Overview
What this category covers
Structure-based design is most valuable as a loop rather than a single calculation. This category covers running that loop — designing, prioritising and re-analysing each cycle as assay and structural data return — alongside the detailed analysis of co-crystal structures and the specialised problem of allosteric site discovery and modulator design.
What we deliver
▸Integrated design–make–test–analyse programmes on two to four week cycles
▸Detailed co-crystal analysis: interactions, water networks, strain and growth vectors
▸Allosteric site discovery through simulation, with communication pathway analysis
▸Method selection matched to decision value, from docking through to FEP
▸Cycle reports comparing what was predicted with what was observed
Who this is for
▸Chemistry teams without an in-house computational chemistry function
▸Programmes with structural data that is not being fully exploited
▸Projects seeking selectivity within a conserved target family
▸Series that have stalled and need their structural basis re-examined
How engagements in this category are run
These engagements are structured as rolling programmes with a defined question per cycle and a fixed handover point for your assay and structural data. Compute is allocated by decision value: cheap methods triage, expensive methods decide. Each cycle closes with a comparison of prediction against observation, and where the model failed we diagnose why rather than quietly retune it. Programmes typically become genuinely predictive after two or three cycles of feedback.
Services
3 computational services in this category
G1
Integrated SBDD Program (AI + Structure Loop)
Closed-loop DMTA: AI prediction → compound design → synthesis → X-ray/Cryo-EM structure → re-prediction. 2–4 week cycles.
Learn MoreG2
Protein–Ligand Co-Crystal Structure Analysis
Binding mode determination, water network analysis, H-bond geometry, hydrophobic contacts, vector analysis for analog design.
Learn MoreG3
Allosteric Drug Design
Allosteric pocket identification via MD, allosteric communication pathway analysis, modulator design (activators and inhibitors).
Learn MoreIn Practice
What a project in this category looks like
Three shapes we are asked for most often. Scope is always set after a feasibility review, so timelines below are typical rather than quoted.
Running the computational half of a hit-to-lead programme
The situation
A chemistry team has structures and assays but no in-house computational chemistry function.
What we do
Each cycle: analyse the latest structures and SAR, design and prioritise compounds, apply docking, end-point free energy or alchemical calculations according to the decision cost, and close the loop by comparing prediction against measurement.
What you receive
Prioritised designs each cycle with structural rationale, updated binding-mode models, and a cycle report stating where the model was right and where it was not.
Typical timeline: 2–4 week cycles, typically 3–12 months
Extracting design direction from a new co-crystal structure
The situation
A structure has just been solved and the team wants everything actionable out of it.
What we do
Interaction geometry assessment, water network classification, ligand strain analysis, sub-pocket and vector mapping, and comparison across the series or against paralogues for selectivity.
What you receive
An interaction and water map, a list of growth vectors with what each could gain, and specific design suggestions tied to the structure.
Typical timeline: 1–2 weeks
Finding selectivity in a conserved family
The situation
The orthosteric site is nearly identical across the family and selective inhibition has not been achievable.
What we do
Simulation-based search for allosteric and cryptic sites, communication pathway analysis to establish a mechanistic case, conservation comparison across the family, then screening or design against the validated site.
What you receive
Ranked candidate sites with their evidence, a selectivity assessment per site, designed or screened ligands, and the experiment that would confirm the mechanism.
Typical timeline: 8–12 weeks
Questions
Asked before we scope this work
Do you need co-crystal structures to be useful?
They help considerably, but programmes routinely start from a predicted or homology model. What matters is the quality of the binding site region, which we assess separately from overall model quality.
How soon does the modelling become predictive?
Usually after two or three cycles of feedback, once the model has been corrected against real assay results. We report predictive accuracy per cycle so the trend is visible rather than asserted.
Do you perform synthesis and assays?
No. ICDMO is a computational CDMO. We design, prioritise and analyse; the laboratory work is carried out by your team or a qualified partner, and we help plan and interpret it.
What Published Work Reports
The results below were published by independent groups and are cited to show what these computational approaches have achieved in the literature. They are not ICDMO project results, and outcomes vary substantially between targets, datasets and programmes. We do not promise comparable results for any particular project.
Structure prediction now exceeds classical docking on independent benchmarks
On a held-out protein–ligand benchmark, deep-learning complex prediction placed roughly three quarters of ligands within 2 Å of the crystallographic pose, outperforming established docking programs on the same set. For design programmes this changes what can be attempted before a co-crystal structure exists — while leaving binding-mode certainty, and therefore the most consequential chemistry decisions, dependent on experimental structures.
Abramson et al., Nature, 2024 ↗Free energy methods can resolve the differences that decide a synthesis
Published benchmarks on congeneric series report errors near 1 kcal/mol, sufficient to rank analogues that differ by around tenfold in potency. In an integrated design loop this is the calculation reserved for decisions where making the wrong compound costs more than the computation.
Protocol and benchmark studies, Scientific Reports, 2019 ↗More
Other service categories
Target Identification & Validation5Virtual Screening (Small Molecule)3Small Molecule Design & Optimization2ADMET & Drug-Likeness Prediction4Molecular Dynamics & Simulation3AIDD — Small Molecule Platform2Bioinformatics & Multi-Omics Analysis11Antibody Design & Engineering (AI)5Peptide & Cyclic Peptide Drug Design4Protein Design & Engineering (AI)4ADC & Bioconjugate Design2Compound & Biological Library Design3
Bring us the scientific question, not a service code
Share your target, dataset, sequences or objective. We will recommend a scoped computational plan, the expected deliverables, a timeline and the validation options.
Request a Feasibility Review