This page goes with Making the origin of life a solvable problem. That page turns the origin of cells into a ledger of inequalities that any history must have satisfied. Most of the ledger is already pinned by published measurements. What remains is a short list of numbers nobody has measured yet. Each one moves a specific inequality past or below its threshold.
If your lab can make one of these measurements, or already has the data, we would like to hear from you: write to josephbrisendine@gmail.com. We can't do bench work ourselves. What we offer is the reading: for any value you measure, we can say what it opens or closes on the path, in the same units as everything else.
Measurethe quantity, with conditions as close as practical to those stated
Thresholdthe value that flips the inequality; a number on either side is informative
What it decideswhich step it opens or closes, and how many bits it moves
Our expectationregistered before the measurement, so it can be wrong
A measured value that differs from ours is still useful. It goes straight into the same inequality and moves the slack, which is log₂(available ÷ required). The ledger is built so that being wrong in a number is a correction, not a collapse.
Completion: the largest bill
copying
A QT45-class ribozyme copier loses about 11.3 bits per copy, roughly one working copy in 2,500. A lineage making 40 copy attempts per division can afford about 3.8. Fidelity accounts for only 2.3 of those bits; most of the deficit is copies that never finish.
1
Re-priming across a cycle
Measure
The fraction ρ of stalled primer–template complexes that resume extension after one freeze–thaw (or pH plus freeze–thaw) cycle
Threshold
ρ ≥ 0.5, with a ×5 tether gain
Decides
Whether the environment can pay the completion bill (−3.3 bits at the junction; 9 of QT45's 11.3 bits per copy). If ρ ≈ 0, that route closes.
We expect
ρ > 0, since cycling already improves yield about ×1.2–1.4 per cycle (Attwater et al. 2025); the magnitude is unknown
2
Completion on a tethered partner
Measure
Per-junction completion of a QT-class copier copying a partner tethered to it
Threshold
≥ 0.95 for the copier; ≈ 0.97 for a cis genome of about 130 nt
Decides
Tests the tether gain the ledger assumes (×5, not yet measured)
We expect
A gain of ×5–10 per junction
3
The commitment ratio per junction
Measure
kc/koff at a junction: the rate of chemical extension relative to the rate the primer falls off, on a tethered template
Threshold
about ×17 above QT45's free-template value
Decides
Whether structure alone (effective molarity from a tether) can pay completion without help from cycling. Completion is a matter of commitment, a rate ratio, not of dissipation per step.
We expect
A ×5 tether gets a little over half of the way (in log terms); the rest has to come from re-priming over time
4
Copies per protocell division
Measure
Complete genome copies made per protocell division, in any fatty-acid protocell system
Threshold
≳ 5–9 complete copies per division
Decides
Every rent threshold in the ledger moves by log₂ of this number: doubling the copies per division buys one bit per copy
We expect
Unknown; not yet measured for any fatty-acid protocell
5
Fidelity at useful yield
Measure
Copying error rate per site at a yield that a lineage could live on
Threshold
ε ≈ 0.02 per site
Decides
The fidelity term of Eigen's inequality. QT45 (0.059) falls short by only 0.1–0.6 bits.
We expect
Reachable
Membranes and magnesium
compartments
RNA copying without help wants ≳ 30–50 mM Mg²⁺; pure fatty-acid membranes tolerate about 1–5 mM. Which of these gives way decides whether an attached amino acid is needed on the first copier at all.
6
RNA retention through freeze–thaw at high Mg²⁺
Measure
RNA retention through freeze–thaw of fatty-acid vesicles with some phospholipid, at 25–50 mM Mg²⁺ (also cold and unfrozen)
Threshold
Tolerance ≈ 4 + 150x mM (x = phospholipid fraction; a fit to one study) predicts ≥ 40 mM at x ≈ 0.25
Decides
Whether a Mg²⁺-rich pond needs the amino-acid handle at all. No retention data exist.
We expect
Retention-grade tolerance somewhat below the dye-based 40 mM
7
Content release per freeze–thaw cycle
Measure
The fraction of encapsulated content released per freeze–thaw cycle, as a function of phospholipid fraction
Threshold
≲ 4–8% per cycle for a lasting genetic code
Decides
Fluid phospholipid membranes are measured at 18–57% per cycle, which would rule out keeping a code inside a freezing nursery unless some membrane or setting retains far better
We expect
Gel-phase membranes under 10%; fluid membranes 20–50%
8
Release under slow freezing, gel-phase blends
Measure
Release per cycle for saturated (gel-phase) phospholipid blends under slow freezing, at about 30 mM Mg²⁺
Threshold
≲ 4–8% per cycle
Decides
Where a lasting code can be kept. In our models no setting has yet been shown to keep one; this is the most likely candidate inside the freeze–thaw nursery, and otherwise the code has to wait until the lineage leaves it.
We expect
Open; this is the ledger's main unresolved question
9
Membrane conversion under pond conditions
Measure
Conversion of fatty acid to phospholipid by an acylating ribozyme, or by thioester transacylation of lysophospholipids (shown without enzymes by Liu et al. 2020), ideally with a catalyst that works at neutral pH
Threshold
≳ 0.5–1 conversion per year
Decides
Makes phospholipid-driven growth heritable. A neutral-pH catalyst would also test whether coupling could come first, which the ledger leaves open.
We expect
The thioester route works if thioesters are supplied
10
Rupture and dispersal per cycle
Measure
Vesicle rupture and content dispersal per freeze–thaw cycle
Threshold
Sets the escape edge: −log₂ P(content survives a generation)
Decides
Prices the last tightening, when a lineage gives up the dispersal its setting supplied
We expect
Escape pays +0.5 to +1.5 bits per generation in separated droplets
The amino-acid handle and the first code
charging
An aminoacylated primer extends about ×100 faster at 0 mM Mg²⁺ in nonenzymatic extension (Radakovic et al. 2021). The ledger says that benefit pays only below about 25–35 mM Mg²⁺, and that the first code bit has to be kept accurate by discrimination or a recheck.
11
A ribozyme copier using an attached amino acid
Measure
Whether a ribozyme copier (not just nonenzymatic extension) finishes copies at low Mg²⁺ when an amino acid is attached
Threshold
Copying restored toward its high-Mg²⁺ rate, below ≈ 25–35 mM
Decides
The existence test for the handle step
We expect
It works, with any α-amino acid: a free α-amine is what matters
12
Charging supply and charge lifetime in the cold
Measure
Supply of activated amino acids per vesicle in a cold pond; lifetime of the charged handle at 0–5 °C and its Mg²⁺ dependence
Threshold
≈ 2–15 charges per vesicle per charge lifetime
Decides
Whether the handle can be funded
We expect
Half-life of days at 0 °C (about ×29 longer than at 25 °C); Mg²⁺ shortens it
13
Charging specificity for close pairs
Measure
Specificity of self-aminoacylating ribozymes for close amino-acid pairs (Ile vs Val vs Leu)
Threshold
≥ 3 kT single pass, or 2 kT plus one recheck; ≈ 5 kT for a majority code without a recheck
Decides
Whether early charging can be accurate without proofreading. A code is only as good as its weakest assignment, so the closest pair is the one that matters.
We expect
Close pairs at ≤ 2.2 kT (from aptamer data), so a recheck is needed
Lower priority, still useful: nucleotide permeability against temperature and phospholipid fraction (sets the feeding limit, predicted at 0.83–0.94 phospholipid fraction); misincorporation with bound peptides (matters only for peptides kept private to a protocell); ligation length statistics in the cold.