Overview
What this category covers
Once a pocket is validated, the question changes from finding molecules to creating better ones. This category covers generative design against a defined site and the specialised design problem of bifunctional degraders, where a linker, two binding events and a ternary complex must work together. Both services are built around filtering and synthesis realism, because generated structures are cheap and makeable candidates are not.
What we deliver
▸Novel molecules generated in the three-dimensional context of your binding site
▸Multi-property optimisation covering potency proxies, ADMET and novelty
▸Retrosynthetic assessment and route sketches for the recommended designs
▸PROTAC and molecular glue designs with modelled ternary complexes
▸Linker enumeration, cooperativity assessment and permeability prediction for degraders
Who this is for
▸Programmes where screening has not produced usable chemical matter
▸Teams designing around a competitor patent position
▸Groups converting an established binder into a degrader
▸Projects needing to expand chemical space around fragments or weak hits
How engagements in this category are run
Design projects run as rounds rather than as a single delivery. A first round establishes what the pocket rewards and what your chemistry can make; measured results from that round then reshape the filters and the generation constraints for the next. Synthetic accessibility and property criteria are agreed before generation starts, so the delivered set is weighted toward compounds a chemist will accept. For degrader work the first round is usually about linker geometry rather than potency, and we scope it that way.
Services
2 computational services in this category
C1
De Novo Small Molecule Design
AI generative models for novel drug-like molecule generation targeting specific binding sites. Multi-property optimization (potency, ADMET, IP novelty).
Learn MoreC6
PROTAC / Molecular Glue Design
Bifunctional degrader design: E3 ligase selection, ternary complex modeling, linker optimization, cell permeability prediction, cooperativity scoring.
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.
Growing fragment hits into lead-like compounds
The situation
Fragment screening produced confirmed binders with crystallographic poses, and the team needs to grow them without losing the observed binding mode.
What we do
The crystallographic pose is held fixed while generative growing and linking explore the surrounding sub-pockets, then everything is filtered for synthetic accessibility, properties and novelty and re-docked.
What you receive
Twenty to a hundred designs with poses, property profiles, retrosynthetic assessment and a proposed synthesis order.
Typical timeline: 5–8 weeks
Designing around a competitor patent
The situation
The chemistry works but the scaffold is covered, and the team needs different chemical matter that keeps the interactions responsible for potency.
What we do
The pharmacophore is extracted from the existing series, generation is constrained to preserve it, and output is filtered against the patented chemotypes with similarity and novelty analysis.
What you receive
Scaffold-hopped designs with novelty distance reported, interaction analysis showing which contacts are preserved, and synthesis routes for the recommended set.
Typical timeline: 6–9 weeks
Turning a binder into a degrader
The situation
A validated binder exists, occupancy-based inhibition is insufficient, and the team wants to know whether a degrader is feasible before committing chemistry.
What we do
Target degradability assessment, E3 ligase shortlisting, exit vector analysis on the warhead, ternary complex modelling across linker length and rigidity, cooperativity assessment and permeability prediction.
What you receive
Fifteen to sixty ranked degrader designs with ternary models, predicted properties and a testing order that makes the first synthesis round systematic.
Typical timeline: 6–10 weeks
Questions
Asked before we scope this work
How many designs are worth making in the first round?
Most teams synthesise fifteen to thirty. The purpose of the design work is to make that round test distinct hypotheses rather than variations of one idea.
Will a chemist accept the proposals?
Every delivered design is accessibility-scored and route-assessed, and we would rather hand over thirty makeable compounds than three hundred that look good on screen. Some will still be rejected, which is normal.
Who owns the designs?
You do. Deliverables belong to the client under our standard terms, and client-specific designs are never reused on another project.
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.
AI-designed molecules have now produced clinical-stage compounds
A compound designed by generative methods against an AI-nominated target completed a randomised Phase 2a trial in idiopathic pulmonary fibrosis, reporting a mean forced vital capacity change of +98.4 mL in the 60 mg once-daily arm against −20.3 mL for placebo over 12 weeks in 71 patients. The result shows that generative design can produce molecules that survive to clinical testing; it does not show that any individual generated series will.
Nature Medicine, 2025 ↗More
Other service categories
Target Identification & Validation5Virtual Screening (Small Molecule)3ADMET & Drug-Likeness Prediction4Molecular Dynamics & Simulation3AIDD — Small Molecule Platform2Structure-Based Drug Design (SBDD)3Bioinformatics & 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