> About TAP
How similar is your antibody to clinical-stage therapeutics?
The earlier error has now been fixed and jobs should be running normally. Thanks for your understanding, and sorry for the inconvenience. --Claire
- To use this application, you only require an amino acid antibody variable domain heavy and light chain sequence.
- The Therapeutic Antibody Profiler (TAP) compares your antibody variable domain sequence against five developability guidelines derived from clinical-stage therapeutic values.
- TAP uses ABodyBuilder, with side-chains modelled by PEARS, to generate a model structure of your antibody. The standard implementation takes around 30 seconds.
- If the H3 loop can not be modelled using FREAD, it defaults to modelling the loop with MODELLER (+~3mins). Sphinx can also be used for more accurate modelling, but expect a large increase in runtime.
- TAP then calculates the following properties (explained in the paper) to see if your antibody design is commenserate with those of clinical-stage therapeutics:
- Total CDR Length
- Patches of Surface Hydrophobicity (PSH) metric across the CDR Vicinity
- Patches of Positive Charge (PPC) metric across the CDR Vicinity
- Patches of Negative Charge (PNC) metric across the CDR Vicinity
- Structural Fv Charge Symmetry Parameter (SFvCSP)
- TAP also outputs...
- An interactive representation of your antibody model, allowing you to visualise hydrophobic/charge patches on the Van der Waals surface. The model is downloadable.
- A representation showing predicted model accuracy, communicating the confidence you should have in our metric values
- Sequence liabilities likely to lead to post- or co-translational modification
- North-defined length-independent canonical forms (excluding the H3 loop)
- For a sample TAP output, please click here. More information about the guidelines and more resources can be found here.
For a full description of the profiler, or if you use this software, please refer to: Raybould MIJ, et al. Five Computational Developability Guidelines for Therapeutic Antibody Profiling. PNAS. (2019)
> Resources
Please cite our paper "Five Computational Developability Guidelines for Therapeutic Antibody Profiling" if making use of these datasets.
- The 242 Clinical-Stage Therapeutics (CST) models that set the TAP Guidelines in the PNAS paper can be found here.
- The 242 CST sequences can be found in the paper SI here (Dataset S2).
- The TAP Guidelines were last updated on 16th June 2019. We are now tracking 377 post Phase-I therapeutics. The 377 models that set the most recent TAP Guidelines can be found here. Their TAP metric values can be found here. The guidelines set by these models are below:
| Property | Amber Region (16th June 2019) | Red Region (16th June 2019) |
|---|---|---|
| Total CDR Length (L) | 39 ≤ L ≤ 42 | L < 39 |
| 55 ≤ L ≤ 60 | L > 60 | |
| Patches of Surface Hydrophobicity (PSH) | 88.2933 ≤ PSH ≤ 100.788 | PSH < 88.2933 |
| 158.470 ≤ PSH ≤ 179.175 | PSH > 179.175 | |
| Patches of Positive Charge (PPC) | 1.2336 ≤ PPC ≤ 3.7545 | PPC > 3.7545 |
| Patches of Negative Charge (PNC) | 1.8333 ≤ PNC ≤ 3.4688 | PNC > 3.4688 |
| Structural Fv Charge Symmetry Parameter (SFvCSP) | -19.50 ≤ SFvCSP ≤ -6.016 | SFvCSP < -19.50 |
Please visit Thera-SAbDab to access all of our CST sequence data.
