Antibody design service: the questions clients ask us most

The questions that come up in nearly every first call, answered in the order they are asked. How long it takes, what the three entry points are, who owns what at the end, what happens if nothing binds, and what we will not do.

Design around an antibody patent: a redesigned VHH binding the same epitope as the patented antibody
Meeting table with an open notebook and a printed protein structure, a laboratory behind a glass wall

These are the questions prospects put to us about our antibody design service in nearly every first conversation, answered in the order they are usually asked. The short version: six weeks from your antigen to candidate sequences and twelve weeks to binders validated in the laboratory, three different points at which a project can start, and the sequences and any intellectual property arising from the work are yours. The rest of this page covers what sits behind those answers, including what happens when a campaign does not work and what we do not do.

How long an antibody design service takes, and why

How long does it take?

Six weeks from go-ahead to candidate sequences, twelve weeks to binders validated in the laboratory. Those are working timelines rather than best cases. What sits inside them is set out week by week in our note on antibody development timelines.

What has to be ready before the clock starts?

A structure of the target, experimental or predicted, and a decision about which site you want engaged. If the site is not settled, that conversation is part of the free feasibility study rather than a delay.

Can it go faster?

Not meaningfully, and the limiting stage is rarely the one people expect. Generation is compute and can be scaled, whereas expression and testing are biology and take the time that biology takes. A shorter engagement is possible if you want the design work without the validation round, although what you then receive is predictions rather than measured binders, which we would state explicitly in the quotation.

Three points at which a project can start

What are the three entry points?

  • De novo design from the antigen sequence: you have a target and a site but no binder, and we design candidates against it.
  • Redesign around a reference sequence: you already have a binder, yours or someone else's, and it binds perfectly well. The difficulty lies elsewhere: it may carry a developability liability, or it may sit too close to a sequence someone else has claimed. We keep the binding mode and move the sequence away.
  • Verification of existing candidates: you have sequences and want them assessed before committing to scale-up, covering developability, specificity risk and distance from prior art. This happens under NDA and requires no design campaign at all.

All three apply to single-domain antibodies as well as to conventional antibodies.

Which one applies to me?

The first applies if you have no binder at all. The second applies if you have one that binds but cannot be used as it stands, which is the route described in designing around an antibody patent. The third applies if you already hold candidates and have a decision to make about them.

Do I have to commit to the whole programme?

No. Each step is agreed on the strength of the previous one, and you never commit to the next stage before the previous one has succeeded. The feasibility study is free and comes first.

Who owns the sequences an antibody design service delivers

Who owns the sequences you design?

You do. The candidate sequences, the data generated during the project and any intellectual property arising from the work are yours. We do not retain rights to the molecules we design for you, and we do not reuse them for other clients.

Do you keep anything?

Our methods, our filtering criteria and our internal tooling remain ours. That is the distinction: the know-how used to produce the molecule is ours, the molecule is yours.

Can we patent what you deliver?

That is your decision and your counsel's. What we supply is the technical record that supports it: the design rationale, the candidate sequences, the measured binding data, and the identity measurement to any prior art you asked us to work against. Whether a given candidate is patentable, and whether it falls outside a given claim, are legal questions and we do not answer them.

Will you work with our competitors?

We work on many targets. We will not take a project that uses your sequences or your data, and confidentiality agreements are standard before anything technical is exchanged. If a conflict is specific and foreseeable, say so at the outset and we will address it in the agreement rather than afterwards.

What do you need under NDA before starting?

For the feasibility study, an antigen sequence or structure is usually enough, and many prospects send that before any agreement. For a full campaign we will normally have an NDA in place, and for the verification route it is always in place first.

What happens when a campaign produces no binder

What happens if nothing binds?

It happens, and you will know by week twelve rather than later. Design campaigns have a per-attempt success rate rather than a guarantee, and some targets do not yield binders. The published evaluations show the same pattern. An independent evaluation of BindCraft, run on peptide binders, produced nanomolar binders against two targets and nothing detectable against two others (Filius et al., ACS Chemical Biology 2025, 20:2991-2998, DOI, reference located through PubMed). The authors of the same tool report experimental success rates ranging from roughly 10% to 100% depending on the target, for protein binders rather than antibodies (Pacesa et al., Nature 2025, 646:483-492, DOI, reference located through PubMed). We report the attrition across the whole funnel rather than only the successes, and the reasoning is set out in what a binding experiment proves.

Do you guarantee a binder?

No, and an antibody design service that guarantees one is telling you something the literature does not support. What we do instead is tell you before you commit whether the target is one we can work against, which is what the free feasibility study is for. We say no to targets, and that is the part that makes the yes worth something.

What if the candidates bind but do not work in our assay?

Binding is not function, and the gap between the two is the ordinary case rather than the exception. We validate binding to the purified target. Whether a binder blocks, inhibits, internalises or kills is tested in your assay, and it is the next stage of the programme rather than part of ours.

Receive a feasibility study on your antigen, for free

What an antibody design service does not cover

Do you immunise animals?

No, never, and for no species. Every sequence we deliver comes from computational design against a structure, which is the whole basis of the platform. It also means we have no reason to push a target towards a route that does not suit it, and we will say when a display method would serve you better, as covered in our comparison of phage display and computational design.

Do you give freedom to operate opinions?

No. That is legal work and it belongs to a qualified patent attorney. We compute and report sequence identity to the claims your counsel identifies, and we design to the threshold they set. Deciding what a claim covers is their work, and the molecule is ours.

Do you run epitope mapping, functional assays or manufacturing?

No. We read structures and design against them, and we validate binding by cell-free expression and ELISA. Experimental epitope mapping, functional characterisation, cell line development and manufacturing are other providers' work, and we will say so rather than stretching the scope.

Do you only work on nanobodies?

No. Single-domain antibodies are often the better unit for design work, for the reasons given in our note on what a nanobody is, and conventional antibodies are a normal request. The comparison between formats is in VHH, scFv or IgG.

Starting with a free feasibility study

What is the free feasibility study?

An assessment of your target and the site you want engaged, returned as our view of what is achievable, including when the answer is that it is not. It costs nothing and it takes a day or two. It exists because running a campaign against an unsuitable target wastes your quarter and our reputation.

What do you need from me to run it?

The antigen sequence or a structure, and an indication of what the binder has to do. If you know the site, tell us. If you do not, that is part of what the study addresses.

What does a project cost?

It depends on the entry point and the scope, and we quote per project rather than publishing a price list. The feasibility study is free in every case, and the quotation follows it rather than preceding it.