New computational method improves ability to detect precancerous tissues
MD Anderson Research News October 13, 2025
- CoCo-ST is a novel method of mapping tissues that is superior to current state-of-the-art benchmarks in preclinical models
- Technique allows clinicians to see parts of the tissue that are overlooked by current methods
- New approach has the potential to vastly improve understanding of earliest stages of cancer development
A new computational method known as Comparing and Contrasting Spatial Transcriptomics (CoCo-ST), developed by researchers at Â鶹ӳ» MD Anderson Cancer Center, could increase understanding of the earliest stages of cancer development. A report of the method was published today in
¡°This is a really exciting method that is helping us discover details that have previously been hidden,¡± said corresponding author , associate professor of Imaging Physics. ¡°These details are crucial to unlocking the earliest stages of cancer development.¡±
What is spatial transcriptomics and how does this method improve it?
Spatial transcriptomics has emerged as an important field in cancer research over the last decade. This field uses advanced computational tools to analyze cells while also preserving information about the microenvironment surrounding those cells at the same time.
Previously, transcriptomic data could provide details about which cells were present, but any information about where they were present or how various cells interacted with each other was lost. The field of spatial transcriptomics has led to advances in the understanding of how cancers develop, how immune cells interact within tumors, and how tumor cells interact with each other.
However, current methods have been limited to what are known as high variance structures ¨C those that are easy to distinguish from one another. CoCo-ST, instead, compares a target sample to a background sample, detecting both high and low variance structures. Essentially, this new method sharpens the vision of computational tools and gives scientists the ability to see developments that previously would have been overlooked, such as cell types that are very closely related.
¡°CoCo-ST lets us see further into the earliest developmental stages of tumors,¡± said Wu, who is also an affiliate member of the Institute for Data Science in Oncology at MD Anderson. ¡°The more we learn about how tumors develop, the better we can devise strategies to fight them.¡±
***
This study was funded by the National Institutes of Health, the Cancer Prevention and Research Institute of Texas, philanthropic support from Andrea Mugnaini and Dr. Edward Smith, Rexanna¡¯s Foundation for Fighting Lung Cancer, QIAC Partnership in Research (QPR) funding, and the Permanent Health Fund. For a full list of collaborating authors, disclosures and research funding support, read the full paper at