Discovery and affinity maturation of antibody fragments from an unfavorably enriched phage display selection by deep sequencing and machine learning

Abstract

Phage display selection has been used for directed evolution of antibody fragments. However, variants with binding affinity cannot be always identified due to undesirable enrichment of target-unrelated variants in the biopanning process. Here, our goal was to obtain functional variants by deep sequencing and machine learning from a phage display library where functional variants were not appropriately enriched. Deep sequencing of the previously biopanned pools revealed that amplification bias might have prevented the enrichment of target-binding phages. We performed a sequence similarity search based on the deep sequencing analysis so that the influence of bias was decreased, leading to discovery of a variant with binding affinity, which could not be discovered by a conventional screening method alone. We applied machine learning to the deep sequencing data; the machine learning proposed effective mutations for increasing affinity, allowing us to identify a variant with improved affinity (EC₅₀ = 3.46 μM). In summary, we present the possibility of obtaining functional variants even from unfavorably enriched phage libraries by using deep sequencing and machine learning.