natural language search engine for genetics. Instead of writing complex code or statistical commands, researchers can ask questions in plain English, and the system performs the analysis automatically. The platform has evolved significantly since its inception around 2013 or 2014.

The technology has progressed from foundational algorithms for analyzing genetic data to more sophisticated capabilities. Bystro now handles complex genetic interactions, since most diseases are influenced by many small genetic effects acting together rather than a single mutation. More recently, the company connected these systems with large language models (LLMs), which are AI systems trained on vast amounts of text to understand and generate human language, and developed what Kotlar describes as an agentic AI platform.

The AI can now call genetic analysis algorithms automatically, run statistical tests, and generate new analyses depending on the question being asked. In some cases, the system can answer extremely complex biological questions with a level of accuracy that researchers don't expect from AI.

How to Use Conversational AI for Genomic Research

• Natural Language Queries: Researchers can ask questions in plain English without needing to write code or understand complex bioinformatics syntax, making genetic analysis accessible to scientists without specialized computational training.
• Automated Statistical Analysis: The platform automatically runs appropriate statistical tests and identifies disease-causing mutations, eliminating the need for researchers to manually perform these computationally intensive steps.
• Complex Interaction Modeling: The system analyzes how multiple genetic factors work together to influence disease risk, moving beyond single-mutation analysis to capture the polygenic nature of most conditions.

Kotlar reflected on a particularly meaningful moment in the company's development. A scientist who had been involved in the early days of the Whitehead Institute and played a role in forming the Broad Institute emailed to say that Bystro was "special" and that he was using it regularly because it was saving him an enormous amount of time.

Who Is Actually Using This Technology?

Currently, most of Bystro's users are researchers at major scientific institutions. However, the company has also seen meaningful adoption among ordinary people seeking to understand their own genetic health. Some users have discovered genetic predispositions to conditions like gout or hearing loss that explained symptoms they were experiencing, making personal discoveries themselves through the platform.

The company is exploring partnerships with pharmaceutical companies, recognizing that the platform can dramatically speed up drug discovery processes. But Kotlar's personal passion extends beyond institutional research. He wants to make this technology accessible to hundreds of millions of people globally, allowing anyone who wants to understand their genetic health risks, optimize their lifestyle, or conduct their own research to have the tools to do so.

What's the Bigger Vision for AI in Healthcare?

Kotlar believes AI has the potential to democratize scientific knowledge that has historically been concentrated among a small group of experts with funding, institutional support, or specialized training. For centuries, knowledge in fields like medicine and science has remained largely inaccessible to the general public.

"Historically, science has been accessible only to a very small group of people, typically those with funding, institutional support, or specialized training. AI has the potential to change that," noted Kotlar.

Alex Kotlar, Founder, Bystro AI

Some scientists worry that democratizing genetic information could lead to misuse or misunderstanding. But Kotlar argues that people deserve access to information about their own health. When someone is trying to help a family member fight cancer, for example, they often become incredibly knowledgeable about the disease because they have to. With the right tools, they can make meaningful discoveries themselves. AI can empower people to participate directly in scientific discovery rather than remaining passive recipients of expert conclusions.

The market opportunity for this approach is substantial. The personal genetics market, including DNA sequencing services, is already approaching $100 billion, with the longevity market estimated to be even larger. As genomic data becomes more prevalent and AI tools become more sophisticated, the ability to interpret that data accessibly could reshape how both professional researchers and individuals engage with their genetic information.