Technology
A new standard for human midbrain organoids.
Made by machines, guided by the experts who built the protocol. No hands in the process, no hands in the data.
01The platform
Human tissue, produced like an instrument reading.
The biology was never the bottleneck. The process was.
PEARL’s automated midbrain organoids (AMOs) are grown from human iPSC-derived small-molecule neural progenitor cells (smNPCs), directed towards midbrain identity by defined small-molecule patterning. None of that is new: it has been in the literature for years. What was missing was a way to run it the same way every time.
The difference is in how they are made. A single-cell suspension is dispensed by liquid handling into SBS 96 V-bottom plates (no Matrigel, no animal-derived matrix), and from there the organoid never leaves its well: patterned, matured, fixed, stained, cleared and imaged in place.
The pre-differentiated starting cells are why it is fast: smNPCs can only become neural tissue, so no maturation time is spent waiting for the culture to decide what it is. And because whole-mount immunostaining and confocal microscopy read the organoid intact, there is no sectioning and no reconstruction from a single plane.
One thing the process does not fix is the starting material. Because the route runs through smNPCs rather than through a single proprietary line, it can be run on any human iPSC line, including yours. If your programme rests on a particular donor background or a particular mutation, that is the line the organoids are made from, and the reproducibility is a property of the process rather than of the cells.
This is proprietary, patented technology as much as it is published science.
<3.6%
Intra-batch CV
In viability, against more than 50% commonly reported for conventional organoids.
20,000
Organoids/day, capacity
What one platform can handle, against roughly 200 per day for one person working by hand.
99.7%
Sample retention
Survive 30 days of culture; 96.5% survive transfer to imaging, 93.9% image acquisition.
30–35 days
To a screenable organoid
Against 60 to 90 days for standard protocols.
02Production workflow
Seeded, patterned and read out without being touched.
The organoid never leaves the well between seeding and readout. That is the whole idea.
Day −1 · Starting cells
Human iPSC-derived smNPCs in 2D culture, dissociated to a single-cell suspension.
Day 0 · Automated seeding
Dispensed into 96-well V-bottom plates by liquid handling. Matrix-free.
Day 2 · Ventral patterning
Defined small molecules and growth factors direct the aggregate towards midbrain identity.
Day 6 · Maturation
The organoid compacts and matures in the well it was seeded into. Growth tracks tightly between batches.
Treatment
Compound addition or other intervention, applied plate-wide at the timepoint the study calls for.
From day 35 · In-plate staining
Automated fixation and immunostaining without transferring the tissue: six days primary antibody, six days secondary.
Day 47 · Clearing
BABB tissue clearing, then whole-organoid image acquisition by high-content confocal microscopy.
Analysis & data
Quantitative image analysis, or other downstream molecular readouts including single-cell sequencing.
03Characterisation
What we measure, and how.
Measured on the intact organoid rather than on a section of it, at whatever depth a question requires, even down to single-cell level.
Viability
Measured per organoid. The spread across a batch is treated as a property of the process, not as noise to be averaged away: below 3.6% intra-batch CV.
Cellular composition
A high density of TH-positive dopaminergic neurons alongside the wider neural population, including glia. Transcriptional correlation against human fetal midbrain reaches 0.78 (Pearson).
Morphology
No necrotic core, and no neural rosettes: the organoids are matured past that stage. Both remove well-known sources of heterogeneity.
Function
Spontaneous, synchronised activity by calcium imaging from around day 30: evidence that the tissue is connected, not just alive.
Single-cell state
Single-cell RNA sequencing resolves which populations are present and what state they are in: the basis for the fetal-midbrain comparison, and available as a study readout.
Quality control
Each batch undergoes a strict, multi-modal release framework. We provide a Certificate of Analysis (CoA) covering morphological, molecular and functional metrics, so you screen on human biology, not batch-to-batch noise.
Which markers we stain for depends on the question; the panel is chosen per study rather than fixed. What does not change is that the tissue is stained and imaged whole: the organoid below is a single intact specimen, cleared and acquired in four channels at once, not a section of one.
04Differentiation
How PEARL compares.
The key differences, at a glance.
| Challenge | Common in conventional systems | PEARL approach |
|---|---|---|
| Variability | Commonly reported above 50% CV | Below 3.6% intra-batch CV in viability |
| Matrix | Undefined animal-derived matrix (Matrigel), adding lot-to-lot variation | Matrix-free formation from a single-cell suspension |
| Time to screen | 60 to 90 days | 30 to 35 days |
| Throughput | Around 200 organoids per day, per person | Around 20,000 organoids per day, per platform capacity |
| Assay quality | Not usually reported | Z′-factor of 0.92 |
| Handling losses | Organoids transferred between vessels to be fixed and stained | Fixation, staining and clearing performed in-plate; 99.7% sample retention through culture |
| Format | Bespoke vessels needing dedicated handling | SBS 96 V-bottom plates on standard liquid handlers |
| Tissue structure | Necrotic cores and neural rosettes are common | Neither, by design |
| Analysis | Often qualitative, or limited to a subset of organoids | Whole-organoid imaging with quantitative readouts |
Comparisons describe general characteristics of manually produced organoid systems, not any specific commercial product.
One difference is not in the table. The method is published in the open literature and backed by an intellectual property portfolio originating from the Max Planck Society, including granted and pending patents covering the automated, gel-free production and the progenitor cells they are grown from. Alongside it sits process knowledge that does not expire: more than 150 unpublished production parameters accumulated over thirteen years of research, and deepening with every production run.
[PEARL: CONFIRM LICENCE WORDING BEFORE LAUNCH]
Your read deck describes an exclusive licence option from the MPG with a term sheet in place. This page therefore says the patents „originate from“ the Max Planck Society and makes no claim about who holds the rights. Once the licence is signed, this can state it plainly, which is considerably stronger. Do not upgrade the wording before then.
See what this enables in practice.
The technology is built around one question: is the result you get real, or is it an artefact of how the model was made? If that distinction matters to your work, let's talk.