Applications
Explore the obstacles behind binder discovery, with structural examples and detailed explanations. Target assessment connects these questions to your project.
Accessibility, immune tolerance, specificity and antigen quality create different obstacles to binder discovery. Each example shows a question we assess before choosing a design and testing strategy. Our work includes membrane and soluble proteins.
A, accessibility-limited
A small, recessed or partly covered epitope can limit which binder shapes reach it. We assess accessibility in the relevant protein structure and biological context.
Shown: a multi-pass membrane protein, where only the short extracellular loops are accessible in the native fold.
Also:
B, immune tolerance-limited
A protein conserved between human and mouse may elicit little immune response. Computational design starts from its structure, without relying on an animal to produce antibodies.
Shown: a protein conserved to the residue, since human and mouse calmodulin share one sequence.
Also:
C, discrimination-limited
Closely related proteins can share the same binding surface. We include the proteins to avoid in the assessment and computational comparison; selectivity is established by the agreed experimental tests.
Shown: two paralogues that differ at a handful of positions, β-actin and γ-actin.
Also:
D, antigen supply-limited
Misfolding, aggregation or poor expression can compromise a binding assay. A structure or model supports an initial assessment, while representative target material remains essential for experimental testing.
Shown: a flexible stretch held in the membrane, from the amyloid precursor protein.
Also:
A, accessibility-limited
A recessed site can constrain the size, shape and orientation of a binder. Some antibody formats and designed proteins can access such sites; suitability depends on the particular structure and experimental context.
A, accessibility-limited
Glycans change which surfaces are exposed and can differ between biological contexts. Structural illustrations help explain that uncertainty; they do not establish that a depicted surface is accessible in an experiment.
A, accessibility-limited
A binding partner can cover an epitope in the native complex. The assessment distinguishes the biological state of interest from the state represented by the available structure and assay material.
B, immune tolerance-limited
Some antigens make conventional immunisation unsuitable. The example illustrates why the discovery route and experimental scope require specialist review; it is not a claim that computational assessment removes those constraints.
B, immune tolerance-limited
Small antigens may produce a weak immune response and can require specialised discovery approaches. Computational methods offer another starting point, but epitope recognition and specificity still need experimental evidence.
C, discrimination-limited
Different conformations can expose different binding surfaces. Both experimental selection and computational design can address state preference, provided the relevant states are represented and the preference is tested.
C, discrimination-limited
Cleavage can expose a new terminus while leaving most of the sequence shared with the precursor. Distinguishing those forms requires a clearly defined specificity question and measurements against the relevant forms.
D, antigen supply-limited
An assembly that dissociates during purification may no longer represent the biological target. A structural model can inform an initial assessment, but experimental testing needs a system that preserves or reports the relevant assembly.
D, antigen supply-limited
A panel of binders may be useful when the assay needs to distinguish several surfaces or forms. Structural models can suggest candidate sites; the resulting binding sites and relationships require experimental confirmation.
We start with academic groups and biotech companies in the German-speaking region. What they share is a protein they need to detect, pull down or block in a specific experiment, and no reagent that does it.
Academic research groups
You work on a membrane protein or another hard-to-reach target, the commercial reagents fail the control you care about, and the group has neither the structural work nor the screening capacity to run a design campaign of its own.
Biotech companies
You need a research reagent against a defined target, or a written feasibility judgement before a programme is committed to it, with scope, acceptance criteria and rights settled before the work starts.
Teams with data and a question
You hold sequence or proteomics data and a question about substrates, cleavage sites or a protein family, but building and validating the analysis would cost more time than the answer is worth.
You get a straight answer about what is realistic.
Research use only. Scope, criteria and deliverables are agreed per project.
