The Thesis

Nature varies one base at a time. So must the data.

Human variation is overwhelmingly single-nucleotide, and WGS and GWAS return it that way. A model trained on gene knockouts is one full resolution coarser than the question it is asked.

GENE-LEVELONE GENEON / OFF1 MEASUREMENTOne answer per gene: present, or absent.VARIANT-LEVELMEASURED EFFECT1000+ MEASUREMENTS · SAME GENEOne answer per base: which one, and how much.
The existing paradigm

Gene-level, in cell lines

  • Deletes whole genes — patients carry variants, not knockouts
  • Runs in immortalized lines standing in for patient biology
  • A trillion cells at the wrong resolution is still the wrong resolution
Codebreaker

Variant-level, in primary human cells

  • The same unit a WGS or GWAS hands you, and the same unit the model must predict
  • Tested in the disease-relevant primary cell, not a stand-in line
  • Breaks linkage — variants tested one at a time, at scale
CODEX

Not one dataset. An Atlas at a time.

Each Atlas is a finished, licensable asset the day it completes — measured data and the models trained on it. Every one deepens CODEX.

01

Our AI reads everything

Models trained in-house mine the world's genetic evidence and nominate every variant worth testing. For most of the field, in-silico prediction is the finished product. For us it is step one.

What step 01 reads
GWAS CatalogClinVarBiobank cohorts Published case literaturePartner datasets
02 in parallel

Map the variants

Multiplexed CRISPR in the disease-relevant primary human cell. Runs with or without a partner.

Partner with the KOL

Where a cohort investigator is involved, they bring direction, credibility and clinical reach.

03

Test in parallel

10K variants — every program sharing a cell type runs at once.

04 license either, or both

Ship the Atlas

The reference causal variant-to-function dataset for that indication.

Build / train the AI

Models trained on that Atlas. License the insights in addition to the data, or instead of it.

↻ Every completed Atlas compounds into CODEX
Collaborate

Which variants would you test if you could?

A GWAS never functionally resolved. A platform that needs a causal layer. A model that needs ground truth.

Your privacy choices

We would like your consent before any analytics or marketing technology is loaded on this site. Nothing in those categories loads unless you allow it. Strictly necessary storage — used only to remember this choice — is always on. See our Privacy Policy.