What Makes NucleoPools Different

Traditional platforms limit exploration to the sequences you design in advance. NucleoPools deliver two dimensions of information:

• Designed diversity: Thousands of homologs or synthetic variants that intentionally span evolutionary or design space.

• Incidental diversity: Full-length mutant variants (≤5 amino acid changes) that arise naturally during synthesis and are captured through our barcoding and sequencing framework.

The result is a 30× expansion in functional data from a single assay. Importantly, those “extra” mutants are not noise, they are in fact signal. They define the edges of sequence space, highlight where designs fail, and illuminate rescue mutations that reveal new mechanisms of function.

Proof in the Data

This principle was recently demonstrated in Science Advances by Karl Romanowicz and colleagues. Using SynPlexity’s DropSynth-powered approach, the team built a library of 1,536 dihydrofolate reductase (DHFR) homologs, spanning 759 bacterial species, including clinically relevant pathogens.

By incorporating not just the designed homologs but also the thousands of incidental mutants, they revealed that 90% of variants could rescue function, uncovering resistance pathways invisible to conventional methods. Mutations once seen as byproducts became the key to understanding how antibiotic resistance evolves across a diverse protein family.

This work sets a new standard for how comprehensive, multiplexed data can unlock evolutionary insights—and it’s a direct extension of what’s possible for any lab using NucleoPools.

From Data to Decisions

For our customers, the takeaway is simple: you can move faster, with more confidence, when you have the whole picture.

• Fewer blind spots: Negative results and dropout variants are just as informative as successes.

• Better models: Rich, balanced “yes/no” outcomes fuel machine learning for predictive design.

• More paths forward: Incidental variants often reveal functional rescue or hidden tolerances you didn’t know were there.

And because assay-side cost is fixed, scaling from 1,000 to 30,000 variants costs essentially the same. That’s leverage.

A Collaborative Approach

We know that every project is unique. That’s why we don’t just hand over a library, we work with you on bioinformatics, design, and interpretation. Whether you’re probing antibiotic resistance, mapping enzyme families, or building training datasets for AI, our team is ready to help you translate scale into insight.

As Karl’s paper shows, NucleoPools aren’t exclusively focused on more data (thought that is valuable), they enable the right data, delivered at the speed modern biology demands.

Ready to dig deeper into your project?

Let’s explore how NucleoPools can accelerate your discovery.