Strategic Collaboration Advances RNA Foundation Models for Precision Oncology

Bulk RNA sequencing is increasingly used to study tumor biology, but standard analyses often reduce results to gene-level summaries that miss important transcript variants and mutation patterns. Long-read sequencing and advanced AI models are now being explored to capture patient-level differences with greater detail for clinical decision support and trial design. A new collaboration pairs high-fidelity long-read RNA data with RNA foundation models to improve patient-level prediction in oncology.

Blank Bio (San Francisco, CA, USA), an applied artificial intelligence research lab training foundation models for RNA, secured USD 7.2 million in seed financing and entered a strategic collaboration with Pacific Biosciences (PacBio; Menlo, CA, USA). The initiative will generate long‑read bulk RNA‑sequencing datasets to advance patient‑level prediction from tumor transcriptomes and support applications in biomarkers, clinical trial design, and diagnostics. Proceeds will fund continued model development and expanded collaborations with pharmaceutical and diagnostic partners.

The collaboration will produce PacBio HiFi long‑read, bulk RNA sequencing data from up to 100 fresh frozen patient tumor samples spanning multiple cancer indications. Sequencing will be performed at Seattle Children’s Research Institute, where Kinnex RNA libraries are automated on the SPTLabtech firefly+ platform. Blank Bio will use these data to further train and evaluate its RNA foundation models, focusing on applications in which RNA‑level signals could improve patient stratification, biomarker discovery, and clinical interpretation.

Blank Bio is applying its RNA foundation models in three main areas. For predictive biomarkers, the company helps pharmaceutical teams better define trial populations, identify patients most likely to respond, and improve patient selection to increase trial success. For prognostic biomarkers and patient trajectory modeling, its models analyze tumor molecular profiles to forecast disease progression, supporting risk stratification and trial design. In clinical diagnostics, Blank Bio is partnering with leading diagnostic companies to enhance existing RNA-seq tests and improve their sensitivity and specificity.

“Bulk RNA-seq is one of the most clinically accessible and information-rich assays in oncology, but much of its signal is still reduced to simplified gene-level summaries,” said Jonathan Hsu, CEO and Co-Founder, Blank Bio. “Blank Bio was founded to apply foundation models to the full molecular detail contained in each patient’s tumor transcriptome and turn that information into more precise, clinically useful predictions. This financing will support continued model development and partnership expansion, while our collaboration with PacBio will generate the high-resolution long-read RNA data needed to further train and evaluate these models in patient tumor samples.”

Related Links
Blank Bio
PacBio