Verge Labs Launches as a Frontier AI Lab for Human Disease Biology

Formerly Verge Genomics, the company is training frontier models on one of the largest proprietary multimodal CNS patient datasets ever assembled – more than 12,000 human brain samples – to predict how individual patients will respond to therapy. Four senior hires join from Altos Labs, Calico, PostEra, and Flatiron.

SAN FRANCISCO, May 27, 2026 /PRNewswire/ -- Verge Labs, the company formerly known as Verge Genomics, today launched as a frontier AI lab building foundation models of human disease biology on a decade of proprietary brain data. Four senior hires join across AI, business, product, and computational biology.

Verge's new direction

Nine out of ten drug trials fail, most often because the wrong target is tested in the wrong patients. Verge Labs is building what AI researchers call a world model of human disease biology: a generative model of how disease unfolds inside a patient, trained directly on human tissue rather than animal or cellular proxies.

Brain tissue is the molecular ground truth for neurological diseases. VergeDB, the company's proprietary dataset, includes more than 12,000 brain transcriptomes across 6,000 patients, 15 million single-cell profiles, and matched proteomic, genomic, and clinical data.

The company applies modern generative AI architectures to this dataset, paired with blood, CSF, imaging, and clinical data from living patients, to produce what it calls a virtual biopsy: a simulated molecular picture of a patient's brain, generated from a simple blood draw. From there, the platform is designed to match patients to therapies, predict disease trajectory, and forecast how biomarkers will respond to a given intervention before a trial is run.

"For a decade, Verge has been building the human tissue data that our field has been missing," said Alice Zhang, co-founder and Chief Executive Officer of Verge Labs. "Breakthroughs in AI architecture mean that, for the first time, Verge's dataset can be used to build models that reason about individual patients' brain biology from measurements accessible in living people. The next decade of neuroscience drug development will look like precision oncology, and we are building the platform to take it there."

Across a decade of programs, targets surfaced by the platform have validated at 83 percent against downstream experimental confirmation. The company's first AI-discovered asset completed a Phase 1b in late 2025. The trial's clinical, biomarker, and patient data have now been folded into VergeDB and form a core training signal for their models. Verge Labs is publishing a separate analysis of its findings from the trial today.

Select drug developers are working with Verge Labs as partners on the platform, with current engagements focused on CNS.

Senior hires

The company also announced four senior hires as it prepares to launch the first of its models:

• Carla Leibowitz, Chief Business Officer. Most recently Vice President of Business Development at Flatiron Health, where she helped launch the company's analytics business, now a substantial share of Flatiron revenue. Previously Chief Business Development Officer at Paige AI, where she oversaw the global launch of the first FDA-cleared AI product in pathology, and Head of Strategy and Business Development at Arterys, the first company to bring a commercial AI radiology platform to market. She holds an MBA from Stanford University and leads Verge's partnerships and commercial execution.
• Emily Ripka, Ph.D., Vice President of Product and Engineering. Joins from PostEra, an AI medicinal chemistry company, where she built and led the product and engineering organization. At PostEra, she oversaw the enterprise deployment of the company's AI platform at Pfizer, where it is now used by hundreds of scientists. She holds a Ph.D. in Chemistry from Syracuse and leads Verge Labs' technical and product teams.
• Yan Wu, Ph.D., AI Research. Joins from Altos Labs, where he served as Senior ML Research Engineer. At Altos, he built state-of-the-art virtual cell models that won the 2025 Arc Virtual Cell Competition generalist prize and led the open-source release of PerturBench,  the field's standard benchmark for ML models of cellular perturbation response. He holds a Ph.D in Bioengineering from UC San Diego and a B.S.E. in Computer Science, Princeton. Wu leads foundation model development at Verge Labs.
• Sean Hackett, Ph.D., Computational Biology. Joins from Calico Life Sciences, where he served as the Director of Discovery Science. At Calico, he led the development of the causal-network and graph-based models underlying the company's discovery programs and collaborated with Google AI on deep learning approaches to proteomics. He holds a Ph.D. in Quantitative and Computational Biology from Princeton University and leads computational biology at Verge Labs.

"Each of these hires has built successful products from the ground up and pioneered the use of AI in human health," said Zhang. "They joined Verge Labs to break open neuroscience the way oncology was broken open a generation ago: precision therapies matched to individual patients, in diseases where millions are still waiting."

About Verge Labs

Verge Labs is a frontier AI lab for human disease biology. Founded in 2015 and headquartered in San Francisco, the company develops world models of patient biology built on proprietary multimodal human tissue data, and partners with drug developers on target discovery, patient stratification, and biomarker prediction across CNS and beyond. Verge Analytics, Inc. operates publicly as Verge Labs.

Media contact

[email protected]

SOURCE Verge Labs