As part of our mission to make binder discovery faster, more accurate, and more accessible than ever before, Diffuse has invented a new assay to accelerate experimental validation of binders at scale.
Today we're launching RamaX, our platform for ultra-fast, high-sensitivity screening of thousands to billions of minibinders, VHHs, and scFvs. Where traditional workflows take months, RamaX enables us to deliver validated binders from naive or AI-designed libraries in just 1-2 weeks.
As protein generative methods improve, experimental screening becomes the rate-limiting step. We needed a faster way to accurately detecting functional protein designs—including in regimes where expected success rates are <1%.
Similar to our rationale for training better protein foundation models, which enable a multitude of computational protein design tasks, we developed RamaX as a new detection methodology which accelerates many experimental workflows. This advance in speed and throughput comes from a new assay that intrinsically supports large-scale screening, rather than lab automation of inefficient existing protocols.
High-capacity, ultra-rapid screening protocols raise the ceiling for protein discovery. RamaX expands our ability to identify binders against increasingly challenging targets, while generating the massive biological datasets that will power our next generation of de novo protein design models.
We discuss RamaX's capabilities, performance benchmarks, and technical specifications in our full technical report. Here’s a quick summary of the key results:
Switching from yeast surface display (YSD) to RamaX for our internal campaigns has completely transformed our ability to screen and learn from AI-designed libraries. RamaX can also enable rapid, large-scale screening for many other binder discovery workflows. We’re excited to put RamaX into your hands as a screening service to accelerate fast binder discovery at scale.
View our full list of RamaX offerings and pricing on our website.
Improvements are underway to increase parallel antigen screening capacity and expand to applications beyond binder identification. For example, prediction and control over multimodal developability properties remains challenging. We envision using future extensions of RamaX to collect large-scale data to solve downstream challenges in drug candidate development.
Developing new infrastructure like RamaX represents a key milestone in our vision: seamlessly integrating AI protein design with experimental validation to accelerate therapeutic discovery.
If this excites you, we encourage you to get in touch to explore partnership opportunities, learn more about RamaX, or work with us by joining our team.
— The Diffuse team
We are eager to connect with you.
