cyberneticlibrary
← library

Engineer protein glycosylation patterns

glycoengineeringskillsetup L2★27,559
K-Dense-AI/scientific-agent-skills ↗
What it does

Analyze and engineer protein glycosylation sites (N- and O-linked)

Best for

Antibody Fc optimization (enhanced ADCC/CDC), therapeutic half-life engineering, or vaccine design where glycan patterns directly control immune recognition.

Inputs
  • · protein sequence (FASTA)
  • · target tissue/cell type
  • · therapeutic goal (stability, ADCC, immune evasion)
Outputs
  • · N-glycosylation sequon map (position, NXS/NXT motifs)
  • · O-glycosylation hotspot scores
  • · mutation recommendations (Asn→Gln, add/eliminate sites)
Requires
  • · NetOGlyc (O-glycosylation prediction)
  • · GlycoShield
  • · GlycoWorkbench
  • · regex sequon scanner (local)
Preconditions

Protein sequence valid; 50%+ human proteins are glycosylated → tissue-type context matters

Failure modes
  • · proline at X+1 in NXS motif → not a true glycosite (N-X-[S/T] requires X≠Pro)
  • · O-glycosylation hotspot threshold unclear → heuristic scores may be noisy
Trust signals
  • · Sequon N-X-[S/T] motif definition precise
  • · IgG1 Fc example provided
  • · NetOGlyc tool integration documented