Recover missing potency
Find interactions that a lead is not yet using and prioritize modifications that can strengthen target binding.
- Binding-site hotspots
- Water-network opportunities
- Strain and pose quality
Lead optimization
Potency & Selectivity Optimization
Gaining potency is only half the job—if your molecule binds related proteins, you inherit their side effects. This service couples affinity improvement with off-target profiling, so each design round widens the activity gap in your favor.
AffinitySelectivityFEPHotspot analysisOff-target profiling
What we optimize
Improve the activity gap, not potency alone
Remove off-target binding
Compare the target with homologs, isoforms, or safety targets to locate exploitable structural differences.
- Panel selection
- Binding-mode comparison
- Liability flags
Rank analogs before synthesis
Use methods suited to the chemistry and decision—from docking and MM/GBSA to FEP—to focus chemistry resources.
- Matched-pair analysis
- Relative affinity ranking
- Uncertainty-aware shortlist
Explain confusing SAR
Connect assay trends with binding poses, residue contacts, protonation, and protein flexibility to generate testable hypotheses.
- Activity cliffs
- Inconsistent analog trends
- Assay follow-up plan
Optimization workflow
Turn assay data into the next compounds to make
01
Define the activity window
Standardize assay data, identify the target potency goal, select meaningful off-targets, and document chemistry constraints.
02
Explain the current SAR
Assess structures and poses, map hotspots, compare active and inactive analogs, and identify the interactions driving each trend.
03
Design and compare changes
Enumerate feasible modifications and apply docking, interaction analysis, MM/GBSA, or FEP according to the series and decision.
04
Prioritize the next test
Balance target affinity, selectivity risk, confidence, and chemical feasibility in a shortlist linked to a practical assay plan.
Continue with the right specialist service
These four closely related pages are listed in the provided website outline and can support a focused workstream or a larger lead-optimization program.
- 1Drug Optimization Service — coordinate multi-parameter lead improvement beyond potency and selectivity.
- 2Binding Free Energy Analysis Service — compare suitable analogs with physics-based relative affinity calculations.
- 3Multiple-target Virtual Screening Service — profile candidates across a defined panel of related targets.
- 4Drug Off-Target Effect Prediction Service — investigate broader target liabilities and prioritize follow-up testing.
How teams use it
Use the service at the point where your series is stuck
Widening the affinity window
You have a lead but need more potency against the target. We identify the modifications most likely to raise affinity and rank them before synthesis.
- Hotspot mapping
- FEP affinity ranking
- Modification proposals
- Predicted ΔΔG
Removing selectivity liabilities
Your series is potent but hits a related protein. We diagnose where selectivity is lost and propose changes that spare the off-target while preserving activity.
- Off-target profiling
- Selectivity matrix
- Interface differences
- Corrective designs
New cores, better profile
You need a fresh scaffold that keeps potency and improves selectivity or IP. We design and screen alternative cores against both your target and the off-target panel.
- Scaffold enumeration
- Docking & scoring
- Selectivity screening
- Prioritized shortlist
Why optimize potency and selectivity together?
A modification that improves target affinity may also strengthen binding to a related protein. We therefore compare every proposal against the target goal and the selected off-target panel, then rank designs by the size and confidence of the predicted activity window.
TraceableEvery proposal tied to structural evidence
Panel-awareTarget and off-targets compared together
PrioritizedRecommendations ordered for synthesis
IterativeNew assay data updates the model
Choose the depth that matches your decision
The method should match the evidence available and the cost of a wrong choice. We do not force every project into an expensive free-energy workflow.
1Focused SAR and hotspot interpretation
2Comparative docking and selectivity profiling
3Free-energy-guided design cycle
What should you send us?
A useful first discussion does not require a perfect dataset. Send the evidence you trust and flag what is uncertain.
- Chemical structures with compound identifiers and stereochemistry
- Potency and selectivity values with units, assay type, and conditions
- Target, isoforms, homologs, and known safety targets of concern
- Experimental structures, predicted models, or relevant PDB entries
- Chemistry constraints, compounds already attempted, and decision timeline
How do we keep recommendations realistic?
Before ranking designs, we check assay comparability, structure quality, ligand states, binding-pose plausibility, chemical coverage, and model convergence.
- Separate comparable assay values from incompatible endpoints
- State when structure or protonation uncertainty changes the conclusion
- Use matched analogs for quantitative comparisons where possible
- Report disagreement between methods instead of hiding it in one score
Important: computational predictions prioritize experiments; they do not guarantee biochemical potency, cellular selectivity, safety, or clinical performance.
Decision-ready deliverables
What will your team receive?
Clear decision statement
The project question, recommended path, confidence level, and key limitations in plain language.
Interaction analysis
Binding modes, hotspots, activity cliffs, and structural interpretation of the available assay data.
Ranked modifications
Proposed compounds with target rationale, selectivity hypothesis, method, score, and synthesis priority.
Target vs off-target
A side-by-side comparison across selected homologs, isoforms, or liability targets.
Review-ready figures
Annotated poses, interaction maps, comparison plots, and structural explanations for team discussion.
Assay roadmap
Experiments that discriminate between design hypotheses and feed the next optimization cycle.
Published data
What evidence supports this optimization strategy?
Study [1] · Free-energy calculations
FEP ranks binding potency before synthesis
Wang L, Wu Y, Deng Y, et al. Journal of the American Chemical Society. 2015;137(7):2695–2703.The authors demonstrated that a modern free-energy calculation protocol could reliably predict relative ligand binding potency in prospective drug discovery, validating FEP as a tool to prioritize modifications before committing chemistry.
Service implication: our design stage ranks modifications by predicted ΔΔG so your synthesis effort targets the highest-probability affinity gains.
Design → FEP → prioritizeOriginal schematic
ModificationsEnumerate candidate changes to the lead.
→
FEP calculationPredict relative binding affinity.
→
PrioritizationRank the best-affinity proposals first.
FEPΔΔGAffinityRanking
Study [2] · Selectivity profiling
Broad profiling reveals where selectivity is lost
Davis MI, Hunt JP, Herrgard S, et al. Nature Biotechnology. 2011;29(11):1046–1051.The authors profiled kinase inhibitors across a large panel of kinases, showing that many compounds engage far more off-targets than intended—and that systematic profiling exposes those selectivity liabilities early.
Service implication: we apply this same profiling logic in silico, screening proposals against related proteins to flag selectivity risk before synthesis.
Panel → profiling → flagOriginal schematic
Target panelAssemble related proteins and isoforms.
→
ProfilingScore each candidate across the panel.
→
Selectivity flagIdentify and avoid off-target hits.
PanelProfilingOff-targetSelectivity
References
- Wang L, Wu Y, Deng Y, Kim B, Pierce L, Krilov G, Lupyan D, Robinson S, Dahlgren MK, Greenwood J, et al. Accurate and reliable prediction of relative ligand binding potency in prospective drug discovery by way of a modern free-energy calculation protocol and force field. J Am Chem Soc. 2015;137(7):2695–2703. https://doi.org/10.1021/ja512751q
- Davis MI, Hunt JP, Herrgard S, Ciceri P, Wodicka LM, Pallares G, Hocker M, Treiber DK, Zarrinkar PP. Comprehensive analysis of kinase inhibitor selectivity. Nat Biotechnol. 2011;29(11):1046–1051. https://doi.org/10.1038/nbt.1990
Project questions
What do teams ask before starting?
The right scope depends on the quality of the structures, the consistency of the assays, and the decision your team needs to make.
We optimize the difference between target and off-target binding rather than potency alone. SAR and hotspot analysis identify useful modification sites, free-energy calculations prioritize likely affinity gains, and multi-target profiling flags designs that may narrow the selectivity window.
No. Experimental structures are preferred when available, but a carefully assessed homology or AlphaFold model can support an initial study. We document model uncertainty and avoid overstating predictions that depend on poorly resolved regions.
FEP or MM/GBSA can compare suitable analogs and prioritize modifications before synthesis. Method choice depends on series similarity, structure quality, chemistry coverage, and the decision required; results are treated as relative predictions with stated uncertainty.
We select a biologically relevant panel of homologs, isoforms, or known safety targets and compare candidate binding across that panel. The result highlights likely selectivity liabilities and structural differences that can guide redesign.
Useful inputs include chemical structures, assay values with units and conditions, target and off-target identities, available structures or models, known liabilities, and the decision deadline. Sparse data can still support a focused hypothesis-driven study.
Yes. We can begin with a focused structural interpretation and uncertainty-aware ranking, then recommend the most informative compounds or assays to close data gaps. Larger, consistent datasets support more quantitative modeling.
You receive a ranked design table, structural and energetic rationale, target-versus-off-target comparison, uncertainty notes, and a validation plan linking each recommendation to the experiment that can test it.
Start a project
Which compounds should your team make next?
Share your lead chemistry, available potency and selectivity data, the proteins you need to avoid, and the decision your team is facing. CD ComputaBio will propose a fit-for-purpose plan that separates essential analyses from optional deeper work.
Online Inquiry
Submit your project details below, and our team will respond within 24 hours.
x
Need help getting the data you need?
Talk to our technical team about your project!
I Want To Talk