The Orbion Blog
Protein Science, in Practice
Field-tested deep-dives on structural biology, protein engineering, expression, and the AI behind it all — written for the people at the bench
PDB vs UniProt: Map Mutations Correctly
Map PDB, UniProt, and construct residue numbers before designing mutations, including insertion codes, missing coordinates, isoforms, and tags.
Low SEC Recovery: Where Did Your Protein Go?
Find where protein is lost during size-exclusion chromatography using a mass balance, collection checks, and tests for adsorption and aggregation.
DSF vs nanoDSF vs CPM for Membrane Proteins
Compare DSF, nanoDSF, and CPM for membrane-protein stability screening, including detergent interference, controls, and apparent melting temperatures.
How to Read a Membrane-Protein FSEC Trace
Interpret FSEC peaks, free GFP, aggregates, and retention shifts to choose membrane-protein constructs for purification and scale-up.
Membrane Proteins on SDS-PAGE: Heat Effects
Diagnose membrane-protein bands that smear or disappear after heating with controlled sample preparation, gel analysis, and transfer checks.
A280 vs BCA vs Bradford: Which Is Right?
Resolve conflicting A280, BCA, and Bradford results with buffer controls, dilution tests, protein-specific standards, and a clear measurement workflow.
AlphaFold Docking: A Readiness Checklist
A practical checklist for using AlphaFold structures in docking: pocket confidence, missing components, receptor state, preparation, and validation controls.
The Post-AlphaFold Model Zoo: MSA-Free Predictors and When to Trust Them
Navigate MSA-free protein predictors after AlphaFold and decide when single-sequence, complex, and inverse-folding outputs deserve trust.
Cryptic Pockets: Finding Druggable Sites the Active Site Hides
Find cryptic pockets by combining static structure, conformational sampling, chemical probes, and experiments—without overcalling druggability.
A quantum future, written in proteins
Orbion-led Nebula joins SPRIND QuantumSense Exploration to discover natural proteins for future biological research tools.
Which Host From Sequence Alone? Shortlisting E. coli, Insect, Mammalian, and Cell-Free
Shortlist E. coli, insect, mammalian, and cell-free expression using topology, disulfides, PTMs, toxicity, and assay needs.
Designing Constructs That Fit Your Vectors: The Compatibility Check Most Teams Skip
Check reading frames, junctions, tags, processing, and assembly constraints so every designed protein construct maps to a buildable vector.
Monomer, Dimer, or Something Messier? Predicting Quaternary State Before You Purify
Predict and validate protein quaternary state using sequence, interface evidence, concentration dependence, and orthogonal mass measurements.
Predicting Aggregation-Prone Regions From Sequence (Before They Wreck Your Prep)
Map aggregation-prone sequence regions, interpret their structural context, and choose better constructs, mutations, and process conditions.
GPCR Expression Optimization: A Computational Triage Before You Clone
Triage GPCR topology, boundaries, host, tags, state, stabilization, and assay design before committing to a large clone screen.
ProteinMPNN and Inverse Folding, Explained for Bench Scientists
A bench-focused guide to ProteinMPNN, inverse folding, functional constraints, sequence sampling, and testable design libraries.
ESM-2 vs AlphaFold: When a Language Model Beats a Structure Predictor
Compare ESM-2-derived models with AlphaFold and choose the right evidence path for orphan, designed, multidomain, and complex targets.
Orbion at the Quantum Global Summit
Orbion joined QAI Ventures’ Quantum Global Summit in Basel to pitch our work and discuss the role of physics, AI and quantum research in protein engineering.
pTM and ipTM: The AlphaFold Scores You Ignore When Ranking Complexes
Learn how pTM, ipTM, and inter-chain PAE work together—and how to rank AlphaFold complex models without trusting one score.
Native PAGE, DLS, SEC-MALS, or AUC: Which Oligomerization Result to Trust
Native PAGE says dimer. DLS reports a radius too large for a dimer. SEC-MALS gives mostly monomer with a rising mass across the peak, and AUC resolves two…
Signal Peptide or Transmembrane Helix? Resolving Ambiguous N-Terminal Predictions
The first 25 residues are hydrophobic. One predictor calls them a cleavable signal peptide, another calls them transmembrane helix 1, and AlphaFold draws a…