Chemical proteomics helps drug discovery teams understand how small molecules interact with proteins in biologically relevant systems. By measuring cellular target engagement and selectivity across many endogenous proteins at the same time, it can reveal whether a candidate reaches its intended target, which additional proteins it engages and how its profile changes with biological context.
This evidence is particularly valuable during hit-to-lead, lead optimisation and preclinical development, when teams must decide which compound to advance and which risks require further investigation. Chemical proteomics does not replace biochemical, functional, pharmacokinetic or safety studies. Its value is that it fills an important evidence gap between biochemical potency and compound behaviour in a biological system.
In one sentence: Chemical proteomics supports better candidate selection by measuring the interaction landscape of a compound in a biological context, not only its activity against an isolated target.
What is chemical proteomics?
Chemical proteomics is the use of chemical probes, protein enrichment and mass spectrometry to study how small molecules interact with proteins across a proteome or defined protein family. In drug discovery, the approach can be used to identify targets, confirm target engagement, assess selectivity and investigate mechanism of action.
Target engagement means that a compound physically interacts with its intended protein in the experimental system. This is distinct from biochemical inhibition, downstream pathway modulation or a phenotypic response. Those measurements answer related questions, but they are not interchangeable.
| Question | Evidence needed | What it contributes |
| Can the molecule inhibit the isolated target? | Biochemical potency assay | Early potency and structure–activity information |
| Does the molecule engage the target in cells? | Cellular target-engagement measurement | Evidence that the compound reaches and binds the target in cellular context |
| What else does the molecule engage? | Broad selectivity profiling | Intended and unexpected protein interactions |
| Does engagement change biology? | Functional, pathway or phenotypic assay | Evidence connecting engagement with downstream effect |
| Can relevant exposure be achieved safely? | Pharmacokinetic and safety studies | Evidence for dose, exposure and development feasibility |
Why biochemical potency alone is not enough
Biochemical assays remain essential for testing potency against a purified or recombinant target. They are controlled, scalable and well suited to comparing compounds. However, an isolated protein system does not reproduce every factor that influences binding in a cell.
Cell permeability, intracellular compound concentration, endogenous protein abundance, protein conformation, binding partners, cofactors and competition with cellular molecules can all affect target engagement. A compound with strong biochemical potency may therefore show weaker cellular engagement than expected. Conversely, an interaction that appears modest in an isolated system may become more relevant in a particular cellular context.
The practical question is not whether biochemical or cellular evidence is superior. Drug discovery teams need to understand what each assay measures, then combine the results to reduce uncertainty around the decision at hand.
The three decisions chemical proteomics can strengthen
1. Does the candidate engage its intended target in a relevant system?
A compound must reach and bind its target before the intended mechanism can occur. Cellular target-engagement data provide direct evidence of that interaction under the selected experimental conditions. When several candidates have similar biochemical potency, differences in cellular engagement can help explain why their functional profiles diverge.
2. Is the candidate sufficiently selective for the programme?
Selectivity is not a universal property that can be reduced to one label. It depends on concentration, exposure time, biological model, the proteins that are expressed and the consequences of each interaction. Broad endogenous protein profiling helps teams compare the intended interaction with the wider engagement profile rather than assessing one off-target at a time.
3. Which candidate has the strongest overall evidence package?
Candidate prioritisation rarely depends on a single result. Chemical proteomics contributes quantitative engagement and selectivity data that can be considered alongside potency, functional activity, physicochemical properties, pharmacokinetics and safety. The most useful output is therefore not simply a ranked protein list. It is an interpreted comparison linked to the programme’s advancement criteria.
Where chemical proteomics fits in the discovery process
| Development stage | Typical decision | Potential contribution of chemical proteomics |
| Hit-to-lead | Which chemical series deserve further investment? | Confirm cellular engagement, identify early selectivity liabilities and compare engagement profiles across representative compounds. |
| Lead optimisation | Which changes improve the overall candidate profile? | Track intended-target engagement and off-target activity across concentration ranges while medicinal chemistry optimises the series. |
| Candidate selection | Which molecule should advance? | Provide a comparative engagement and selectivity layer for the multidisciplinary evidence package. |
| Preclinical development | What mechanism or selectivity risks need further study? | Test engagement in more relevant models and support focused follow-up experiments. |
| Translational research | Can engagement be assessed closer to disease biology? | Where the method and sample type are suitable, compare engagement across increasingly relevant biological systems. |
How to choose the right biological model
The most sophisticated model is not automatically the most informative. The model should reflect the decision, target expression, compound properties and development stage. A simpler system may be appropriate for comparing a medicinal chemistry series, while a more complex model may be needed to investigate whether engagement is maintained in disease-relevant biology.
| Model | Best suited to | Planning consideration |
| Cell lines | Controlled comparison of compounds in living cells | Confirm endogenous target expression and whether the cell line reflects the biology relevant to the programme. |
| Primary cells, including PBMCs | Engagement in native human cell populations | Donor variation, cell availability and target abundance influence design and interpretation. |
| Organoids | Engagement in a three-dimensional, disease-relevant model | Model complexity, heterogeneity, material requirements and analytical depth should be justified by the decision. |
| In vivo tissues | Engagement under physiological exposure and tissue context | Tissue distribution, sampling time, dose and target expression become central to interpretation. |
For Omivera, the approved source material identifies cell lines, peripheral blood mononuclear cells (PBMCs), organoids and in vivo models as relevant model categories. The current operational availability and validation status of each category should be confirmed before publication or inclusion in a client proposal.
