AI model maps tumor tissue to improve cancer care
A tumor is more than its cancer cells. Immune cells, blood vessels and other surrounding tissue influence how a cancer grows and responds to treatment. Two tumors containing the same cell types can respond very differently to the same therapy depending on how those cells are arranged. Spatial proteomics captures this organization in molecular detail, but the resulting data, which contain many…
Tumors are not just a collection of cancer cells. Their growth and response to treatment can be influenced by surrounding immune cells, blood vessels and other tissue. Spatial proteomics provides a detailed molecular map of this tissue organization, but the complex data sets are challenging to interpret and compare across studies.
To tackle this issue, researchers at EPFL developed Virtual Tissues (VirTues), an AI model that learns from spatial proteomics data across different cancers and studies. VirTues can analyze individual cells to whole tissue sections and patient outcomes, addressing both the biological and computational challenges of interpreting tissue data.
By incorporating data from numerous studies, VirTues creates a shared representation of proteins, cells, and their spatial context, applicable to various analyses. With VirTues, researchers can compare patient groups, determine distinguishing spatial patterns, and discover spatial biomarkers linked to disease progression and treatment response.
The model's flexibility allows researchers to rapidly integrate new tissue samples, building on the model's learnings instead of creating new models from scratch. This advancement in foundation modeling for spatial proteomics holds great potential for precision oncology, from understanding cell types and states within tumors to predicting treatment outcomes and tailoring therapies to individual patients.
Written by urgent.news from Medical Xpress's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
