AI & De Novo Protein Design
How de novo binder design actually works in practice: RFdiffusion, BindCraft, and ProteinMPNN on the compute side, display validation on the bench side.
Working on a ai & de novo protein design project? See our AI protein binder design services.
AI protein binder design services →Wet lab as an API for binder design agents
The Ranomics Platform API lets AI binder design agents submit candidates from RFdiffusion or ProteinMPNN runs directly to a real wet lab. Yeast display, mammalian display, or DMS. Enriched hits, NGS counts, and called binders return as structured JSON or signed webhooks.
Using ColabFold Batch to Triage RFdiffusion and BindCraft Binder Pools
How to use colabfold_batch to filter hundreds or thousands of de novo binder designs before any gene synthesis. pLDDT cutoffs, self-consistency RMSD thresholds, and where the filter typically lands.
ColabFold vs AlphaFold 2: When the MMseqs2 Frontend Beats the Full MSA Pipeline
ColabFold runs the AlphaFold 2 weights with an MMseqs2 MSA frontend instead of jackhmmer plus HHblits. We break down when the speedup is free and when full-MSA AlphaFold 2 still wins.
AI-Driven Protein Design: An Honest CRO Perspective
What AI-driven protein design actually delivers in 2026: where it works, where it fails, and what wet-lab validation reveals.
BindCraft vs RFdiffusion: When to Use Which for Binder Design
BindCraft and RFdiffusion solve different parts of the binder design problem. A practical comparison from a CRO that uses both.
Why Most AI-Designed Binders Fail Wet Lab — and How to Fix It
Wet-lab hit rates for AI-designed de novo binders fall well below the headline numbers in method papers. The failure modes, the developability filters, and how to close the gap.
Run ProteinMPNN sequence design in your browser — no installation required
The Ranomics tools hub now hosts a one-click ProteinMPNN interface. Upload a backbone, set your parameters, download designed sequences. No GPU, no conda environment, no command line.
5 Developability Red Flags That Kill mAb Programs
The five antibody developability red flags most commonly responsible for late-stage CMC failures, how each one shows up in the clinic, and which of them you can detect from sequence before synthesis.
RFdiffusion Outputs Need a Developability Check Before Wet Lab
RFdiffusion, BindCraft, and ProteinMPNN optimize for structure and binding, not for developability. A concrete triage workflow for filtering generative outputs before synthesis or yeast display.
Binder Design on a Grant Budget: Scoping a Single-Target Campaign
What to prioritize, what to cut, and what actually determines cost when a PI or postdoc is running a single-target de novo binder design campaign on a defined budget.
From an AlphaFold Model to Your First Binder: A Walkthrough for Teams Without Structural Biology Expertise
A practical, step-by-step guide for small biotech and academic teams who have an AlphaFold model of their target but no structural biologist on staff — what to check, what to decide, and how to move into a binder design campaign.
When to Use Epitope Scout vs. a Structural Biologist
A practical framing of when automated epitope scoring is enough for your binder campaign and when you actually need a human structural biologist in the loop — with a checklist for deciding on your own target.
Closing the Loop: AI Protein Design Plus Display Screening
Most teams treat computational design and experimental screening as separate workflows. The programs that produce the best binders treat them as one coupled system.
Protein Engineering Design in the Age of Machine Learning
Modern protein engineering design increasingly relies on machine learning, but experimental data and workflow integration remain the true bottlenecks. A guide to the six-stage design cycle.
From Computational Protein Design to Validated Binders: What Actually Works
What separates successful AI protein design campaigns from failed ones? A practical breakdown of the computational and experimental steps required to go from generative models to validated binders.
ProteinMPNN and the sequence design problem: what it does and why it matters
ProteinMPNN solves the inverse folding problem. Given a backbone, which sequences will fold into it? How it works, how it fits into de novo binder design pipelines, and the practical parameters that matter.
Hotspot-guided binder design: using structure to focus the design campaign
Hotspot residues (the subset of interface contacts that contribute most of the binding energy) dramatically improve de novo binder design campaigns when used to constrain diffusion-based generation.
AI de novo design vs. library screening: when to use which approach
De novo computational design and library screening are not competing methods. A decision framework for choosing between them, and why the best programs often couple both.
RFdiffusion in Practice: What Works and What Fails
Operational lessons from running RFdiffusion binder design campaigns. Scaffold topology biases, hotspot conditioning tradeoffs, partial diffusion for scaffold grafting, failure modes on flat targets and membrane proteins, and how to avoid redundant candidate pools.
De Novo Protein Design: How the Pipeline Works in Practice
A practitioner's guide to de novo protein binder design using RFdiffusion, BindCraft, ProteinMPNN, and structural validation. What the real bottlenecks are, what determines campaign success, and how experimental validation has replaced computation as the rate-limiting step.
Ready to start a project?
Tell us about your protein engineering challenge. We will scope a program and get back to you within 24 hours.
Start a project →