Phage display is a laboratory technique that uses viruses called bacteriophages to discover new proteins and peptides that bind to specific targets. Developed in 1985 by George Smith, the method earned him a share of the 2018 Nobel Prize in Chemistry. At its core, the technique works because each bacteriophage particle physically links a protein on its outer surface to the DNA instructions for making that protein inside. This connection lets researchers screen billions of different protein variants at once, then immediately identify the genetic sequence of any “winner” that sticks to a target of interest.
How Genotype-Phenotype Linkage Works
A bacteriophage is a virus that infects bacteria. It consists of a protein coat surrounding a strand of DNA. In phage display, scientists insert a foreign gene into the phage’s DNA so that the encoded protein or peptide gets produced as part of one of the phage’s coat proteins. When the phage assembles itself inside a bacterium, the foreign protein appears on the phage’s outer surface while the gene encoding it stays packaged inside the same viral particle.
This physical linkage between what’s displayed (the phenotype) and the gene responsible (the genotype) is the entire basis of the technique. If a particular phage binds tightly to a target molecule, you can pull it out, read its DNA, and know exactly which protein sequence was responsible. And because identical phage particles come from the same bacterial clone, you can reproduce that result reliably.
Building a Phage Library
The power of phage display comes from scale. Rather than testing one protein at a time, researchers create a “library” containing billions of phage clones, each displaying a slightly different peptide or protein variant on its surface. A typical library contains around 1 to 2 billion unique clones, though theoretical diversity can be orders of magnitude higher depending on the length and composition of the randomized DNA sequences inserted.
Building these libraries involves inserting randomized DNA sequences into the phage coat protein gene, then letting bacteria produce the resulting phage particles. Not every bacterium successfully produces a working phage. In practice, roughly half to two-thirds of transformed bacteria yield viable particles, so the actual diversity of a library is always smaller than what the DNA sequences alone would predict. The composition also shifts during amplification: library diversity can drop more than sevenfold as certain clones replicate faster than others inside bacteria, regardless of their binding ability. A phage clone that’s only a moderate binder can still dominate a library simply because it replicates more efficiently.
The Biopanning Cycle
The selection process, called biopanning, works like an evolutionary filter repeated over several rounds. In each round, the phage library is exposed to a target, such as a protein, a cell surface receptor, or even a whole cell. Phages whose displayed peptides bind the target stick around. Those that don’t bind, or bind only weakly, get washed away.
The retained phages are then stripped from the target using an acidic solution (a step called elution) and fed to fresh bacteria, where they replicate and produce a new, enriched pool. This enriched pool becomes the input for the next round. After three to five rounds, the pool is heavily skewed toward phages that display high-affinity binders. Researchers then isolate individual clones and sequence their DNA to identify the winning peptide or protein.
Common Phage Platforms
Several bacteriophages serve as display platforms, but two dominate the field. M13, a filamentous phage with single-stranded DNA, is the most widely used system. It’s well characterized and versatile, but its reliance on the bacterial secretion pathway to export displayed proteins can limit what it handles well.
T7 phage offers a different set of strengths. It uses double-stranded DNA, which is more stable and less prone to mutation during replication. T7 doesn’t depend on a protein secretion pathway, so it can display proteins that would be difficult for M13. Its particles also tolerate extreme conditions like high temperature and low pH, making it easier to use harsh elution steps during biopanning without destroying the phage. Other platforms include T4 and f1, though they see less widespread use.
What Phage Display Is Used For
The technique’s biggest impact has been in drug discovery, particularly for therapeutic antibodies. By displaying antibody fragments on phage surfaces, researchers can screen enormous libraries against disease-related targets to find antibodies that bind with high specificity. As of 2025, 21 FDA-approved therapeutic antibodies trace their origins to phage display, up from just 5 in 2015. That accounts for about 13% of all approved therapeutic antibodies, behind hybridoma technology (56%) and transgenic mice (20%), but growing steadily.
Beyond antibodies, phage display is used to find peptides that activate or block membrane receptors, identify enzyme inhibitors, map the specific regions of proteins that antibodies recognize (epitope mapping), study protein-protein interactions, and develop targeted drug delivery systems. Peptides selected from phage libraries tend to bind biologically relevant sites on target proteins, such as enzyme active sites or allosteric pockets, which means they often interfere with the target’s function rather than just sticking to an inert surface.
How It Compares to Other Display Systems
Phage display isn’t the only way to link proteins to their encoding genes. Yeast display uses the surface of yeast cells instead of viruses, which brings an important advantage: yeast is a eukaryotic organism with the cellular machinery to add sugar chains and form complex disulfide bonds that mammalian proteins need to fold correctly. For antibodies or other complex proteins that require these modifications, yeast display can produce more functionally accurate results.
Phage display’s primary edge is library size. Because phage libraries can contain billions of unique variants, they cover far more sequence diversity than yeast or bacterial display systems typically achieve. This makes phage display especially powerful for initial discovery, when you’re casting the widest possible net. Many research groups use phage display for the first screen, then switch to yeast display for fine-tuning the best candidates.
How Sequencing Has Changed the Game
Traditional phage display relied on Sanger sequencing, which reads one clone at a time. Researchers would pick a few dozen or a few hundred clones after biopanning and sequence them individually. This approach misses the vast majority of what’s happening in the library, making it easy to overlook genuine binders that simply weren’t abundant enough to be sampled.
Next-generation sequencing has transformed this bottleneck. By reading millions of sequences from every fraction of every biopanning round, researchers can now track how individual peptide motifs rise or fall across rounds, distinguish true target binders from sequences that are simply good at replicating inside bacteria, and flag known “parasitic” sequences that dominate libraries for reasons unrelated to binding. This deeper view reveals position-specific amino acid patterns and helps separate real hits from artifacts, such as the library-wide loss of certain amino acid types that happens during amplification. The combination of phage display with deep sequencing has made the technique substantially more reliable and information-rich than it was even a decade ago.