NEW YORK – Researchers affiliated with the Singapore Agency for Science, Technology and Research (A*STAR) are leading a collaborative effort in Singapore to develop a spatial omics platform for biomarker analysis, drug target discovery, and personalized medicine applications.
The project has been awarded $18 million in grant funding and will run for four years. In addition to researchers from A*STAR's Genome Institute of Singapore, the alliance includes experts from A*STAR's Bioinformatics Institute, SingHealth's specialty center, National Cancer Center Singapore, and Singapore General Hospital.
The Target Inference from Spatialomics and Histology Using Multimodal AI and Phenotypes (TISHUMAP) platform will use artificial intelligence to infer cell-cell interactions based on microscopic tissue images from 1 million tissue images in the SingHealth Digital Pathology Archive and 20,000 samples from the SingHealth Tissue Repository. The platform will rank the interactions based on correlations with de-identified clinical data and tissue image features.
The research team will use the results to identify promising drug targets and define subtypes of diseases for customized therapy. A*STAR said TISHUMAP will focus on a variety of cancers, including lung, breast, colon, liver and gastric cancers, as well as chronic kidney disease and fatty liver disease.
A*STAR said it also expects TISHUMAP to yield insights into drug mechanisms and potential additional disease indications.
The researchers previously generated spatial omics and single-cell datasets from clinical samples and identified two subtypes of colorectal cancer. They are currently exploring potential partnerships for multimodal tissue analysis and scale-up of the technology.
"The images and samples from SDPA and STR will provide high-resolution data and clinical-grade quality that will allow the program to develop insights and predictions," Tony Lim, chairman of the division of pathology at Singapore General Hospital, said in a statement. "This data has been stored and collected over many years and can be annotated by pathologists to allow an array of different analyses."