Skip to content

Patient-derived models, made in MOS

MOS start from the lab's own samples: cell lines, PDX, PDO, and biopsies, fresh or frozen. Using a single-cell suspension as input, MOSgen makes up to 280,000 MOS per run.

MOS in a well

MOS in a well.

Tissues used in MOS

MOS have been made from cancer and normal tissue across nine organs, from fresh or frozen samples.

Lung

Non-small cell lung cancer (NSCLC) and normal lung.

About 50 PDOs and A549 cells established in MOS.

Breast

Breast cancer and normal breast.

About 40 PDOs and MCF7 cells established in MOS.

Colon

Colorectal cancer and normal colon.

About 35 PDOs and Caco2 cells established in MOS.

Bone marrow

Multiple myeloma and normal bone marrow.

MM1S and H929 cells established in MOS.

Skin

Melanoma and normal skin.

A375 cells established in MOS.

Liver

Normal liver, including primary human hepatocytes.

Pancreas

Pancreatic cancer (PDAC).

5 PDOs established in MOS.

Ovary

Ovarian cancer.

Prostate

Prostate cancer.

Source: Uniform Bio, Research with MOS.

What goes into a MOS

Each MOS is a miniaturized droplet of matrix with a selectable number of starting cells, set by a simple dilution. Cells proliferate and organize inside every MOS.

Cells in droplets in the MOSgen chip

In the chip

Cells in matrix form droplets, four lanes at a time.

A single MOS with organoids growing inside, imaged by fluorescence

One MOS

A single MOS with organoids growing inside, imaged by fluorescence.

MOS (magenta) containing live patient organoids

Imaged whole

MOS (magenta) containing live patient organoids.

Why MOS instead of hand-plated domes

The matrix in one hand-plated dome makes many MOS, and each MOS is its own data point.

A 10 microliter dome beside one MOS, drawn to scale

Drawn to scale: a 10 µL dome beside one MOS at the standard droplet size. The bar is 1 mm.

Plating
MOS arrive as a suspension and go into plates with standard tips, bulk dispensers, and liquid handlers. There is no dome-by-dome pipetting.
Uniform size
MOS from one run are consistent in size, so differences between wells reflect the biology, not the size of the matrix the organoids grow in.
Reproducibility
In a colorectal organoid model, pIC50 CV was 2.5% or lower over 35 runs, 22 weeks, and 6 operators.
No necrotic core
MOS are small enough that every cell sits close to the medium, so drugs and nutrients reach the center of each MOS.
Immune cell access
TILs and PBMCs move into MOS and engage the cells inside. In Matrigel domes, T cells largely stay at the edge, and a T-cell engager that killed in MOS did not kill in dome organoids.
Sample per data point
Neuroblastoma screens in MOS used 2.5 times fewer cells per well than suspension screens, with concordant results across 44 compounds.

Sources: Gobits R, et al. JCO Precision Oncology, 2026 (reproducibility). Ding S, et al. Cell Stem Cell, 2022 (immune cell access). Schoonbeek MC, et al. EJC Paediatric Oncology, 2026 (sample per data point). Other rows: Uniform Bio data.

Bring your samples

Tell us your sample types and the assays you run, and we will plan an evaluation of MOSgen with you.

Request an evaluation