Use TNP (Therapeutic Nanobody Profiler) Online
Commercially Available TNP (Therapeutic Nanobody Profiler) No-Code Web Server
TNP: Nanobody Developability Profiling and Prediction from Sequence
TNP (Therapeutic Nanobody Profiler) is a computational tool that characterizes and predicts the developability of nanobodies, meaning how well a candidate is likely to behave during manufacturing, storage and stability testing. It follows the framework of the Therapeutic Antibody Profiler (TAP) but is tailored to the structural features of single-domain VHH antibodies. According to the authors, a web application and a Python package for high-throughput analysis are available, and code, descriptors and non-proprietary data are shared in a public GitHub repository.
Nanobodies are attractive therapeutic formats, but developability problems such as aggregation, poor expression or unwanted stickiness can sink a candidate late in a project. Existing developability tools were built around conventional antibodies and do not capture features that are specific to nanobodies, such as the way their long CDR3 loop interacts with framework residues.
To address this gap, the authors characterized nanobody structures, compared them with clinical-stage nanobodies, and experimentally tested 108 nanobodies. They then built TNP so that new sequences can be compared against the properties of nanobodies that have reached the clinic.
Property | Detail |
|---|---|
Task | Developability profiling of nanobody (VHH) sequences |
Framework | Follows the Therapeutic Antibody Profiler (TAP) approach |
Metrics | Six: total CDR length, CDR3 length, CDR3 compactness, and surface patches for hydrophobicity, positive charge and negative charge |
Reference set | 36 clinical-stage nanobodies define the thresholds |
Structure prediction | NanoBodyBuilder2 (three models generated; the one with a CDR-H3 closest to the average is used) |
Flags | Amber and red, relative to the clinical-stage distribution |
How TNP Profiles Nanobody Developability
TNP predicts a 3D structure for each input sequence and then computes six descriptors that are compared with the distribution seen in clinical-stage nanobodies.
The six TNP metrics
Total CDR length and CDR3 length: overall loop length, using IMGT numbering for the CDRs.
CDR3 compactness: loop reach divided by loop length. A value of 1.25 separates two CDR3 subpopulations in the authors' analysis.
Surface hydrophobicity patches: a TAP-style measure of exposed hydrophobic area.
Positive and negative charge patches: TAP-style measures of exposed charged regions.
Amber and red flags
A property receives an amber flag when it falls in the lowest or highest 5% of the clinical-stage distribution, and a red flag when it lies outside the range observed for clinical-stage nanobodies.
Benchmark Data and Results for Nanobody Developability Prediction
The authors compared eight sequence and structure datasets, including clinical-stage molecules from TheraSAbDab, natural repertoires from Observed Antibody Space, and entries from PLAbDab-nano and SAbDab. They also assayed 72 proprietary VHHs alongside the 36 clinical-stage nanobodies, using SEC, CIC, HIC, DLS, nDSF, AC-SINS and BVP-ELISA. The paper describes the assay data as the first publicly available nanobody developability assay dataset.
Result | Reported value |
|---|---|
Nanobodies assayed in vitro | 108 (36 clinical-stage plus 72 proprietary) |
Red flags among proprietary candidates | 14 across 11 candidates |
Red flags for surface hydrophobicity patches | None |
Bound vs. unbound structure pairs | 76 pairs; compactness scores were consistent |
What is Tamarind Bio?
Tamarind Bio is a no-code bioinformatics platform built to give life scientists and researchers access to powerful computational tools. Many cutting-edge machine learning models are hard to deploy and use. Tamarind provides an intuitive, web-based environment that removes the complexity of high-performance computing, software dependencies and command-line interfaces.
The platform is designed for biologists, chemists and other researchers who may not have a background in programming or cloud infrastructure but want to run models on their own data. Key features include:
A user-friendly graphical interface for setting up and launching experiments
A robust API for integration into existing research pipelines
An automated system for managing and scaling computational resources
Tamarind treats information and data security as a top priority, as detailed in its Trust Center and Terms of Service.
Accelerating Discovery with TNP on Tamarind Bio
Triage nanobody libraries: flag candidates with unusual CDR length, compactness or surface patch values before committing to expression.
Compare against clinical precedent: see where a sequence sits relative to 36 clinical-stage nanobodies.
Prioritize assay budgets: decide which candidates deserve SEC, HIC or other developability assays first.
Guide engineering: use red and amber flags to target the loop or surface features worth redesigning.
How to Use TNP on Tamarind Bio
Log in and open the tool: sign in at tamarind.bio and select TNP (Therapeutic Nanobody Profiler).
Provide nanobody sequences: enter the VHH amino acid sequence or sequences you want to profile, if the input form accepts several.
Set parameters: keep the defaults unless the form exposes options; the source defines IMGT numbering and a NanoBodyBuilder2 structure step.
Run the job: submit and wait for structure prediction and descriptor calculation to finish.
Download results: save the six metric values and flags, and any predicted structures if exposed.
Interpret the flags: amber means the lowest or highest 5% of the clinical range; red means outside it.
Use results downstream: rank candidates and combine TNP output with experimental developability data.
Things to Keep in Mind
Thresholds rest on a small reference set of 36 clinical-stage nanobodies, which the authors plan to update as more data appear.
Agreement between computed metrics and in vitro assays was relatively weak, and some candidates with green flags still showed amber or red in vitro results.
All structures are predicted rather than experimentally determined.
The authors note that the six-metric set is one of many possible choices.
Source: Therapeutic Nanobody Profiler (TNP) bioRxiv preprint, doi.org/10.1101/2025.08.11.669635 (2025). Code and non-proprietary data: github.com/oxpig/TNP. Web application: opig.stats.ox.ac.uk/webapps/tnp.