Identity
Measure the biological starting conditions of a study, at the resolution where differences between animals sharing one strain label actually appear.
sibylops · software · sensing · biological control
Make biology
repeatable & then
organs replaceable.
Find what changed. Explain divergence. Eliminate causes. Verify convergence.
New Age Sibyl Project (NASP) is building SibylOps (formerly LineageOps + SIBIL): Git-like versioning and root-cause analysis across biological R&D. It is one closed-loop system, a “Git for biology” that finds why biology diverges, helps eliminate the causes and verifies convergence, with the same architecture progressing from experiments toward animals, biological manufacturing and organ replacement.
sibylopsgit for biologyroot-cause analysisrodent-firstvendor-neutralstage-gatedautomationai-assistedreplacement biologylongevity infradeep tech
life science
remember
measure
predict
eliminate
verify
simulationphysical systemspartner lab
The problem
Same model. Different biology. Expensive failure.
Nominally comparable biological systems can diverge because lineage, microbiome, environment, handling, developmental history, interventions and process conditions accumulate across time. When that context is fragmented across systems, teams can see conflicting outcomes without knowing what changed. SibylOps turns those hidden differences into something that can be versioned, compared and investigated.
Development
Treat developmental environment and lineage history as part of the experimental system, recorded on the same footing as the protocol itself.
Reproducibility
Create stable reference baselines against which deliberate biological variation can be tested across sites and time.
Documented cases
What unmeasured variation has already cost.
Each of these is a published account of an experiment moving because something outside the protocol changed. They span oncology, immunology, ophthalmology and behavioural science, and every one of them was found after the fact.
Two suppliers, two outcomes
Genetically similar mice from different vendors grew tumours at substantially different rates, and responded differently to an immunotherapy. The difference travelled with the gut microbiota, and transferring it between animals removed the gap.1
Room temperature moved the biology
Standard mandated housing sits below the thermoneutral range for mice. Animals held at standard temperature showed faster tumour growth and measurably weaker antitumour immunity than animals held warm.2
Who handled the animals
Exposure to male experimenters produced a stress response and stress-induced analgesia in mice and rats, shifting apparent baselines in pain and behavioural testing.3
A variant inside a substrain
A spontaneous mutation in one commercial line produced striking immune phenotypes, and was carried into multiple gene-targeted lines before anyone identified it as the cause.4
The wrong gene credited
A retinal mutation present across an entire substrain family produced eye disease that had been attributed to the genes under study, affecting work across vision research.5
Routine husbandry, measurable response
Continuous home-cage monitoring across three institutions recorded activity responses to ordinary cage changes, alongside behavioural differences between sites running the same nominal strain.6
Why it matters commercially
Start where divergence already costs money.
In pharmaceutical and biotechnology work, a result that moves for an unrecorded reason is discovered late, usually during scale-up, replication or a partner handover, when the study has already been paid for. Recording the conditions at the time of the experiment is cheaper than reconstructing them afterwards, and it is what lets a disagreement between two studies be settled by evidence.
Where variance is already paid for
CROs, pharma R&D teams and advanced animal facilities, through a paid pilot or deployment.
Find what changed
Show what differed between nominally comparable studies, and why it mattered to the outcome.
Drive an intervention
Act on the likely cause and verify whether outcomes converge again.
From one site outward
A recurring site contract, then multi-site, then deeper sensing, biological control and biological manufacturing.
The system
SibylOps turns biological variation into an engineering backlog.
SibylOps versions biological and experimental context, connects changes to outcomes, identifies likely causes of divergence, supports intervention and verifies whether outcomes converge again. It is one system with three connected layers, developed together because each one determines what the others have to handle.
A “Git for biology”
Git-like versioning of lineage, environment, procedures and outcomes, with root-cause analysis on top, vendor-neutral and built with automation and AI-assistance. A working software PoC exists, using simulations and open research datasets, on widely used open frameworks.
Measure the state that matters
Longitudinal measurement of biological, environmental and process state, so that divergence is seen while it emerges. Sensing and laboratory hardware integrate into the same record once funded.
Act, then check it worked
Intervene on likely causes and verify convergence, at progressively deeper biological stages. Work runs through qualified partner laboratories under appropriate ethics and welfare review, and begins in earnest once capitalised.
One system
One machine. Progressively harder biology.
Each stage has to produce evidence before the next one is attempted, because the biological and regulatory difficulty rises sharply at every step.
Explain hidden variance
Find the hidden sources of outcome variance between studies that look identical on paper.
Control what drives outcomes
Control the biological dimensions that drive outcomes, starting with rodents and extending to larger animals only when earned, under the heavier oversight such work properly requires.
