v1.9.3 Public Beta

Engineering
Life. Digitally.

An accurate and intelligent protein structure prediction platform. Accelerate drug discovery with next-generation AI models.

Request demo
Explore

Our Pipelines

Two design routes. One intelligent platform.

Pipeline 1 - Foldexa CDR - 4-step

Pipeline 2 - Foldexa Scaffold - 4-step

Pipeline 3 Research preview

Your next antibody.
A clearer path forward.

Redesign binding regions, explore predicted structures, and compare candidates in one workspace.

Start a design View your runs

Fv, VHH and full-antibody input. Laboratory validation is required.

  1. 01

    Bring your structure

    Upload an antibody–target complex and choose the regions to redesign.

  2. 02

    Choose your design

    Choose the antibody format, regions and number of candidates.

  3. 03

    Review the evidence

    Inspect 3D structures, sequence changes and the control result.

From Algorithms
to the Cure.

Foldexa was created by Azamat and Rauan with a vision to democratize protein science.

We combine Foldexa CDR, Foldexa Scaffold, and Foldexa Structure into a seamless platform for de novo antibody design.

2
End-to-end pipelines
4+
AI models integrated
Open
Source UI on GitHub

Our Pipelines

Pipeline 1 — CDR-focused redesignFoldexa CDR
01
Foldexa CDRAntibody CDR DesignOpen
02
Foldexa RefineEnergy MinimisationOpen
03
Foldexa StructureStructure ValidationOpen
FilteringRMSD · i_PAE · ddGProprietary
4-step · ~3 h runtime
Pipeline 2 — De novo scaffold designFoldexa Scaffold
01
Foldexa ScaffoldScaffold GenerationOpen
02
Foldexa SequenceSequence DesignOpen
03
Foldexa StructureStructure ValidationOpen
FilteringpLDDT · i_PAE · ddGProprietary
4-step · ~5 h runtime

Simple. Powerful. Fast.

From raw protein structure to publication-ready variants.Automatically.
Generate

Step 03

Generate

Analyze Results

Step 04

Analyze Results

Upload Structure

Step 01

Upload Structure

Choose Pipeline

Step 02

Choose Pipeline

Generate

Step 03

Generate

Analyze Results

Step 04

Analyze Results

Upload Structure

Step 01

Upload Structure

Choose Pipeline

Step 02

Choose Pipeline

Our Partners

ProteoGenix Logo

ProteoGenix

Antibody discovery and production

Solbridge Logo

Solbridge

International School of Business

KAIST Logo

KAIST

Korea Advanced Institute of Science

Become a Partner

Join ProteoGenix, KAIST, Solbridge, and other partners in advancing the future of protein engineering.

Foldexa.bio

Pipeline Benchmark vs. World-Class Platforms

Anti-Tie2 CDR redesign · Foldexa Structure predicted metrics · In Silico metrics · February 2026

Core Metrics Comparison

Interface PAE (i_PAE)

Lower = higher confidence in binding pose

★4.54

Foldexa.bio
Best (#1)

4.62

Foldexa.bio
Top-10 avg

5.75

Reference
hTAAB-hTie2

8.2

Reference Engine
(median)

CDR-L1 Backbone RMSD

Angstroms (Å) — lower is better

2 Å filtering threshold
★0.179

Foldexa.bio
Best (#1)

0.37

Foldexa.bio
Top-10 avg

0.81

Reference
hTAAB-hTie2

~1.5

Foldexa CDR
(scRMSD)

Binding Free Energy (Foldexa Refine ddG)

kcal/mol — more negative = stronger binding

0 kcal/mol
−25.75★

Foldexa.bio
Best (#4)

−23.52

Foldexa.bio
Top-10 avg

−15.00

Foldexa Scaffold
(Bennett)

0.00

Reference
hTAAB-hTie2

Key Performance Indicators

100%
Hit rate
(10/10 designs passed)
0.179 Å
Best RMSD
(near-native)
−25.75
Best ddG
(kcal/mol)
4.54
Best i_PAE
(interface confidence)
0.96
pLDDT
(structure quality)

Pipeline Feature Comparison

FeatureFoldexa.bioFoldexa ScaffoldFoldexa CDRReference Engine
CDR redesign✓ All CDRs✓✓— (structure only)
Hit rate100% (10/10)~30–60%~25%N/A
Structural validationFoldexa Structure ensembleFoldexa StructureRMSD onlyBuilt-in
Binding energy scoringFoldexa Refine ddGFoldexa Refine——
End-to-end pipeline✓ AutomatedManual assemblyManualInference only
Sources: Bennett et al., Nature (2025); Luo et al., ICML (2022); Chai Discovery, bioRxiv (2025); Jin et al., Nat. Commun. (2021)
Note: Foldexa.bio metrics from AF2-multimer predictions. Cross-platform metrics not directly comparable.
Foldexa.bio

We are Foldexa

Four different paths.
One shared obsession.

We came from different worlds — bioengineering, software, business, and systems architecture. But we shared one belief: protein engineering should be accessible to everyone.

Azamat Armanuly

Azamat Armanuly

CEO

Bioengineer — KAIST

"Understanding life at molecular level"
Read story
Rauan Bolat

Rauan Bolat

CPO

Project Manager & Product Developer

"Architecting systems from concept to execution"
Read story

Swipe to see all

Four disciplines. Four founders.
One platform.

Foldexa brings biology, software, and vision together to accelerate the future of protein engineering.

See what we built
Our Vision

To democratize protein engineering and make cutting-edge AI tools accessible to researchers worldwide.

We believe that breakthrough discoveries shouldn't be limited by computational barriers.

Get in Touch

Interested in partnering with us?

Whether you want to learn more or explore collaboration opportunities, we'd love to connect.

LinkedInEmail Us