smDeepFLUOR: single-molecule deep learning fluorescence classification
Jinseob Lee, Byungju Kim, Gayun Bu, Muhammad Tehseen, Vlad-Stefan Raducanu, Samir M. Hamdan, Jong-Bong Lee · 2026
Read smDeepFLUOR: single-molecule deep learning fluorescence classification by Jinseob Lee, Byungju Kim, Gayun Bu, Muhammad Tehseen, Vlad-Stefan Raducanu, Samir M. Hamdan, Jong-Bong Lee (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13438114.txt.
Description
Jinseob Lee and colleagues taught a deep learning system to read faint single-molecule glow signals human-designed rules cannot sort. Fluorescence imaging tracks single proteins by light, yet distinct events often produce near-identical brightness traces. Their 2026 study fed the network short three-dimensional image sequences, tiny blocks of space-time, and let it hunt for hidden patterns without built-in physics rules or hand-made features.
The system learned to tell true binding from stray, nonspecific sticking with accuracy up to ninety-seven percent, and kept that skill across data from different days. It also followed DNA being built in real time, sensing tiny changes in distance between a glowing tag and the growing strand's end, a feat beyond standard analysis. They pitch the approach as a way to widen single-molecule imaging, including minimal labels or unlabeled targets.
Catalog record
| CREATOR | Jinseob Lee, Byungju Kim, Gayun Bu, Muhammad Tehseen, Vlad-Stefan Raducanu, Samir M. Hamdan, Jong-Bong Lee |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65345 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |