ASMS 2026: Automated LC-MS Disulfide Characterization in Trispecific Antibodies
Positive controls, MS1 isotope envelope confidence, and decision-tree triage for complex disulfide analysis
About this Poster
Roughly 30% of FDA biologics approvals over the past decade depend on correct disulfide bond formation for structural integrity and efficacy — and the more complex the format (mono-, bi-, trispecific antibodies), the harder disulfide characterization gets. Disulfide scrambling during digestion plus low signal intensity of native disulfide-linked peptides make non-reducing LC-MS peptide mapping unreliable. This poster, from Protein Metrics with Fresenius Kabi Biopharma and EPFL, combines a scrambling-induced positive control with MS1 isotope envelope confidence scoring, automated decision-tree filtering, and a novel cysteine-based grouping strategy to make disulfide analysis in a trispecific antibody both confident and quantitative.
Key Learnings:
- See how neutral-pH digestion without alkylation dramatically increases shuffled species (~8% → 69%), exposing more scrambling for detection and validation.
- Understand how MS1 isotope envelope confidence scoring lets Byos identify low-intensity disulfide-linked peptides with high confidence.
- Learn how decision-tree-based automated filtering cuts manual-review candidates by ~85% (from 415 down to 60) without sacrificing reliability.
- Discover the cysteine-based grouping strategy that links each expected disulfide species to its shuffled variants, enabling quantitative disulfide analysis in trispecific antibodies.
In collaboration with Fresenius Kabi Biopharma (Vaud, CH) and EPFL (Lausanne, CH).