Skip to content
Progenitor
Bio · Holding company
Progenitor Bio · Est. [PLACEHOLDER: year] · [PLACEHOLDER: city]

Some futures are built.Ours are grown.

Progenitor Bio builds companies and products for the future of organic reality.

Section 01 · Organic reality

Food, medicine, materials, ecosystems: much of what we depend on is grown, not made. That is organic reality — everything that grows, divides, metabolizes, and evolves. Working with it still runs on slow, expensive trial and error. Progenitor builds the companies and products that change that, each starting as one idea and growing into its own lineage.

Section 02 · Differentiation

Every lineage starts with one cell.

A progenitor cell divides, and its daughters specialize. Progenitor works the same way: one idea, proven small, grows into a company or product of its own.

LINEAGE 01DrylabIn-silico experiments for living cells · Private previewLINEAGE 02In developmentLINEAGE 03In development

Section 03 · Lineage 01 · Drylab

Run itdry first.

Drylab (drylab.app) runs experiments on living cells in software: gene knockouts and drug screens, from yeast to human cell lines.

Every run is quoted first, with its runtime, its dollar cost, and whether the model behind it is checked or still theory. Nothing is charged until you confirm. Reruns from the record alone come out byte for byte identical.

Private preview · invite only

TP1 · Glance · S. cerevisiae S288C

Which deletions are lethal?[1]

TP1 yeast gene essentiality matched held-out data: MCC 0.56 on 1,481 genes the model never saw.
  1. #1 YEL026W1.00
  2. #563 YDR299W0.85
  3. #1125 YDR177W0.60
  4. #1688 YPL059W0.38
  5. #2250 YIL042C0.24
  6. #2812 YLR207W0.17
  7. #3374 YHR053C0.11
  8. #3937 YBR298C0.07
  9. #4499 YOL065C0.03
  10. #5061 YLR240W0.00
Predicted chance each single-gene deletion is lethal. Ten of 5,061 genes, taken evenly down the full ranking, so you see the whole range.

TP1 · Glance

Which of these genes is likely essential? Ranks a whole genome.

Method
Learned surrogate
Runtime
Seconds

A screening and triage tool for research. It does not replace wet-lab experiments. Not for clinical use.

Section 04 · Principles

How we decide what is true.

  1. 01

    Checked or theory

    Every claim we make is labeled. Checked means it matched real data it never saw. Everything else is theory, and says so.

  2. 02

    Reality has the last word

    Models, simulations, and products earn trust by meeting the real world. When they disagree, the world wins and the model changes.

  3. 03

    Reproducible by default

    Same inputs, same result, every time. If we cannot rerun it from the record, we do not report it.

  4. 04

    Start with one cell

    Begin small, prove it, then grow it into its own lineage. Breadth comes from many small things that work.

Section 05 · Drylab by the numbers

reactions modeled
4,105[2]
genes
1,143[2]
compute tiers
5[3]
bytes differ on rerun
0[4]

Gene essentiality in yeast is checked: MCC 0.56 on 1,481 held-out genes[5]. MCC (Matthews correlation coefficient) scores a yes/no prediction from −1 to 1, where 0 is chance. Drug screens are still theory.

Section 06 · Journal

[PLACEHOLDER: journal post 1: title, category, date, one-line summary]
[PLACEHOLDER: journal post 2: title, category, date, one-line summary]
[PLACEHOLDER: journal post 3: title, category, date, one-line summary]

Careers

Help grow the next lineage.

See open roles

Section — · Sources

Sources

  1. [1]Drylab TP1 (Glance) output for S. cerevisiae S288C, model vcl-tp1-go-essentiality v1: all 5,061 single deletions ranked. Source ↗
  2. [2]yeast-GEM v9.1.1, SysBioChalmers, the genome-scale model Drylab pins by checksum. Source ↗
  3. [3]Drylab capability matrix: the five compute tiers and the cells each serves. Source ↗
  4. [4]Drylab, how we check: reruns from the run record alone, compared byte for byte. Source ↗
  5. [5]Drylab validation report: yeast gene essentiality, MCC 0.56 on 1,481 held-out genes. Source ↗