Repeatable biological states
Make intended biological states repeatable, with each result attached to the conditions that produced it.
Identity-preserved biology
The long-range objective that the earlier stages exist to make credible: reliable replacement with identity-preserved biology.
The closed loop
From what happened, to why, to what to change next.
Recording tells you what happened. Root-cause analysis helps determine why. The closed loop tests what to change next, and verification checks whether the intervention worked. Then the loop runs again.
Keep the history
Lineage, provenance and experimental history, in one linked record for each animal and study.
Capture current state
Biological, environmental and process state, through a vendor-neutral layer.
Locate divergence
Identify where divergence emerges, and which changes are its likely causes.
Act on the causes
Drive an intervention through qualified partner labs, with appropriate oversight.
Test for convergence
Check whether outcomes converge, with repeatable evidence, and feed the result back in.
Early Steps
The first 24 months.
These gates convert the current thesis into a working product, partner-lab evidence and a credible biological programme.
Months 0–3
Architecture gate
Technical architecture, literature map, failure-mode register, ethics route, simulation specification and external review.
Months 2–7
Physical-control gate
Benchtop sensing, imaging and environmental-control prototype demonstrating stable closed-loop operation.
Months 6–15
Biological-validation gate
Formal partner-lab collaboration and one predefined biological subsystem test with an animal-welfare plan.
Months 12–24
Replication gate
Repeatability data, independent review, comparison against simulation and preparation for a larger wet-lab financing round.
Horizon
The compounding roadmap
Each stage creates products, evidence and collaborations that support the next.
Years 0–2
Software PoC → physical integration → partner pilot
Take the working software PoC through physical sensing integration and facility pilots, and validate one defined biological subsystem with a qualified partner lab.
Years 2–4
Mouse-stage convergence evidence, then multi-site
Turn validated lineage, environment, behaviour and outcome data into evidence that interventions make outcomes converge, then expand across sites into deeper biological control.
Years 4–6
Biological manufacturing
Make intended biological states repeatable, advancing toward verified clonal-lineage control through staged R&D and external validation.
Years 6–10
Identity-preserved replacement biology
Expand through specialist collaborations toward veterinary cloning, tissue work and identity-matched replacement biology.
Team
Dedicated people across biology, instrumentation, software and protocols.
The core covers systems and software, precision instrumentation, experimental biology, and veterinary protocols. Several members contribute alongside ongoing degrees or research posts, deepening as the programme is funded.
S. W.
Biotechnology and information technology across the whole system. Published AI/ML research engineer with work spanning microscopy, computer vision, sensing, robotics integration and simulation, alongside product and commercial execution.
R. K.
Veterinary medicine with small-animal and equine clinical exposure. Leads animal-model workflow, welfare planning, protocol design and the bridge into partner laboratories.
S. D.
Experimental neuroscience with rodent surgery and formal laboratory-animal training. Optogenetics, electrophysiology, human iPSC-derived neuronal work, molecular cloning and broad wet-lab technique.
S. V.
Bioinformatics and full-stack scientific software. Structural-biology computation, transcriptomic analysis and pathogen genomics, plus dashboards, APIs and deployment.
S. A.
Precision scientific hardware across optics, mechanics and electronics. Detector characterisation, data acquisition and control, cryogenics, and the technical-commercialisation perspective that comes with it.
A. B.
Six years of production backend and data-platform engineering. High-throughput event-driven systems, schema validation, fault recovery and cloud operations, with a record of collapsing integration timelines from weeks to days.
Project Philosophy
Start practical. Compound toward the extraordinary.
New Age Sibyl Project starts with rodent operations and biological identity, then advances through stage-gated R&D toward clonal-lineage control and replacement biology with qualified collaborators.
- The aim is to produce a validated biological prototype through successive stages.
- Existing research supports individual components and partial transitions, and implies the feasibility of a complete repeatable pipeline.
- A working SibylOps software PoC exists, using simulations and open research datasets. It compares experimental context, preserves provenance, identifies changed versus held-constant conditions, and links differences to downstream outcomes.
- Next comes physical sensing, then partner facility or CRO integration, then intervention, then mouse-stage convergence evidence. Those stages await capitalisation.
- Initial biological work is non-human and rodent-first.
- Animal experiments would occur through qualified partner laboratories with required ethics and welfare review.
- Biomedical and pharmaceutical application is the long-term objective.
Organ failure should become an engineering problem instead of an endpoint.
This website describes an early research and venture direction, at the stage of a working software prototype and a capitalisation-dependent biological programme.