Redefining Large-Scale Genomics for the Future of Medicine

Ultima Genomics’ mission is to continuously drive down the cost of omics data to unlock new sequencing applications that accelerate life science research and improve human health. The UG 100 Sequencing Platform, with Solaris chemistry, delivers on this mission by providing ultra-high-throughput sequencing at a low cost per base.

The platform achieves this by leveraging an open flow cell architecture, fast and efficient sequencing chemistry, robotic automation and machine learning-powered analysis to provide production-scale sequencing. The platform was designed to enable data-hungry applications that demand high quality, low cost and scale, and it has already accelerated the reshaping of the genomics landscape.

For example, several industry-first, large-scale projects are being powered by our platform, showcasing the application of machine learning and AI techniques in biology. We are also seeing a growing wave of clinical applications coming that leverage lower-cost sequencing, including whole-genome sequencing and liquid biopsy. 

Single-cell, spatial and multiomic data provide a high-resolution view of biology as well as spatial context in biological systems. When you train on millions of cells across tissues and time points, with spatial coordinates and matched transcriptomic, epigenomic and proteomic readouts, models learn more actionable biology versus training on any proxy. This enables a better picture of biology and can improve applications like target discovery, disease subtyping and response prediction.

Through collaborations with the Arc Institute, the Chan Zuckerberg Initiative (CZI) and the UK Biobank Pharma Proteomics Project, for example, we’re supporting the creation of some of the largest and most comprehensive multiomic datasets in the world.

Our role is to drive down the cost of generating omics data so that foundational datasets can be broad, diverse, generated economically and sustainably, so that AI models can truly be rooted in biology. 

Higher throughput and lower cost directly expand the scope of atlases and screens. For programs with the Arc Institute and CZI, lower costs on the UG 100 enable analysis of more samples, more cells and tissues, more cellular perturbations and more comprehensive reading of omics of the cells to capture rare states and cell–cell interactions.

For example, in studies like pooled perturbation and drug screening, a lower cost per read allows for larger perturbation libraries, more replicates and time-course designs that increase statistical power. This helps collaborators transition from pilot maps to production-grade, population-scale atlases more quickly.

Precision oncology is an excellent example of a field and clinical application that is incredibly sequencing data-hungry. Since the field emerged, there has been a constant migration to more comprehensive sequencing-intensive approaches and applications. Whether that be a more comprehensive analysis of tumor tissue for therapy selection, target selection and monitoring for minimal residual disease or deep sequencing for earlier detection.

The UG 100 platform is playing a role across many of these applications and has been adopted especially rapidly in the fast-growing field of liquid biopsy. A number of clinical testing companies disclosed that they are working with us in this area, and there are a host of others that are undisclosed.

Customers are interested in Ultima for applications like minimal residual disease and early detection because these applications require extreme performance characteristics of the sequencing technology, including identifying rare cancer signatures in blood. In these applications, the sequencer needs to be highly sensitive, be accurate down to a rare single read and be low-cost so that you can read a lot.

One good example of how our platform is differentiated in these applications is our technology called ppmSeq. ppmSeq is unique in that it enables a part-per-million level of detection of rare SNVs in blood through a whole-genome sequencing approach. 

For population-scale initiatives such as the UK Biobank proteomics program, our platform provides cost-effective omics data so that these initiatives can perform larger studies and generate richer databases with more comprehensive omics data that, in the end, generate more insights and impact.

The UK Biobank in particular will be one of the largest and first biobanks to pair large-scale sequencing data with proteomics and across longitudinal timepoints with clinical outcomes information.

In the case of Gene by Gene, our platform is enabling Gene by Gene’s customer, MyHeritage, to transition its direct-to-consumer franchise, which addresses more than 1,000,000 samples per year, from arrays to whole genomes. This enables a better product for MyHeritage customers, powered by richer datasets and insights and at a cost that does not impact the end price to the consumer. This was an important moment for the genomics industry because it marked a turning point when, for a large application, whole-genome information can be generated more cheaply than targeted panels and enable a better end product.

Sequencing scale and AI progress directly reinforce each other. Sequencing scale drives bigger and richer datasets that can train better models that accelerate AI’s impact on new clinical sequencing applications and clinical assay design. Take many of the leading clinical oncology companies leveraging next-generation sequencing (NGS), for example. Many of them have and are building large biobanks and databases built on NGS data. As these datasets grow, the application of AI becomes more powerful, for example, to inform the design of clinical assays that more directly predict treatment response or better detect cancer earlier in the blood. These large datasets are also impacting the bioinformatics layers of these sequencing applications. Larger datasets better inform AI models that can be used to build better analysis pipelines and clinical reporting techniques.