A peptide array and a set of purified peptides answer different questions, and a surprising amount of wasted effort comes from asking one format to do the other's job. An array surveys hundreds of sequences in a single experiment but cannot tell you how tightly any one of them binds. A purified peptide in a tube gives you a number you can defend, for one sequence at a time. The sensible workflow uses both, in that order.
The distinction is not really about throughput. It is about what the peptide is attached to and how well you know how much of it there is.
On an array, each peptide is synthesised in place on a support and stays there. It is covalently anchored, usually through its C-terminus, and it is never purified, never weighed, and never quantified individually. You read out a signal per position and compare positions with one another. On a cellulose SPOT membrane, peptides are covalently bound to the membrane via the C-terminal end, with roughly 5–10 nmol per spot at a spot diameter of about 2–3 mm.
In solution, each peptide has been cleaved from the resin, purified, characterised by HPLC and mass spectrometry, and supplied with a known content. You dilute it to a known concentration, present it free in both orientations, and fit a curve. The result is a dissociation constant, an IC50, a kcat/KM — a number with units that can be compared against a number from another laboratory.
Scale, and the ability to ask a question you could not otherwise afford to ask. The SPOT technique, first described by Ronald Frank in 1992, builds addressable peptides directly on a membrane by pipetting activated amino acids onto defined positions1. Later reviews credit that paper as the seminal description of simultaneous parallel peptide synthesis on a membrane support4, and the method has since been written up as a standard protocol3. The published protocol reports up to 1,000 peptides on a single 20 × 29 cm cellulose sheet3; spot size and spacing, and therefore density, differ between formats. Slide-based formats are denser: CelluSpot slides hold up to 768 spots at 0.8 mm diameter and 1.2 mm spacing, which allows 384 peptides to be printed in duplicate. Reviews of the method describe arrays of roughly 25,000 peptides prepared by SPOT synthesis2. For a sense of pace rather than capacity, the protocol gives 400 nine-residue peptides in six days3.
That scale changes which experiments are reasonable. A full overlapping scan of a 300-residue antigen in 15-mers stepped by three residues is about a hundred peptides — trivial on one membrane, a serious synthesis project in purified form. A positional scan that replaces every position of a 12-mer with each of the other nineteen amino acids is 228 analogues. Neither is a candidate for individual purification unless the hit rate is already known to be high.
Arrays also tolerate chemistry that would be awkward to buy one peptide at a time. Because the same reservoirs serve every position, D-amino acids, phosphorylated residues and other non-standard building blocks can be distributed across the design without commissioning each variant as a separate synthesis, though non-standard monomers differ in cost, stability and coupling behaviour. The read-out is similarly flexible: the same membranes have been used for antibody binding, for enzyme activity and for antimicrobial screens, with the peptide loading per spot deliberately varied between designs5.
Three limits matter, and all three are structural rather than a matter of technique.
The peptide is not free. It is tethered at one end and presented at high local density against a surface. A sequence whose binding epitope sits near the C-terminus may be occluded; a sequence that only works as a free acid will behave differently when its C-terminus is the attachment point. Density effects cut the other way too, raising apparent avidity for a weak interaction that would not survive dilution in solution.
You do not know the purity of any individual spot. The array is made under one set of conditions for every sequence, and difficult sequences fail quietly. As the standard review of the method puts it, the large number of sequences means coupling yields can vary, so the amount as well as the purity of the peptides may differ, and the low amount of peptide per spot makes them very difficult to purify2. A blank spot is therefore ambiguous: it may be a non-binder, or it may be a sequence that did not assemble.
The signal is not a measurement. Spot intensity depends on peptide density, on how much of that density is the correct sequence, on exposure, on local membrane properties and on the detection chemistry. Intensities within one membrane, synthesised together, can be ranked with some confidence — the method is described by its own developers as able to distinguish binding affinities only semi-quantitatively, and only within a single array6 — which is why taking arrays from the same membrane is recommended, so that spots have similar peptide density5. A dedicated reliability study put a figure on what "semi-quantitative" means in practice: signal intensities could be assigned accurately to three binding-affinity classes, not to a continuous scale of KD8. Across membranes, or converted to an affinity, they are not trustworthy.
The practical consequence: an array result is a ranked shortlist with a specific failure mode. Strong spots are good candidates. Weak spots are not evidence of non-binding. Treat the output as a hypothesis-generating filter, and budget for the confirmation step before you run the screen.
Defensibility. A purified peptide comes with an HPLC trace, a mass, and a stated content, so the concentration axis is quantitatively meaningful. You can present the peptide in whichever orientation your biology requires, label it or immobilise it deliberately rather than as a consequence of how it was made, and run the orthogonal assay that a reviewer will ask for.
It also lets you measure the things an array has no access to: a true KD by surface plasmon resonance or isothermal titration calorimetry, a competition IC50 against the native ligand, kinetics, stoichiometry, solubility, aggregation behaviour and stability in serum. None of these can be extracted from a spot.
| Question | Format | Why |
|---|---|---|
| Which regions of this antigen are recognised? | Array | Needs full coverage; relative ranking is enough |
| Which positions in this binder can be substituted? | Array | Hundreds of analogues, comparative read-out |
| How tightly does my lead bind? | Solution | Requires known concentration and free peptide |
| Is analogue A better than analogue B? | Solution | Spot intensities do not support that comparison |
| Does this peptide work in cells? | Solution | Peptide must be free, soluble and dosed |
| Is this spot a real negative? | Solution | Array blanks are confounded with synthesis failure |
In practice the sequence is: screen broadly on the array, take the top candidates plus a few informative negatives, have those resynthesised and purified, and measure them properly. This is the structure of published selectivity work too. In one antibody cross-reactivity study, a spot permutation array was followed by surface plasmon resonance on the survivors; the kinetic measurements largely validated the stability ranking that the array had suggested, while also establishing how sensitive the spot intensities actually were7. That is the honest description of the relationship between the two formats: the array orders the candidates, and the solution measurement tells you what the order is worth. The negatives are worth the money — if a predicted non-binder turns out to bind in solution, something about the array needs tracking down — synthesis on the support, presentation of the tethered peptide, or a genuine difference between the two formats — and that changes how you read the rest of the membrane.
For the design of overlapping and substitution scans, see our page on epitope mapping and peptide scanning. For what a dissociation constant from the confirmation stage does and does not tell you, see peptide versus protein binding affinity, and for the negative and scrambled sequences that make either format interpretable, see control peptides in experimental design.
| Array strength | Hundreds to thousands of sequences in one experiment; comparative ranking |
| Array weakness | Tethered peptide, unknown per-spot purity, no quantitative affinity |
| Solution strength | Known purity and concentration; free peptide; defensible constants |
| Solution weakness | Cost and time scale with the number of sequences |
| Recommended order | Array to shortlist, purified peptides to conclude |
LifeTein builds peptide libraries for screening and purified peptides with full HPLC and mass spectrometry characterisation for the confirmation stage. Tell us the sequence or the protein you want scanned and we will quote both halves of the workflow, so the follow-up cost is visible before the screen starts.