A new AI assisted approach aligns data standards and accelerates interoperability in biomedical research
Rodney Alan Long, Shannon Ballard, Syed Shah, Owen Bianchi, Lietsel Jones, Mathew J Koretsky, Nicole Kuznetsov, Elise Marsan, Bryant Jen, Philip Chiang, Abhradeep Mukherjee, Cornelis Blauwendraat, Hampton Leonard, Dan Vitale, Kristin Levine, Sara Bandres-Ciga, Paige Jarreau, Patrick Brannelly, Mukta Phatak, Caroline Pantazis, Laurel Screven, Kate Andersh, Alifiya Kapasi, John F Crary, David Gutman, Brittany N Dugger, Sarah Biber, Timothy Hohman, Faraz Faghri, Michael Griswold, Lana Sargent, Kendall van Keuren-Jensen, Andrew B Singleton, Yang Fann, Mike A Nalls, Hirotaka Iwaki · 2026
Read A new AI assisted approach aligns data standards and accelerates interoperability in biomedical research by Rodney Alan Long, Shannon Ballard, Syed Shah, Owen Bianchi, Lietsel Jones, Mathew J Koretsky, Nicole Kuznetsov, Elise Marsan, Bryant Jen, Philip Chiang, Abhradeep Mukherjee, Cornelis Blauwendraat, Hampton Leonard, Dan Vitale, Kristin Levine, Sara Bandres-Ciga, Paige Jarreau, Patrick Brannelly, Mukta Phatak, Caroline Pantazis, Laurel Screven, Kate Andersh, Alifiya Kapasi, John F Crary, David Gutman, Brittany N Dugger, Sarah Biber, Timothy Hohman, Faraz Faghri, Michael Griswold, Lana Sargent, Kendall van Keuren-Jensen, Andrew B Singleton, Yang Fann, Mike A Nalls, Hirotaka Iwaki (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/PMC13616909.txt.
Description
Rodney Alan Long and a large multi-institution team report a 2026 approach that uses large language models to speed up biomedical data standardization. Research groups describe their variables in incompatible ways, which blocks pooling of studies. Their system drafts common data elements, the shared definitions that let datasets line up, for clinical and research dictionaries.
The team processed thirty-one datasets through a GPT-4 based pipeline and had experts check the output. Most generated fields needed no revision, and accuracy was highest on well-structured sources. Tests on Alzheimer and Parkinson datasets mapped about a third of unseen variable names onto existing elements automatically. The authors argue that automating this tedious groundwork lowers barriers to collaboration across studies.
Catalog record
| CREATOR | Rodney Alan Long, Shannon Ballard, Syed Shah, Owen Bianchi, Lietsel Jones, Mathew J Koretsky, Nicole Kuznetsov, Elise Marsan, Bryant Jen, Philip Chiang, Abhradeep Mukherjee, Cornelis Blauwendraat, Hampton Leonard, Dan Vitale, Kristin Levine, Sara Bandres-Ciga, Paige Jarreau, Patrick Brannelly, Mukta Phatak, Caroline Pantazis, Laurel Screven, Kate Andersh, Alifiya Kapasi, John F Crary, David Gutman, Brittany N Dugger, Sarah Biber, Timothy Hohman, Faraz Faghri, Michael Griswold, Lana Sargent, Kendall van Keuren-Jensen, Andrew B Singleton, Yang Fann, Mike A Nalls, Hirotaka Iwaki |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 80766 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |