The problem Evo-Gene solves
Problem We're Solving
The current process for genetic variant interpretation is a critical
bottleneck in personalized medicine.
● Challenges: Current manual and data intensive analysis is plagued by:
long turnaround times (2 4 weeks), high operational costs ($200 --$500
per variant), and severely limited access to expert geneticists.
● Uncertainty Crisis: The technology has outpaced our ability to interpret
it; consequently , 40 50% of patient genetic variants are classified as
"Uncertain Significance ( hindering diagnosis and treatment
planning.
● High Stakes: This affects millions undergoing genetic testing for
diseases like cancer (e.g., a single BRCA1 mutation can increase breast
cancer risk by up to 85%), heart disorders, and rare genetic conditions,
demanding a rapid, scalable, and highly accurate AI solution.
Target Users
● Professionals aged 25- 55 in the healthcare, biotechnology, and research sectors.
● B2B, offering tools and APIs for labs, hospitals, and research institutions
Challenges we ran into
- I faced a difficult bug with localStorage-based authentication where the app kept redirecting to the login page even after logging in.
- The issue was caused by inconsistent client-side checks and timing problems due to Next.js rendering. I resolved it by moving all token checks into useEffect(), unifying the login validation logic, and adding redirect handling.
- This made the authentication flow stable and seamless across the entire project.
- We also faced problems while inferencing our model- Evo2 on MODAL.