PacBio and Verily team up to pair long-read genomes with clinical data

PacBio and Verily Health announced a multi-year collaboration on Oct 19 to connect PacBio’s HiFi long-read sequencing with Verily Pre, Verily’s platform for governed health data and AI. The goal is datasets that combine long-read genomic data with real-world clinical data at population scale, for research and for training and running AI models.

Why long reads

Long-read sequencing can pick up complex variants, structural changes, methylation and full-length RNA transcripts that short-read methods have trouble capturing, the companies say. Their release also says that processing those large files and joining them to clinical records is still fragmented and largely custom work today. The deal is meant to connect those steps.

How it will work

Verily will set up Pre Workbench as the secure research environment where HiFi genomic datasets are pooled, analyzed and curated. The companies also plan to use Verily Pre’s trusted agentic environment for model training and inference. PacBio’s HiFi whole-genome sequencing workflows already run end to end inside Verily Pre, which gives the work a starting point. Over time they plan to support larger long-read programs for research groups, biopharma and population genomics projects.

Verily CEO Stephen Gillett named neurology, oncology and rare disease as target areas.

GEN reports that the announcement did not disclose financial terms, specific customers or a timeline for broader rollout. PacBio’s products are labeled for research use only and not for diagnostic procedures, so this work sits in research for now, not in patient care.

Why it matters

Clinical AI models are only as good as the data under them, and this deal targets the step that joins deep sequencing data to patient records rather than the sequencing itself.