A practical framework for candidate prioritisation
Chemical proteomics data become decision-ready when the analysis is designed around a comparison that the project team must make. A practical review should consider five dimensions together:
- Intended-target engagement: Does each candidate engage the target at concentrations and exposure conditions relevant to the programme?
- Engagement profile: Which additional endogenous proteins are engaged, and how does this change with concentration?
- Biological context: Is the profile consistent across the models needed to support the next decision?
- Functional alignment: Do engagement patterns agree with pathway, phenotypic or pharmacological evidence?
- Development relevance: Which findings are differentiating, manageable or significant enough to change the candidate choice?
This framework prevents a common interpretation problem: treating the compound with the fewest measured interactions as automatically preferable. Some multi-target profiles may be intentional or therapeutically useful, while one unanticipated interaction may carry more risk than several benign ones. The assessment must remain connected to mechanism, exposure and disease biology.
How a chemical proteomics study should be planned
A strong study begins with the decision, not the technology. Before samples are generated, the scientific team and the chemical proteomics partner should agree on the comparison, the compounds, the biological system, exposure conditions and the evidence that would change the programme’s direction.
- Define the decision. Specify whether the study should confirm engagement, compare candidates, investigate selectivity or explore mechanism.
- Select compounds and controls. Include reference compounds or inactive controls where they improve interpretation.
- Choose the biological context. Confirm target expression and match the model to the programme stage.
- Design concentration and exposure conditions. Ensure the design can distinguish meaningful engagement patterns rather than producing a single snapshot.
- Generate and quality-control the proteomics data. Analytical reproducibility and transparent data processing are prerequisites for comparison.
- Interpret results with orthogonal evidence. Review engagement alongside biochemical, functional, pharmacological and safety findings.
- Define follow-up actions. Translate observations into candidate, chemistry or experimental decisions.
What can chemical proteomics reveal that targeted assays may miss?
A targeted assay can quantify one predefined interaction with high focus. Chemical proteomics can extend the view across many endogenous proteins and reveal interactions that were not selected in advance. This makes the approach useful when selectivity is uncertain, when compounds show unexplained phenotypes or when a team needs a broader mechanism-of-action hypothesis.
Breadth does not remove the need for targeted validation. Proteins may be absent, expressed below detection limits or inaccessible to the probe used in the workflow. Unexpected findings should therefore be treated as evidence for follow-up, not as definitive proof of biological consequence.
CellEKT: endogenous kinome target-engagement profiling
CellEKT, short for Cellular Endogenous Kinase Targeting, is Omivera’s chemical proteomics workflow for profiling the cellular target engagement of kinase inhibitors. The workflow uses broad-spectrum kinase probes and mass spectrometry to assess engagement of endogenously expressed kinases.
A peer-reviewed 2025 study authored by researchers from Leiden University and collaborators reported target-engagement profiles for covalent and non-covalent kinase inhibitors across more than 300 kinases. Results were expressed as half-maximal inhibitory concentration (IC50) values and evaluated using orthogonal approaches including phosphoproteomics and NanoBRET. The publication provides the principal evidence base for CellEKT’s use in broad endogenous kinome profiling.
For drug discovery teams, the relevant question is how this evidence can support a specific programme. Depending on study design, CellEKT data may help compare kinase-inhibitor candidates, investigate intended and unexpected engagement, and guide follow-up during lead optimisation or preclinical development.
Limitations and interpretation boundaries
Chemical proteomics is most useful when its boundaries are understood before the study begins.
- Coverage is method-dependent. A protein must be expressed, accessible to the probe and detectable within the analytical workflow.
- Target engagement is not the same as functional inhibition. Binding should be connected to downstream biology with appropriate orthogonal studies.
- An IC50 from a competitive engagement experiment is conditional on the assay design. It should not be treated as a universal compound constant.
- Model relevance matters. Results from one cell line cannot automatically be generalised to primary cells, organoids, tissues or patients.
- Broad datasets still require prioritisation. Statistical significance, effect size, biological relevance and development exposure all affect interpretation.
- Chemical proteomics complements rather than replaces pharmacokinetic, efficacy, toxicology and clinical evidence.
Questions to ask before commissioning a study
- Which drug discovery decision should the data support?
- Which targets and protein families are expected to be measurable in the proposed model?
- What concentration range, exposure time and controls are needed?
- How will biological and technical replicates be handled?
- How are engagement curves, selectivity profiles and uncertainty reported?
- Which findings will require orthogonal validation?
- How will the results be integrated with medicinal chemistry, functional biology and development criteria?
Better data are valuable when they lead to a clearer decision
Chemical proteomics gives drug discovery teams a broader and more biologically relevant view of compound–protein interactions. Its contribution is strongest when the study is designed around a real candidate-selection question, uses an appropriate biological model and is interpreted together with orthogonal evidence.
For small-molecule programmes, this can reduce uncertainty around intended-target engagement, selectivity and mechanism before the next major investment. It does not eliminate development risk, but it can make the basis for advancing, redesigning or deprioritising a candidate more explicit.
Primary CTA: Discuss your drug discovery programme with Omivera to determine whether cellular target-engagement and selectivity profiling can support your next candidate decision.