Back

EZ-DMS - A Simple and Accessible Protocol and Software Package for Deep Mutational Scanning of Virus Proteins

Gillman, A. N.; Broghammer, M. M.; McCarthy-Potter, S. A.; Vujjini, R. R.; Birler, C. M.; Kleinpeter, A. B.; Haim, H.

2026-01-02 microbiology
10.64898/2026.01.02.697422 bioRxiv
Show abstract

Mutations that occur during viral genome replication can result in new phenotypes, such as infection of new cell types or resistance to therapeutics. To determine the phenotypic potential of viruses, deep mutational scanning (DMS) can be used. This approach measures the effects of all amino acid changes at sites of interest on protein function, virus infectivity, or resistance to therapeutics. A virus library that contains all possible variants is produced and used to infect a culture of cells. The frequency of each amino acid at the DMS site in the infected culture relative to the input stock captures the relative fitness for each form under the selective pressure applied. Current DMS protocols are complex and costly due to reliance on deep sequencing for genomic characterization. They also require expertise in bioinformatics to process and analyze the data, thus limiting the number of labs that can apply the technology. Here, we describe a simple and efficient protocol to perform DMS on single sites and on two-site combinations in the Env protein of HIV-1. Dual-site DMS allows probing for epistatic effects. The protocol greatly reduces recombination of the proviral genome in transformed bacteria, thus efficiency of the method. Sequencing is performed using Oxford Nanopore technology, and the data are uploaded to a graphical user interface-based platform that calculates the amino acid preferences, which describe the relative fitness of all amino acids under the experimental conditions tested (e.g., under selection pressure of a drug). The protocol does not require expertise in bioinformatics and can be completed in approximately 16 days, with 3-4 days of laboratory activity and 12-13 days of incubation, with considerably lower costs than existing systems.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.