Use structural modeling to accurately predict exposed liabilities for thousands or millions of antibody sequences at once with the IGX-Annotate App
Antibody discovery workflows used to rely on low-throughput, time-consuming technologies that required years to develop a handful of antibody candidates. With the rise of next-generation sequencing (NGS), researchers are now able to produce tremendous amounts of sequencing data and identify millions of potential antibody candidates in a single experiment.
The generation of large amounts of sequencing data brings, however, a new challenge to antibody discovery. It is impracticable and economically unfeasible, to express and functionally test millions of antibody candidates in the laboratory. Consequently, a diverse set of antibodies need to be selected from millions of antibody sequences. How do you choose the best candidates from such a large pool of antibody candidates? In other words, how do you find that needle-in-a-haystack antibody?
The importance of predicting sequence liabilities and developability risks
Bulk and single-cell NGS technologies generate orders of magnitude larger volumes of data compared to Sanger sequencing. As a result, the pool of potential antibody candidates to pick from increases tremendously with the use of such technologies. To increase the chances of picking the right candidate, it is crucial to assess the sequence liabilities and the developability of candidates early in the discovery phase. This will prevent many candidates to fail at later stages in the development process, when significant amounts of time and resources have already been spent. Predicting sequence liabilities and developability risks with high confidence at early stages in the discovery process helps to:
- Reduce the risk of picking candidates unfit for production and clinical use
- Decrease time spent on antibody engineering
- Obtain more first-time-right, high-quality candidates
Linear and structural liability predictions; what is the difference?
Historically, the assessment of antibody developability profiles has been a trade-off between throughput and accuracy. Screening linear sequences for motifs associated with liabilities is a relatively simple and high-throughput method to identify antibody liabilities. However, sequence-based assessments do not consider the position of these motifs in the 3D structure of the antibody. In fact, liabilities that are exposed on the surface of an antibody are more likely to affect stability, solubility and other biophysical characteristics compared to liabilities located inside of the molecule. By only using a sequence-based analysis, antibodies that have motifs located inside of the molecule will be falsely flagged and removed from the discovery pipeline. Predicting the 3D structure of an antibody by structural modeling can help to accurately assess exposed sequence liabilities. However, such predictions come with certain challenges as these methods require significant computational resources, especially while used in a high-throughput manner.
Accurate liability predictions with IGX-Annotate at scale
ENPICOM developed a specialized App, named IGX-Annotate, within the IGX Platform that leverages highly accurate annotation of liabilities and developability risks. More specifically, IGX-Annotate performs high-throughput structural modeling of antibody candidates to accurately identify and annotate liabilities exposed at the surface. The App is powered by the Structural Antibody Prediction Platform (SAbPred) developed by researchers at the University of Oxford. SAbPred is a validated, peer-reviewed, and globally recognized toolbox that makes accurate and efficient predictions about antibodies based on their 3D structure. In addition to the structural analysis, IGX-Annotate compares the developability profile of your antibody candidates to those of all clinically approved antibodies. Together, these two analyses will greatly de-risk antibody development by prioritizing candidates with the best developability profiles.
IGX-Annotate allows to
- Leverage a validated tool for the identification of liabilities. Use a validated structural modeling toolbox to improve liability annotation of your antibody candidates
- Annotate structural liabilities for thousands of sequences. Run the structural liability predictions of SabPred for many molecules in parallel and scale the analysis to meet your needs
- Directly compute developability profiles. Easily configure the penalty scores for all predicted liabilities and developability properties to prioritize antibody candidates that fit your research strategy
- Annotate structural liabilities for single-chain variable fragments. IGX-Annotate now also supports the structural liability predictions for single-chain variable fragments. These fragments are often used in antibody display workflows, e.g., phage display. Combine IGX-Annotate with our newest App IGX-Track, that supports antibody display data, to select the best candidates after complex panning rounds
- Overlay liability scores. Liability and developability scores can directly be used to guide the selection of your top candidates, e.g., use the scores to identify antibody clusters or candidates in the interactive cluster or phylogenetic tree viewers, respectively
From thousands or millions of sequences to that needle-in-a-haystack antibody
The IGX Platform helps you to find that needle-in-a-haystack antibody within thousands or millions of sequences. With IGX-Annotate, you can easily integrate accurate developability predictions into your workflow to select antibodies with the best characteristics and minimal risk of failure at later stages during development. Be sure to read and download our latest white paper to learn more about IGX-Annotate and the latest features of the IGX Platform.
Find out how the IGX Platform can help to accelerate and improve your antibody discovery workflow.