Billions of addressable probes per wafer.
Scale the number of molecular questions asked in parallel while preserving the identity and location of every probe.
The DNA chip platform
Technology
Centrillion brings semiconductor scale manufacturing to molecular measurement. Photolithographically synthesized DNA chips place enormous numbers of addressable probes on a wafer, creating a flexible foundation for high scale biological assays.
Photolithographic synthesis
Write DNA using light.
A photomask selects which features receive UV exposure. Light removes the protecting group. The next DNA base couples only where the light landed. Repeat.
Because selection is optical, millions of features can be addressed in parallel during each synthesis cycle.
Cycle 000
Base A
Apertures 0 / 12
Mask align
A
C
G
T
Protecting group
Wafer scale production
Build molecular features with semiconductor precision.
The Centrillion lithography facility brings optical patterning, surface chemistry, and wafer processing into one production environment. Yellow safe-light conditions protect photosensitive process steps while the tools create densely patterned DNA surfaces.
Technology for synthesis
- Surface Chemistry
- Photoactive Nucleotides
- Spincell Synthesis
- DNA Printer Controllers
A scaling engine for biology
Apply the logic of Moore’s Law to molecular measurement.
Semiconductor progress came from placing more reliable sensing and computing elements into each manufacturing unit. Centrillion applies the same scaling logic to biology. More independently addressable DNA probes per wafer create more opportunities to ask biological questions in parallel and reduce the cost of each measurement.
A production wafer can carry billions of DNA probes distributed across many chips. Our long term roadmap is to extend that density by orders of magnitude toward trillion scale molecular sensing.
Density, yield, and cost specifications will be published for each named product and manufacturing revision.
Manufacturing foundation
Synthesize the assay directly on the wafer.
Photolithographic synthesis builds many different oligonucleotide sequences at known positions on a wafer. The result is not one generic sensor repeated many times. It is a densely patterned molecular surface whose probes can be designed for a specific biological question.
Centrillion developed dedicated fabrication processes and production capability around this approach. Wafer scale production connects assay design, surface chemistry, process control, imaging, and data processing in one repeatable system.
View a published high resolution chip patent record with a 150 mm wafer fabrication example.
Published example; current manufacturing specifications are product specific.
One surface, many assays
Probe DNA, RNA, and DNA encoded molecular signals.
DNA probes naturally recognize complementary nucleic acid sequences. With the appropriate assay chemistry and labeling, the same dense surface can support measurements of DNA and RNA, as well as workflows that convert protein or other molecular interactions into DNA readable signals.
This programmability makes the platform useful across targeted genotyping, pathogen detection, genome resequencing, spatial biology, and new assays that combine several molecular dimensions.
Supported analytes and performance depend on the assay configuration.
Data for biological AI
Build the volume and context that biological models require.
Biology is diverse, hierarchical, and context dependent. A genome contains long range sequence relationships. A tissue contains molecular states distributed across cells and physical space. Understanding either system requires data at a scale that sparse measurements cannot provide.
Large language models demonstrate how model capability grows when large collections of structured context become available. Biological models need the same foundation: precise, high dimensional measurements linked to sequence, cell, tissue, phenotype, and experimental provenance.
Centrillion is developing both the measurement platform and the data experience.
Design the surface around the assay.
Change probe content, chip format, and analytical workflow without giving up the manufacturing advantages of a shared wafer process.
Create data that models can learn from.
Connect molecular signals to context, quality, provenance, and interpretation so models can train on more than an isolated measurement.
Start a conversation