The Decision Layer for Protein R&D
An agentic integration & decision platform for explainable protein discovery, design, and optimization, from targets to candidates.
The Challenge
Researchers waste weeks stitching together incompatible tools, chasing unreliable results, and re-running analyses from scratch. ProtXplain changes that.
Structure prediction, docking, and sequence analysis live in separate universes, forcing manual handoffs and error-prone copy-paste workflows.
Critical choices, which target to pursue, which variant to test, are made without a unified view of the evidence, leading to costly late-stage failures.
Ad-hoc scripts and undocumented pipelines make it nearly impossible to reproduce, audit, or build upon prior work, even within the same team.
The Solution
ProtXplain combines an intelligent research agent with a visual workflow engine, giving your team both depth of insight and control.
A conversational AI agent with deep protein knowledge. Ask questions in natural language, retrieve structures, identify binding hotspots, and generate data-driven hypotheses, all in one place.
A visual, node-based workflow builder for protein R&D. Connect AlphaFold, ProteinMPNN, ESM-2, and custom tools into reproducible, auditable pipelines, without writing a single line of code.
Workflow
ProtXplain connects the entire protein R&D loop, from exploratory questions to structured, reproducible experiments.
Start with a natural language question. ProtXplore retrieves relevant structures, surfaces binding insights, and generates data-driven hypotheses.
Convert your hypothesis into an executable workflow. Drag, connect, and configure nodes for folding, design, docking, and validation.
Get structured, auditable results. Compare variants, track decisions, and share reproducible workflows with your team.
Use Cases
From academic labs to biotech pipelines, ProtXplain accelerates decision-making across the most critical protein R&D workflows.
Identify novel candidates and functional similarities across proteins, uncovering hidden relationships and better starting points.
Predict and rank protein function with interpretable, auditable reasoning, even in low-data settings.
Design and prioritize antibodies and protein binders from target to decision.
Design and optimize enzymes by screening and ranking mutations for activity, stability, and selectivity.
Rank biomarkers across datasets and cohorts with high confidence and reproducible, traceable results.
Evaluate and prioritize targets based on structural context, druggability, and biological evidence.
The Team
A founding team combining deep computational biology, AI engineering, and biotech experience.
Founding Team
PhD in Computational Biology (TUM). First-author publication in Nature Communications (2025). Co-founder of CORE64.
PhD in Molecular Evolution (U. Queensland). 60+ publications, including Nature Genetics (2020) and Nature Communications (2023).
PhD candidate in Computational Biology (TUM). MSc in CS (AI). Bridges agentic AI engineering and product design.
Development & Application
PhD candidate at Helmholtz Munich. Brings agentic AI systems expertise to ProtXplain's orchestration layer.
PhD candidate at TUM. Applies computational methods to protein sequence and structure analysis across the ProtXplain pipeline.
ETH Zürich alum. Full-stack engineer building the platform infrastructure, frontend, and integrations that power the ProtXplain experience.
PhD from the University of Bristol. Bridges computational protein design with wet-lab validation to ground ProtXplain's predictions in experimental reality.
Advisory Board
Focuses on AI-guided protein design to engineer cellular decision-making by creating synthetic proteins that modulate, sense, or reprogram signal transduction.
Develops bioinformatics algorithms and software for high-throughput biological research, focusing on structural genomics, proteomics, protein interactions, and protein function prediction.
Early Access
Join the researchers and biotech teams already using ProtXplain to make faster, better-grounded decisions in protein design.