MDCDR: predicting cancer drug response via multimodal feature fusion and feature disentanglement
Jinxiong Zhang, Hedong Xu, Chunyan Tang, Dandan Ni, Xuyang Shen · 2026
Read MDCDR: predicting cancer drug response via multimodal feature fusion and feature disentanglement by Jinxiong Zhang, Hedong Xu, Chunyan Tang, Dandan Ni, Xuyang Shen (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/pmc-13262659.txt.
Description
Jinxiong Zhang and co-authors present a study the catalog lists under 2026. Cancer drug response varies widely across patients. A model that predicts response could guide personal care. Cell line data and tumor tissue differ, and drugs act in varied ways, which shifts the data patterns. The team builds a model called MDCDR to join many data types. Drug traits and cell line traits are fused in one frame.
Their design splits features into shared patterns and specific ones. Contrast learning at several grain sizes sharpens that split. Masked feature guessing adds a second training signal. Tests show the model beats past methods by a clear margin, as the authors report. Drug teams get a predictor built for messy, mixed data. The split design is the key idea: shared biology and drug-specific quirks each get their own space instead of being blurred together.
Catalog record
| CREATOR | Jinxiong Zhang, Hedong Xu, Chunyan Tang, Dandan Ni, Xuyang Shen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 40 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |