We believe that a kidney modelled at cell resolution, in the patient it came from, can name the target, the patient and the response for nearly every problem in kidney drug development — and that the medicine should be written directly against what those models find.
Each model learns one way of seeing the organ. The twin holds them together and answers a single question: what would have to change in this patient’s kidney for the disease to stop, and what medicine would change it.
Types cells on tissue it has never seen, without labels, and predicts how a neighbourhood answers when a gene is switched off.
Turns a routine biopsy slide into measured structure and molecular state, where pathology today returns a grade.
Reads the clinical course to say who will progress, and whether a treatment is being answered, before function is lost.
Turns a genetic variant into a predicted effect on a gene, in a named cell type, anchored in kidney regulation.
Built from human kidney biology and updated by every new observation. It exists to answer the three questions a medicine has to answer before it can help anyone.
Everything upstream is in service of this. The target comes from the twin, the medicine is written as a sequence, and it is designed for the patients whose own tissue shows that biology. When the target changes the sequence is rewritten, and the chemistry, the delivery and the manufacturing carry over, so the next medicine arrives faster than the last.
Discuss a program