Bio Protac Degraders

Version Compatibility

Reference examples tested with: PRosettaC (web service), DeepTernary research code, AlphaFold3, Boltz-1 / Boltz-2, RDKit 2024.09+, OpenMM 8.1+ (for ternary MD).

Before using code patterns, verify installed versions match. If versions differ:

• Python: pip show <package> then help(module.function) to check signatures

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

PROTAC and Bivalent Degrader Design

Design bifunctional molecules (PROTACs) that recruit an E3 ubiquitin ligase to a target protein, inducing target ubiquitination and proteasomal degradation. PROTACs differ from traditional drugs: a productive ternary complex (target + PROTAC + E3) is required, not just target binding. The modality has produced clinical programs, but their development and regulatory status changes rapidly and must be checked from current sources. PROTAC design balances target ligand binding, E3 ligand binding, linker geometry (length, rigidity, chemistry), cooperativity, dose-dependent ternary-complex formation, and cell permeability. Negative cooperativity and the high-concentration hook effect are distinct phenomena, although cooperativity can influence the dose-response profile.

For target ligand design, see chemoinformatics/virtual-screening and chemoinformatics/admet-prediction. For linker-only enumeration, see chemoinformatics/reaction-enumeration. For generative linker design, see chemoinformatics/generative-design.

E3 Ligase Choice

Decision: Select an E3 recruiter using evidence for ligand availability, target/E3 geometry, cellular expression, neosubstrate liabilities, and the intended biological system. CRBN and VHL are common starting points with extensive published examples, but neither is a universal first choice.

Linker Design Principles

Linkers tune ternary complex geometry and stability. The ranges below are exploratory starting points, not validated acceptance criteria:

Critical: The "Goldilocks linker length" is target-specific. Too short can create a ternary clash; too long can impose an unfavorable entropic cost or permit unproductive geometries. Enumerate a series around the geometry supported by the binary structures rather than assuming a universal optimal range.

Decision Tree by Scenario

Ternary Complex Prediction Tools

Decision: Use a PROTAC-specific method such as PRosettaC for first-pass ternary modeling. AlphaFold3 or Boltz can provide unrestrained whole-complex predictions, but should be benchmarked on relevant ternary complexes. DeepTernary is released research code rather than a hosted API; validate it against relevant structures before prospective ranking.

Cooperativity (Alpha)

Cooperativity quantifies how the ternary complex stabilizes (or destabilizes) the binary binding:

alpha = (Kd_binary,target) / (Kd_ternary,target)

• alpha > 1: positive cooperativity (ternary stronger than binary)
• alpha = 1: no cooperativity (independent binding)
• alpha < 1: negative cooperativity (mutual destabilization)

Positive cooperativity can favor ternary-complex formation, but the preferred alpha is system- and assay-dependent and does not alone establish degradation efficacy. Alpha must be measured from binary and ternary binding experiments; a predicted structure does not directly provide it.

Measure with ITC (isothermal titration calorimetry) or SPR/BLI titrations of binary vs ternary.

DC50 / Dmax Characterization

In cellular assays:

• DC50: PROTAC concentration for 50% degradation (analogous to IC50)
• Dmax: maximum fraction degraded at any concentration

Hook effect: at high PROTAC concentrations, binary complexes (PROTAC-target alone, PROTAC-E3 alone) dominate, and ternary complex formation drops. Dose-response curves are bell-shaped.

Ternary Complex Modeling Workflow

Goal: Predict 3D structure of target-PROTAC-E3 ternary complex.

Approach:

1. Start with binary co-crystals: target + target-ligand pose; E3 + E3-ligand pose
2. Connect via linker enumeration (combinatorial)
3. Score by geometric feasibility (linker length, no clashes)
4. Refine with energy minimization

# Pseudo-code workflow
def predict_ternary(target_pdb, target_ligand_sdf,
                    e3_pdb, e3_ligand_sdf, linker_smiles):
    # 1. Place binary complexes in same coordinate frame
    # 2. Enumerate linker connectivity from target-ligand exit vector to e3-ligand entry vector
    # 3. Score by total linker length, RMSD to expected geometry
    # 4. Apply a documented refinement protocol and test convergence
    return ternary_poses

For a production workflow, use PRosettaC or provide both proteins and the complete PROTAC as components of an unrestrained AlphaFold3 input. AlphaFold3 does not expose arbitrary chain-chain distance restraints; compare predicted interfaces and confidence with known complexes or a PROTAC-specific method.

Linker Geometry Assessment

from rdkit import Chem
from rdkit.Chem import AllChem

def attachment_distance(target_ligand, e3_ligand,
                        target_attachment_idx, e3_attachment_idx):
    """
    Measure an attachment-point distance after both ligands have been placed in
    the same ternary-complex coordinate frame.
    """
    p1 = target_ligand.GetConformer().GetAtomPosition(target_attachment_idx)
    p2 = e3_ligand.GetConformer().GetAtomPosition(e3_attachment_idx)
    return p1.Distance(p2)

An attachment-point distance does not map uniquely to a linker atom count: bond geometry, rigidity, branching, solvation, and the relative protein orientation all matter. Enumerate chemically synthesizable linker candidates, sample their conformers in the ternary geometry, and retain candidates that can connect without severe strain or clashes.

Generative Linker Design

REINVENT 4 can generate linkers, but it does not provide the ternary_score / deepternary interface shown in some informal examples. Export generated candidates, run an installed and validated ternary-prediction workflow separately, and then join the structural scores back to the candidates. Do not assume DeepTernary is a web API or a built-in REINVENT scoring component.

Per-Tool Failure Modes

PRosettaC -- inaccessible E3 in selected ligase

Trigger: Target's known binding mode incompatible with E3 ligase orientation.

Mechanism: The selected binary poses, exit vectors, linker conformations, or protein orientation may not support a compatible ternary geometry.

Symptom: Low ternary complex scores; high RMSD across replicates.

Fix: Try a different E3 such as CRBN or VHL and compare against experimentally resolved ternary complexes with compatible exit-vector geometry.

DeepTernary -- novel chemotype

Trigger: Target ligand or E3 ligand outside training distribution.

Mechanism: A ligand, linker, target, or E3 outside the method's validated domain may require extrapolation.

Symptom: Predicted ternary complex unrealistic.

Fix: Compare with an independently configured structural method and relevant known complexes; validate prospective ranking experimentally.