Monash AI-driven biomedical model could make advanced disease research faster, cheaper and more accessible

Monash BDI’s Prof Jiangning Song and Dr Zhikang Wang.

Researchers at Monash University’s Biomedicine Discovery Institute (BDI) have developed an artificial intelligence (AI) model that could dramatically expand access to advanced spatial multi-omics research, enabling scientists to generate rich biological insights from data that would otherwise be expensive or difficult to obtain.

It stands as the first computational approach specifically designed for spatially-aware cross-omics translation.

The novel AI model, called NicheTrans, is designed to bridge a longstanding gap in spatial biology by translating one type of molecular data into another while preserving critical information about how cells are organised within tissues. It uses highly-flexible transformer-based deep learning technology to infer biological information - such as protein expression - from more readily available data, including gene expression measurements.

The latest advances in NicheTrans have been published in Nature Methods, in research led by co-senior authors Professor Jiangning Song, head of the AI-driven Bioinformatics and Biomedicine Lab at Monash BDI, and Professor Zhiyuan Yuan of Fudan University, with first author Dr Zhikang Wang, a recent PhD graduate from Professor Song’s Lab.

Unlocking a fuller picture of disease

Scientists increasingly rely on multi-omics approaches to understand how diseases develop and progress. By combining information from different biological layers - including genes, proteins and other molecular signals - researchers can build a more complete picture of what is happening inside cells and tissues.

However, acquiring multiple types of high-resolution biological data from the same sample is often expensive, technically challenging and beyond the reach of many laboratories.

NicheTrans addresses this challenge by learning the relationships between different molecular data types and using those relationships to accurately generate missing information. Crucially, the model also incorporates spatial context, enabling researchers to study not only which cells are present, but also how they are organised and interact within their surrounding tissue environment.

During validation across multiple biological systems, the researchers found that NicheTrans was able to identify spatial patterns and biological relationships that could not be detected using a single data type alone.

A powerful tool for studying cancer and neurological disease

Professor Song said the ability to understand cellular neighbourhoods and tissue architecture is becoming increasingly important in research into complex diseases such as solid tumour cancers, Alzheimer's disease and Parkinson’s disease.

“NicheTrans can reveal important architectural relationships between different cells, or the association between different biomarkers,” Professor Song said.

The technology could help researchers analyse the cellular microenvironments that influence disease progression, treatment response and recurrence. For example, in brain tissue affected by Alzheimer’s disease, NicheTrans can help characterise how different cell types are organised and interact across spatial regions.

In cancer research, the tool may allow scientists to identify specific cell populations or cellular neighbourhoods linked to treatment resistance or disease relapse.

“Because many important omics measurements are often expensive and technically difficult, NicheTrans provides a game-changing solution,” Professor Song said.

“It accurately fills in missing data, making multi-omics analysis highly accessible and affordable.”

Lowering the barriers to advanced biomedical research

A key advantage of the technology is its potential to enable access to sophisticated spatial biology techniques.

Professor Song said the team’s goal was to use AI to translate information from expensive experimental platforms into insights that could be generated from more affordable technologies.

“If we can do that, these tools can be used worldwide in labs that may lack access to high-end, expensive equipment,” he said.

“Instead, they could use this cheaper way to generate high-quality data.”

The researchers believe the approach could also support more precise patient stratification by linking spatial biological data with clinical outcomes. This could help scientists better understand why some patients respond to treatment while others do not, potentially informing future precision medicine approaches.

“The tool may identify specific types of cells or cell clusters within a spatial neighbourhood or spatial niche that can better inform a patient’s treatment,” Professor Song said.

Next steps

Professor Song is keen to apply the NicheTrans technique to analyse solid tumour samples in cancer patients who have a relapse following treatment.

Researchers hope the model will help identify biomarkers and cellular signatures associated with tumour relapse, improving understanding of why some cancers return while others remain in remission.

Importantly, the NicheTrans software has been released as a free, open-source resource for academic and non-commercial research, allowing scientists worldwide to apply and build upon the technology. The platform also presents opportunities for collaboration with pharmaceutical and biotechnology companies, helping to establish public-private partnerships that accelerate its translation into clinically relevant applications. In the longer term, these partnerships may also support the development of a spin-off company to further advance and commercialise the technology.

“We are developing AI tools for a future where we can reconstruct the full picture of our cellular environment within diseased tissue,” Professor Song said.

“That will allow us to gain deeper insights into more effective treatments. That is the road we are on.”

DOI: 10.1038/s41592-026-03153-3

MEDIA ENQUIRIES

For more Monash media stories, visit our news and events site
Monash Media
E: media@monash.edu
T: +61 (0) 3 9903 4840

About the Monash Biomedicine Discovery Institute at Monash University

Committed to making discoveries that will relieve the future burden of disease, Monash Biomedicine Discovery Institute at Monash University brings together more than 120 internationally renowned research teams. Spanning seven discovery programs across Cancer, Cardiovascular Disease, Development and Stem Cells, Infection, Immunity, Metabolism, Diabetes and Obesity, and Neuroscience, Monash BDI is one of the largest biomedical research institutes in Australia. Our researchers are supported by world-class technology and infrastructure, and partner with industry, clinicians and researchers internationally to enhance lives through discovery.