Use TAP2 (Therapeutic Antibody Profiling) Online
Commercially Available TAP2 (Therapeutic Antibody Profiling) No-Code Web Server
TAP2: Therapeutic Antibody Profiling to Flag Antibody Developability Risks
The Therapeutic Antibody Profiler (TAP) is a free computational tool that builds a homology model of an antibody variable domain and checks it against five developability guidelines. It accompanies the 2019 PNAS paper "Five computational developability guidelines for therapeutic antibody profiling" by Raybould, Marks, Krawczyk, Taddese, Nowak, Lewis, Bujotzek, Shi and Deane.
A therapeutic antibody must bind its target and also avoid developability problems such as poor stability or high aggregation. Small-molecule discovery has Lipinski's rule of five, but no comparable in silico guide existed for antibodies. The authors derived one from post-phase-I clinical-stage antibody therapeutics.
Property | Detail |
|---|---|
Publication | PNAS 116(10), 4025-4030 (2019), DOI 10.1073/pnas.1810576116 |
Reference set | 242 clinical-stage antibodies (137 training, 105 held-out) |
Metrics | Five guidelines with amber and red flag regions |
Structure modelling | Homology models built with ABodyBuilder |
Outputs | Flags, potential sequence liabilities and canonical forms |
The Five TAP Developability Guidelines
Total CDR length: amber flag from 54 to 60 residues, red above 60.
Patches of surface hydrophobicity (PSH) in the CDR vicinity: amber from 83.84 to 100.71 or 156.20 to 173.85, red below 83.84 or above 173.85.
Patches of positive charge (PPC): amber from 1.25 to 3.16, red above 3.16.
Patches of negative charge (PNC): amber from 1.84 to 3.50, red above 3.50.
Structural Fv charge symmetry parameter (SFvCSP): amber from -20.40 to -6.30, red below -20.40.
Validation: Do the Guidelines Flag Real Developability Problems?
Only 8 of the 105 held-out clinical-stage antibodies (7.69%) received a red flag, averaging 0.08 red flags per therapeutic. The authors also tested two industrial cases.
Case | Outcome |
|---|---|
MEDI-1912 (aggregation-prone) | Red flag for PSH; engineered MEDI-1912STT had no flags |
MEDI-578 | Amber flag for PSH |
AB001 (poor expression) | Red flag for PNC; AB008 had no flags |
Why Profiling Early Matters
Flagging risky sequences before expensive development lets teams prioritize candidates or redesign liabilities, as the engineered variants in the paper illustrate.
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 TAP2 on Tamarind Bio
Candidate triage: rank antibody hits from discovery campaigns by developability flags.
Liability engineering: find which CDR property, such as hydrophobic or negative-charge patches, drives a flag and redesign it.
Benchmarking against clinical antibodies: compare a candidate with the range seen in clinical-stage therapeutics.
Sequence liability review: check reported liabilities and canonical forms alongside the guidelines.
How to Use TAP2 on Tamarind Bio
Log in and open the tool: sign in at tamarind.bio and select TAP2.
Provide the antibody: enter the heavy and light chain variable domain sequences.
Run: submit the job; a homology model is built and profiled.
Download outputs: collect the metric values and flags, if exposed.
Read the flags: compare each metric with its amber and red regions.
Weigh distance from the range: the authors advise considering how far a value lies outside observed ranges.
Iterate: redesign flagged sequences and re-profile them.
Things to Keep in Mind
The guidelines cannot capture all developability issues, such as immunogenicity.
Homology models bias PSH toward higher values than crystal structures.
Thresholds are guidelines, not hard rules.
Only 25 clinical-stage antibodies have lambda light chains, too few for a separate lambda threshold.
Source: Raybould MIJ, Marks C, Krawczyk K, Taddese B, Nowak J, Lewis AP, Bujotzek A, Shi J, Deane CM, "Five computational developability guidelines for therapeutic antibody profiling", PNAS 116(10), 4025-4030 (2019), DOI 10.1073/pnas.1810576116 (details read via Europe PMC). Web server: http://opig.stats.ox.ac.uk/webapps/sabdab-sabpred/TAP.php.