Ranomics
De novo binder scaffolds bound to a target protein, scored by JAX AF2 Initial Guess in the Ranomics in-house PXDesign pipeline
In-house pipeline

PXDesign de novo binder design

PXDesign is the de novo binder design pipeline Ranomics runs on its own wet-lab campaigns. RFdiffusion-style backbone generation, ProteinMPNN sequence design, and JAX AF2 Initial Guess scoring. The same algorithm our team uses when a client pays us to deliver binders.

Run it self-serve, then hand the candidates to our team for wet-lab validation when you find designs worth testing.

Based on Bennett et al., Nature Communications 2023, the AF2 Initial Guess methodology.

How it works

Target structure to ranked binder candidates

01

Define hotspots

Upload a target PDB or AlphaFold model. Specify hotspot residues that the designed binder must contact, or pull them from Epitope Scout output.

02

Generate backbones

RFdiffusion-style diffusion produces hundreds to thousands of novel scaffolds geometrically complementary to the hotspot patch.

03

Design sequences

ProteinMPNN assigns sequences to each backbone, with fixed positions at predicted contact residues to preserve binding geometry.

04

Score with Initial Guess

JAX AF2 Initial Guess refolds each binder-target complex and filters on interface pLDDT, PAE, and ipTM. Ranked CSV plus PDBs delivered.

Methodology

Four stages, one integrated run

PXDesign chains the four canonical stages of modern de novo binder design into one integrated run. Each stage passes its outputs directly to the next, with no manual handoff between tools.

Stage 1

Backbone generation

RFdiffusion-style hotspot-conditioned diffusion. Reverses a noise process to produce backbone coordinates that are geometrically complementary to the specified epitope. Scaffold length range and topology constraints are tunable per run.

Stage 2

ProteinMPNN sequences

Inverse-folding sequence design. Multiple sequences sampled per backbone at controllable temperature, with hotspot contact positions fixed to preserve the diffusion-defined binding geometry.

Stage 3

JAX AF2 Initial Guess

AlphaFold2 multimer refolds each binder-target complex using the designed structure as the initial coordinate guess. This is faster and more discriminative than de novo AF2 multimer prediction. The differentiation step from vanilla pipelines.

Stage 4

Interface filtering

Candidates are ranked on interface pLDDT, PAE at the predicted interface, and ipTM. Designs whose Initial Guess refold matches the original geometry rise to the top; designs that drift are discarded.

Output

Ranked candidate set

CSV of scored designs plus per-candidate PDB structures. Top candidates are pre-formatted for direct handoff into a Ranomics wet-lab Binder Pilot or AI Binder Sprint.

Beyond the algorithm

Why PXDesign is not a hosted academic tool

Initial Guess vs. vanilla AF2

Default AF2 multimer predicts complexes from scratch using MSAs and templates, which is slow and noisy when the binder is a never-seen-before designed sequence. Initial Guess seeds AF2 with the designed structure and asks: does the model agree this complex is real? The result is faster scoring and a success metric that correlates with wet-lab binding much more tightly than default multimer scores.

Wet-lab feedback loop

Every campaign Ranomics runs sends real binding data back into how we tune PXDesign, covering filter thresholds, scaffold lengths, and hotspot weighting. The version you launch on tools.ranomics.com is the version we calibrated against actual yeast display screens, not the public release weights frozen at paper publication.

Scales to full campaigns

Self-serve PXDesign runs on the same infrastructure we use internally. Hundreds of candidates per run. Need 10,000 to 50,000 designs across a full campaign? The same pipeline scales up inside an AI Binder Sprint with no rewrite and no re-validation.

When to use PXDesign

The algorithm we trust on paid campaigns

PXDesign exists because we needed a binder design pipeline we could defend at a wet-lab handoff, not a stitched-together set of GitHub repos with different conda environments and inconsistent scoring conventions. The Initial Guess scoring step is the load-bearing piece. It is what makes us willing to put a 100% binder guarantee on the AI Binder Sprint.

Run it self-serve when you want the same algorithm Ranomics uses internally, with results in a format that drops straight into a wet-lab campaign if the candidates are worth testing.

You want the same de novo binder algorithm Ranomics runs on its own paid wet-lab campaigns

Comparing AF2 Initial Guess scoring against your existing RFdiffusion plus default AF2 multimer pipeline

Need ranked binder candidates against a target you already have a PDB or AlphaFold model for

Planning a Binder Pilot or AI Binder Sprint and want to scope the design space before paying for wet-lab validation

Running an internal feasibility study before committing budget to a multi-week campaign

Tuning hotspot residue selection from Epitope Scout output and want to see which patches generate scoreable designs

Run the algorithm we run on our own campaigns

Sign in, upload a target structure, define hotspots. Get ranked binder candidates scored with JAX AF2 Initial Guess.

Algorithm based on Bennett, J. et al. Improving de novo protein binder design with deep learning. Nature Communications 14, 2625 (2023). See all Ranomics technology.