The Mission

We are developing a functional secondary diagnostic tool that serves to identify non-coding gene variants and their respective dynamic properties concerning disease development. Ultimately, we aim to validate library-based diagnostic technologies and facilitate comprehensive analysis of clinical cases, delivering more reliable diagnostic data for clinical reference.

The Challenge

Rare diseases affect 3-5% of the global populace, although a seemingly negligible statistic, this translates to circa 300 million people. While the majority of research for genetically linked diseases is conducted with a focus on coding regions, new evidence highlights the pivotal contribution of functional genetic variants in non-coding regions to disease pathology. Given the potential effects exerted on neighboring functional genes, non-coding genes may act as regulatory elements modulating gene expression and genome structure. To investigate these relationships, we are leveraging WGS (whole genome sequencing) technologies, considering their ability to characterize non-coding variants. Despite these aims, intron-contingent disease pathology remains under-researched and underrepresented in current models.

The solution

By developing and training a lightweight ML model capable of screening, identifying, and scoring intron variants based on potential risk, we tackle the limitations of current exome-based models to maximize diagnostic coverage as well as contribute to the relatively unexplored field of non-coding genomics.