Connect
Login

SentinusAI®

SentinusAI® is a de novo antibody and fusion protein engineering AI platform that designs and optimizes candidates based solely on the target sequence, delivering computational results within a week and optimized drug candidates within a month. Milestone success is typically achieved with wet-lab testing of just 50–150 protein molecules across 3 iterations.

Protein Design Engine

De Novo Protein Design

IgG, Fab, scFv, VHH, peptides etc.
Bispecific and multi-specific formats
Antibody-drug conjugate (ADC, PDC)

Affinity Maturation

Therapeutic Antibody Affinity Maturation
Fusion Protein Engineering

Therapeutic Protein Humanization

Immunogenicity Assessment
Antibody Humanization

Epitope Mapping

Binding specificity Optimization

Therapeutic Protein Off Target Prediction

Protein Off-target Toxicity Prediction

Therapeutic Protein Developability

Stability Prediction of Sequence
Optimization

Application Scenarios:

Therapeutic Drug and Vaccine Design
Diagnostic Antibody Design
Industrial Enzyme Optimization

Project Lifecycle

Step 1:
NDA & Project Assessment

Step 2:
Statement of Work & Quote

Step 3:
Sign Contract & Invoice

Step 4:
Upfront Payment &
Target Information for Project Kick-Off

Step 5:
Antibody Designed, Expressed, and Tested

Step 6:
Project Delivered & Milestone Payment

Platform Demos

SentinusAI®
Frequently Asked Questions (FAQs)

What design approaches does SentinusAI® offer, and when should I choose each one?
  • Sequence optimization / affinity maturation – Starting from your existing VH/VL sequences, our sequence-based deep‑learning engine introduces virtual CDR mutations and proposes ~50 variants per round. Typical hit rates: ~10 % (Round 1), 40 % (Round 2), 60 % (Round 3). Best when you already have lead antibodies that need higher affinity or better developability.
  • Structure‑based design – If you have crystal or cryo‑EM data, we can use it to refine epitope insights, but 3‑D structures are not required; our core models remain sequence‑based.
  • De novo design – For novel or hard‑to‑express targets, we generate ~50 de novo antibody sequences directly from the antigen sequence and validate them in cell‑based assay (if biochemical assay is not available). Ideal when no starting antibody exists.

·       Computational phase: ~1 week per iteration (sequence optimization or de novo).

·       Protein expression & testing: ~2 weeks per batch.

·       Total project cycle: 1 – 3 months, depending on the number of design–test iterations.

Item What’s Included Typical Batch Size Timeline*
Computational design Affinity maturation or de novo sequence generation; ranked candidate list; developability assessments — 1 week
Antibody expression Gene synthesis, cloning, CHO cell expression, purification, QC 50 samples (200 µg – 1 mg each) 2 weeks
Custom target protein expression Gene synthesis, cloning, CHO cell expression, purification, QC Per target 2 – 3 weeks
Binding assays (BLI-Gator/Octet Detection) Single‑dose ranking or full KD measurement 50 samples 5 – 7 days
Binding assays (SPR/Biacore Detection) Single‑dose ranking or full KD measurement 50 samples 7 – 10 days

*Timelines may run in parallel whenever possible.

  • For affinity maturation: VH/VL sequences, target (antigen) sequence, any functional data (binding affinity, internalization, stability) is optional.
  • For de novo design: Target sequence alone is sufficient; epitope hints or functional data (binding affinity, internalization, stability) accelerate optimization.
  • For all projects: An NDA and basic target background. 3‑D structures and detailed epitope maps are helpful but not mandatory.
  • Optimized DNA and amino‑acid sequences.
  • Upon request: expression vectors or purified antibody/protein samples.
  • Detailed computational rankings, assay reports (e.g., KD values), and, upon request, expanded functional data such as internalization, stability, or developability metrics.

No. Our AI can perform in‑silico epitope mapping and paratope‑epitope interaction modeling from sequence alone. However, any experimental epitope or structural insights you can share will further refine predictions.

  • Design sequences outside known patent claims using proprietary freedom‑to‑operate algorithms (upon request).
  • Run similarity checks against public patent databases (upon request).

No. Ainnocence’s discovery platform is a proprietary, sequence-based AI system developed entirely in-house. By working directly from primary sequence data, it eliminates the need for external 3-D structure-prediction software and other open-source packages. Instead, our technology performs ultra-high-throughput virtual screening and multi-objective optimisation, evaluating up to 10¹⁰ candidates within hours to days—giving our partners capabilities that off-the-shelf tools cannot match.

Interested in how we can support your goals? We’re here to answer any inquiries.

Let's Talk!