Multimodal Quantum Vision Transformer for Enzyme Commission Classification from Biochemical Representations
Murat Isik, Mandeep Kaur Saggi, Humaira Gowher, Sabre Kais · 2025
Read Multimodal Quantum Vision Transformer for Enzyme Commission Classification from Biochemical Representations by Murat Isik, Mandeep Kaur Saggi, Humaira Gowher, Sabre Kais (2025) 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/arxiv-2508-14844.txt.
Description
Murat Isik and co-authors (2025) built a quantum machine learning model that sorts enzymes into their official function classes. Knowing an enzyme's class matters for drug design and industry, yet many enzymes lack the structure data that older methods rely on. The model reads four kinds of input at once: the protein's letter sequence, electronic features computed with quantum methods, a graph of the molecule's bonds, and a two-dimensional picture of the molecule. A quantum vision transformer joins the four streams through an attention step.
The combined model named the right class for 85.1 percent of test enzymes. That score beat models using the sequence alone by a wide margin and also beat other quantum learning models. The authors argue that mixing several views of an enzyme, including the spatial patterns that shape its chemistry, gives a stronger basis for guessing function than any single view.
Catalog record
| CREATOR | Murat Isik, Mandeep Kaur Saggi, Humaira Gowher, Sabre Kais |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-14844