From surgical resection to perturbation dataset.
Tissue Collection
Patient-derived tumor resections are collected from partner hospital networks. Tissue arrives with anonymized clinical annotations tumor type, grade, stage, and treatment history.
Precision Slicing
Fresh tissue is processed into precision-cut tumor slices. Unlike cell lines or organoids, PCTSs preserve the intact tumor microenvironment cellular composition, native spatial arrangement, and vasculature.
Perturbation
Slices are allocated into matched arms untreated and perturbed. Compounds screened include checkpoint inhibitors, chemotherapeutic agents, cytokines, and more.
Multimodal Readouts
Numerous slices allow running a variety of panels post perturbation. These include transcriptomic, proteomic, and histological assays.
AI Training and Applications
Foundation models trained on perturbation-response data. Applications include target identification, in-silico drug screening, patient stratification, and biomarker discovery. Each experiment feeds back into the next cycle, informing the next round of data collection and perturbations.
10,000 AI-Designed Regulatory DNA Sequences
AI-designed proximal enhancer-like sequences across three cell types, with predicted activity and 3D structures.
Does Muon Improve Regulatory DNA Learning?
Benchmarking Muon and Adam variants for regulatory DNA modeling.
Introducing Axis
The first AI model that generates regulatory DNA elements and predicts their function.
Dr. Manolis Kellis
Professor of Computer Science, MIT · Broad Institute
Leads the MIT Computational Biology Group, focusing on genomics, epigenomics, and regulatory genomics.
Dr. Nicole Paulk
Founder & CEO, Siren Biotechnology · UCSF
Leading expert in AAV gene therapy, advising Dyno Therapeutics, Astellas, and Metagenomi.
Dr. Rashid Bashir
Dean, Grainger College of Engineering, UIUC
Pioneer in bio-nanotechnology, biosensors, and microfluidics. CZ Biohub Chicago advisory committee.