Chromly is a one-person product. It was built by Josh, with heavy help from modern AI coding tools. The privacy story on the homepage is the truth, and the reason it's the truth is that there's only one person who would have to break it — and that person designed the architecture so he literally can't.
Josh McPhail is the founder and sole developer of Chromly. He's worked a wide range of jobs — from farms to grocery stores to life coaching — and earned a degree in economics along the way. The through-line has always been a love of learning about the human condition and just about anything in the sciences that tries to explain it. Chromly started as an attempt to make something complex, and turned into a more interesting and engaging way to learn something about ourselves.
What that means in practice: when you email support, it's me. When the code ships a bug, it's my bug. When the privacy claim says "your DNA never leaves your machine," there's no marketing committee between you and the person who wrote the network code.
Modern code-assistant tools wrote a large fraction of Chromly's source code under my direction and review. I think that should be on the table for anyone evaluating a privacy product, so I'm saying it on the front of the about page instead of hoping nobody asks.
Two things follow from that:
Consumer DNA tests in the late 2010s and 2020s produced a strange situation. People paid for sequencing, got back a wall of trait reports, and walked away with no actual feel for what they had bought. The raw file sits in a download folder. Three billion base pairs, packaged as a tab-separated text file, opened once and never again.
The information is interesting. The way it's presented is not. Pie charts and percentages don't have shape. Chromosomes do.
So Chromly takes the raw file you already paid for and renders it as the thing it actually is: 24 three-dimensional structures, folded the way the published Hi-C literature says chromosomes fold, with your variants pinned to their real coordinates. You can fly through it, drill into specific genes, read curated descriptions of what each one does. The science page has the details on what's grounded in published data and what is procedurally rendered.
If a product can technically see your data, it eventually will. The only honest privacy guarantee is the kind enforced by where the code runs, not by what the privacy policy says. Chromly's parser runs in your browser because that's the only place it can run without me being a single point of failure.
Support and direct contact: [email protected]. It comes to me. There is no support ticket queue, no chatbot, no offshore team. I read every message.
If you find a bug, a wrong gene annotation, a privacy concern, or a part of the UI that just confuses you, please tell me. The product gets better that way and it's been the main feedback loop since launch.
How the 3D rendering actually works (TADs, GRCh38, where the data comes from) →
← Back to chromly.app