OS-01, also called Peptide 14, is a peptide developed by OneSkin and identified through computational screening rather than found in nature or derived from a known protein. It targets senescent cells in skin — cells that have permanently stopped dividing but remain in the tissue secreting inflammatory signals.

What are senescent cells?

Cells that have permanently stopped dividing after accumulating enough damage or divisions. The state is protective, since a damaged cell that cannot divide cannot become a tumour. The problem is that these cells neither die nor leave — they remain metabolically active and secrete inflammatory signals that appear to drive dysfunction in healthy neighbouring cells.

What is the difference between a senolytic and a senomorphic?

A senolytic kills senescent cells outright, which is decisive but requires very good selectivity because cell death is irreversible. A senomorphic leaves the cells in place but quiets their inflammatory secretion, which is gentler and reversible but does not remove the cells, so the effect has to be maintained. OS-01 is presented as a senomorphic; FOXO4-DRI is a senolytic.

What does it mean that OS-01 was discovered by AI?

That it was identified by generating and computationally evaluating a very large space of candidate sequences against a desired biological signature, then testing the highest-scoring few. This can reach sequences no organism ever produced. It describes how the compound was found, not how well it works — a screen produces candidates, it does not validate them.

How strong is the evidence for OS-01?

Modest. The main work uses ex vivo human skin models — real human skin kept alive outside the body, a better model than cultured cells because tissue architecture is intact — and reports reductions in senescence markers and improvements in barrier measures. Most research comes from the company that sells it, independent replication is limited, and ex vivo skin lacks a circulatory and immune system.

Research Library›Skin & Anti-Aging›OS-01 (Peptide 14)

Research Compound

OS-01 — The First Peptide in This Library Found by a Computer

Q: Does OS-01 penetrate skin?

It faces the same barrier constraint as every topical peptide. The outer layer of skin substantially restricts what passes through, so how much reaches living tissue is a separate question from what the compound does once it arrives. This limitation applies across the cosmetic peptide category rather than being specific to OS-01.

It was not discovered in nature or cut from a known protein. It was screened computationally from a large space of candidate sequences, and it targets something no other compound here does: cells that have stopped dividing but refuse to leave.

7 min read·3 studies cited·Updated August 2026

Evidence level

Preclinical only — animal and cell studies

Ex vivo human skin models; company-run

What this level means

Quick Reference

Also called	Peptide 14
Discovered by	Computational screening, not extraction
Target	Senescent cells in skin
Class	Senotherapeutic — presented as senomorphic
Evidence	Mostly ex vivo skin models; company-run

The Cells That Will Not Leave

Cells have a limit. After enough divisions, or enough damage, a cell stops dividing permanently — a state called senescence. This is protective: a damaged cell that cannot divide cannot become a tumour.

The problem is what happens next. A senescent cell does not die and it does not leave. It sits in the tissue, metabolically active, and it secretes a steady stream of inflammatory signals into everything around it.

Think of one retired employee who still comes into the office every day, occupies a desk, and complains loudly to everyone within earshot. They are not doing the job any more. That would be tolerable. The difficulty is the effect on everyone still working nearby — over time the surrounding staff start behaving worse too. One is a curiosity; a few dozen changes the atmosphere of the whole floor.

That secretion is the real issue, and it is why senescent cells attract so much ageing research. They accumulate with age, they are especially visible in skin, and their inflammatory output appears to drive dysfunction in healthy neighbours. The damage is not the cell that stopped working. It is what the stopped cell keeps saying.

Two Ways to Deal With Them

There are two approaches, and confusing them makes most writing on this subject incoherent.

Approach	What it does	The trade-off
Senolytic	Kills senescent cells outright	Effective and blunt — cell death is irreversible and selectivity must be very good
Senomorphic	Leaves the cells but quiets their inflammatory secretion	Gentler and reversible — but the cells remain, so the effect must be maintained

Returning to the office: a senolytic escorts the retired employee out of the building. A senomorphic lets them keep the desk but stops them talking. The first is decisive and harder to get wrong safely. The second is milder and does not solve anything permanently.

The library already covers a senolytic — FOXO4-DRI, which works by disrupting a mechanism senescent cells use to avoid dying. OS-01 is presented as the other kind: aimed at reducing markers of senescence and the associated inflammatory signalling rather than at removing the cells.

How It Was Found

Every other peptide in this library arrived in one of two ways. Either it was found in a living thing — oxytocin, LL-37, the bioregulators — or it was cut down from a larger known protein, as TB-500 was from Thymosin Beta-4.

OS-01 came from neither. It was found by computational screening — generating and evaluating a very large space of candidate sequences against a desired biological signature, then testing the small number that scored best. It was developed by OneSkin, where it is also called Peptide 14.

The difference is prospecting versus surveying. Traditional discovery is walking the ground and noticing something promising in a riverbed — it works, and you can only find what happens to be lying where you walked. Computational screening surveys a far larger territory at once and marks the places worth digging. It can reach sequences no organism ever made and no chemist would have thought to try.

What this does not do is tell you the compound works. A screen finds candidates; it does not validate them. It is worth stating clearly because "AI-discovered" is doing promotional work in this field at the moment, and it describes how something was found rather than how well it performs. The evidence question is unchanged, and unimpressed by the origin story.

What the Evidence Amounts To

Modest, and worth reading with the commercial context in view.

The published work centres on ex vivo human skin models — real human skin kept alive outside the body and treated directly. That is a genuinely better model than cultured cells in a dish, because the tissue architecture is intact. Reported findings include reductions in senescence markers and improvements in measures of skin barrier and structure.

The qualifications are substantial. Most of the research comes from the company that developed and sells the compound. Independent replication is limited. And ex vivo skin, while a good model, is skin disconnected from a circulatory and immune system — closer to real skin than a dish, still not the same as skin on a person.

It also faces the delivery constraint that limits every topical peptide, described in Argireline: the skin barrier restricts what gets through, and how much reaches living tissue is a separate question from what the compound does once there.

The fair summary is a genuinely novel target, a genuinely novel discovery route, and early evidence from interested parties. That is a more interesting position than most cosmetic ingredients occupy and it is not the same as a demonstrated result.

OS-01 is covered because it is the first senomorphic and the first computationally-discovered compound in this library, and both of those are worth understanding.

Frequently Asked Questions

Compiled by the Southern Pacific Labs research team. Last updated August 2026.

Browse the research range →

Published References

Zonari A, Brace LE, Al-Katib K, et al. “Senotherapeutic peptide treatment reduces biological age and senescence burden in human skin models.” npj aging, 2023. PubMed
The 2024 correction replaces an unintentionally duplicated image in Figure 6a. The authors state that the correction does not change their main conclusions. View notice
Coppé JP, Desprez PY, Krtolica A, et al. “The senescence-associated secretory phenotype: the dark side of tumor suppression.” Annual review of pathology, 2010. PubMed
Kirkland JL, Tchkonia T. “Senolytic drugs: from discovery to translation.” Journal of internal medicine, 2020. PubMed

Browse all OS-01 (Peptide 14) research on PubMed →