Blank Bio Partners with Pacific Biosciences on RNA Data for AI Model, Raises $7.2M
The startup will use long-read sequencing data from up to 100 patient tumor samples to train models aimed at improving clinical trial patient selection.
Blank Bio, a startup developing artificial intelligence-based models to assist in designing clinical trials, has announced a strategic collaboration with Pacific Biosciences to generate long-read RNA sequencing data for use in training and evaluating its models.
Blank Bio also raised $7.2 million in seed financing, led by Define Ventures, Leonis Capital, Nova Threshold, Ripple Ventures, SignalFire, Y Combinator, and other investors.
“This dataset will give us a richer view of patient tumours, including splice isoforms, mutations, and expression signals, helping our models learn from a more complete picture of the transcriptome,” Blank Bio CEO and Cofounder Jonathan Hsu said in a May 19 LinkedIn post. The company’s models, trained on RNA data, could help predict a patient’s disease progression and response to treatment. “By capturing more of each patient’s biology, our goal is to help clinical teams design smaller, more efficient trials that are more likely to succeed.”
Proceeds from the seed round will fund model development, expanded collaborations with pharmaceutical and diagnostic companies, and new long-read RNA-seq dataset generation.
Founded in 2025, Blank Bio has participated in the Y Combinator startup accelerator. Its cofounders also include Philip Fradkin and Ian Shi, both of whom worked in the University of Toronto lab of Bo Wang, who developed the AI foundation model scGPT and cofounded Xaira Therapeutics. Fradkin and Shi were first authors on a paper published in April in Nature Methods on an RNA-based foundation model called “Orthrus.” Whether the company is commercializing Orthrus or if Wang is involved in Blank Bio isn’t clear; the company did not immediately respond to a request for comment.
Under their collaboration, Blank Bio will generate PacBio HiFi long-read bulk RNA sequencing data from up to 100 fresh-frozen patient tumor samples across multiple cancer indications. Sequencing will be conducted at Seattle Children’s Research Institute. Blank Bio will use the resulting data to train and evaluate its models, with a focus on RNA-level signals that may improve patient stratification, biomarker discovery, and clinical interpretation. Financial and other terms of the deal were not disclosed.
The collaboration aims to elucidate obscure isoform architecture, mutational complexity, and other features of patient-specific tumor biology that are lost in gene counts provided by existing RNA datasets.