Estimate from the top concentration and the total treated volume, not from the number of experiments, and then add a margin for losses that do not appear in any protocol. Running short halfway through is worse than over-ordering, because it forces a lot change in the middle of a study and a new batch is not interchangeable with the old one.
The calculation, worked backwards
The quantity needed is set by the highest concentration used and the volume it has to fill, because everything below the top of a dilution series is made from material already counted.
Take a concentration–response curve with eight points, triplicate wells, 200 µL per well. That is 24 wells, but only the top concentration consumes material at full strength — so the figure that matters is the top concentration multiplied by the total volume that will contain it, plus the volume needed to make the series.
| Input | Example | Why it matters |
|---|---|---|
| Top concentration | 10 µM | Sets the mass; everything lower is made from it |
| Molecular weight | 1,400 g/mol | Converts molarity to mass |
| Total volume at top | 3 wells × 200 µL | Plus the dilution series volume above it |
| Replicates and repeats | 3 biological | Multiplies everything |
| Net peptide content | 82% | Powder mass is not peptide mass |
| Losses | See below | Routinely 20–50% in practice |
Why does the top concentration dominate?
Because a serial dilution reuses material. Making 1 mL at 10 µM and stepping down tenfold five times consumes the 1 mL once, not six times — each lower point is made from a portion of the one above it.
So doubling the number of points in a curve barely changes consumption, while doubling the top concentration doubles it outright. A design that extends the curve downward is nearly free in material terms; one that extends it upward is not.
What losses does a protocol not mention?
Four, and together they are usually larger than anyone expects.
Dead volume. Every tube, tip and reservoir retains liquid that is never delivered. Across a plate's worth of transfers this is a real fraction.
Adsorption. Material binds to plastic, and the proportion lost rises as concentration falls — so the dilute end of a series loses the largest share.
Filtration. If the solution is filtered, the membrane takes a fixed mass and the housing retains a volume.
Failed runs. The plate that was mispipetted, the assay that did not work, the repeat nobody planned. This is the largest of the four and the one never budgeted.
A practical margin
The honest approach is to calculate the theoretical requirement, then apply a multiplier based on how much is already known about the system.
For an established assay with known handling, 1.5× the calculated figure is usually comfortable. For a new assay, or one where the concentration range is not yet settled, 2× to 3× is more realistic — much of the material in exploratory work goes into finding out what range to use.
That sounds wasteful and is usually cheaper than the alternative. The cost of a second vial is small against the cost of a study that has to bridge two lots mid-way, or be partly repeated because the second half is not comparable to the first.
Is it better to buy one large vial or several small ones?
One lot, aliquoted at the start, for anything where comparability across the study matters. That gives a single impurity profile and a single content figure throughout, which removes the most common invisible variable.
The format question is separate. A larger vial costs less per milligram because synthesis and testing are batch costs, but a vial entered repeatedly accumulates moisture and handling exposure. Buying the larger vial and aliquoting it immediately captures the saving without the exposure.
What if the study runs longer than the material lasts?
Then the lot change is a planned event rather than an accident, which is a far better position. Order the second lot early, run an overlap — the last of the old and the first of the new, in the same session on the same assay — and the offset between them becomes a measured quantity instead of an unexplained step in the data.
That overlap costs one extra arm and it is the difference between a study with a documented transition and one with a discontinuity nobody can account for.
How does compound choice affect the budget?
Through molecular weight more than anything else, because molarity is what assays specify and mass is what you buy. At the same molar concentration, a compound of around 4,000 g/mol consumes roughly three times the mass of one around 700 g/mol.
That ratio surprises people who budget by habit rather than calculation, and it is why equal mass is not equal molecules matters for purchasing as well as for pipetting. A mid-weight peptide sits between the two and is a reasonable mental benchmark.
A short worked estimate
Suppose eight concentrations, triplicate, 200 µL per well, three biological repeats, top concentration 10 µM, molecular weight 1,400, net content 82%.
Material at the top point: 3 wells × 200 µL = 600 µL at 10 µM. Add a working volume for the series, say 1 mL at the top. Call it 1.6 mL at 10 µM per repeat, which is 16 nmol, or about 22 µg of peptide. Three repeats is 66 µg.
Correct for content: 66 ÷ 0.82 is about 80 µg of powder. Apply a 2× margin for the losses above and a failed run: roughly 160 µg. A single 5 mg vial covers that many times over — which is the usual conclusion, and the reason this calculation is worth doing once rather than agonising over.
Where it stops being trivially covered is at higher concentrations, in animal work, or where a compound is unusually heavy. Those are the cases the arithmetic is actually for.
What the estimate is for
The arithmetic matters less than the habit of doing it before ordering rather than after running short. Most of the time the answer is that a single standard vial covers the study several times over, and the five minutes spent confirming that is five minutes well spent.
Where it changes a decision is at the edges: a heavy compound at a high top concentration, an animal study, or a programme that will run long enough that one vial will not last. Those are the cases where guessing produces either waste or a mid-study lot change.
Does the margin differ between in-vitro and in-vivo work?
Substantially. Cell work consumes microlitres at micromolar concentrations, so quantities are small and a generous margin costs little. Animal work consumes far more by volume and by mass, so the same proportional margin is a much larger absolute number and the estimate deserves more care.
The structural difference is that the two formats answer different questions and are rarely interchangeable in planning. Estimating one from experience with the other is where the largest errors come from.
What is the most common planning mistake?
Budgeting for the experiment as designed rather than for the experiment as it will actually run. Pilot runs, failed plates, a concentration range that turns out to be wrong, and a repeat requested by a reviewer are all normal and none appear in the protocol.
A 2× margin absorbs most of that. Ordering exactly the calculated amount guarantees a reorder, and a reorder means a new lot with its own content figure arriving mid-study.
Should the estimate be written down?
Briefly, with the assumptions. Top concentration, volumes, repeats, content figure and the margin applied — five numbers on one line. The value is not the arithmetic but that the next study can start from a record of what was assumed rather than redoing it from memory.
It also makes an over- or under-estimate diagnosable afterwards. A study that consumed three times the prediction has something interesting in it, usually a loss nobody had accounted for, and that is only visible if the prediction was recorded.
Does the same logic apply to ordering supplies?
It does, and supplies are the more common thing to run short of. Solvent, syringes and consumables are consumed per-session rather than per-nanomole, so the estimate scales with the number of sessions rather than with concentration.
The failure mode is also different: running out of peptide stops the study, while running out of solvent mid-session invites a substitution that was never part of the plan. Changing diluent partway through is a variable nobody intended to introduce.
One last framing worth keeping. The purpose of the estimate is not to minimise what you buy; it is to avoid the one outcome that cannot be fixed afterwards, which is a study split across two lots because the first ran out. Material is cheap against that, and the calculation exists to make sure the cheap option is the one taken.
All products referenced here are supplied for laboratory and research use only. They are not drugs, foods, supplements or cosmetics, and are not for human or veterinary use.
