Diffusion‑guided linker generation—backed by predictive biology and assays.

A curated ADC knowledge base, augmented with in-house validation, powering computational design and wet-lab iteration from target to candidate.

Request access Explore AMM‑INVITRO

Designed to preserve familiar, manufacturable conjugation motifs while exploring novel chemistry where it matters.

Platform

A single system for ADC intelligence.

ADCpedia combines structured ADC knowledge with experimental feedback to accelerate discovery decisions—especially linker innovation.

Comprehensive ADC database

Curated entities across targets, antibodies, linkers, payloads, and design attributes—structured for discovery workflows.

Assay‑augmented

In‑house assays provide validation signals to calibrate models and strengthen real‑world predictivity.

Unified stack

Predictive modeling and wet‑lab feedback work together—closing the loop from hypothesis to candidate.

Workflow

Discovery that closes the loop.

Predict activity, expression, binding, and in vivo response—then guide design toward candidates that fit your target biology.

Multi‑model predictions across in vitro, in vivo, expression, affinity, and DAR strategy.
Generative design for linkers, payload refinement, and antibody components.
Guided optimization across multiple objectives (constraints + biology).

Want validation snapshots? We share deeper details under access.

Models

A suite of predictive and generative modules for ADC design.

Built to support practical decision points—from expression to efficacy to conjugation constraints.

Predict

AMM‑INVITRO

In vitro response likelihood across cell lines.

Output: activity likelihood + context

Predict

AMM‑INVIVO

In vivo response likelihood in PDX/CDX models.

Output: tumor response likelihood

Predict

GENCEP

Protein expression intensity across cancer/normal cell lines and xenografts.

Output: expression intensity estimates

Predict

Affinity

Antibody–antigen binding affinity prediction for binder selection and refinement.

Output: affinity signals

Predict

DARwin

DAR strategy optimization for a given antibody–linker–payload combination.

Output: recommended DAR range

Core

Linker Diffusion Engine

Diffusion‑based linker generation guided by predictive signals—designed to respect practical conjugation constraints.

Output: ranked linker candidates

Generate

Payload Optimization

Property‑guided payload refinement for improved fit to your biology and chemistry constraints.

Output: optimized payload candidates

Generate

Antibody Designer

Generative VH/VL + CDR design to explore novel binders when needed.

Output: candidate sequences (under access)

Diffusion‑guided linker generation—where prediction becomes design.

A diffusion engine that proposes linkers aligned to constraints and biology—then ranks them for decision‑making.

Stability Hydrophilicity Cleavability Exposure DAR Binding Expression Activity

Step 1

Conditioning

Define the target context—antigen format, payload class, conjugation site, and design constraints.

Step 2

Predictive guidance

Multi‑signal guidance surfaces candidates likely to work in the biology you care about—without turning the page into a methods paper.

Step 3

Diffusion generation

A diffusion model explores linker space while honoring hard constraints—steering toward designs that balance chemistry and biology.

Step 4

Ranked proposals

Linker proposals preserve familiar conjugation motifs while introducing novelty elsewhere—ranked by predicted fit.

Assays that strengthen the signal.

ADCpedia is augmented with in‑house assays designed to calibrate predictions and reduce design risk—helping teams converge faster.

Use cases

Designed for real ADC discovery decisions.

High‑level workflows that map to how teams actually make tradeoffs.

Linker innovation

Explore novel linker space without losing practical conjugation constraints or manufacturability considerations.

Target expansion

Support conventional and non‑conventional antigens with expression‑aware modeling across cell lines and xenografts.

DAR strategy

Choose DAR aligned to payload chemistry, exposure, and biological context—then validate with assays.

Binder optimization

Use binding and expression signals to refine antibody choices and guide when to redesign.

Try AMM‑INVITRO.

Explore predicted likelihood of in‑vitro activity across cell lines in an interactive playground.

Open AMM‑INVITRO Playground Request access

Reference: [Authors, Year]. “AMM‑INVITRO: …” (DOI)